Shortcuts

챗봇 튜토리얼

Author: Matthew Inkawhich

번역: 김진현

이 튜토리얼에서는 순환(recurrent) 시퀀스 투 시퀀스(sequence-to-sequence) 모델의 재미있고 흥미로운 사용 예를 살펴보려 합니다. 간단한 챗봇을 학습해 볼 텐데, 사용할 데이터는 영화 대본으로 구성된 Cornell Movie-Dialogs(코넬 대학교의 영화 속 대화 말뭉치 데이터 입니다.

대화형 모델은 많은 사람들이 관심을 갖는 인공지능 분야의 연구 주제입니다. 고객 서비스와 관련된 활용, 온라인 헬프데스크 등 여러 상황에서 챗봇을 활용할 수 있습니다. 많은 챗봇이 검색 기반(retrieval-based) 모델을 사용하는데, 이는 특정한 형식을 갖춘 질문에 대해 미리 정해진 반응을 출력하는 방식입니다. 분야를 특정 회사의 IT 헬프데스크처럼 한정짓는다면 이러한 모델을 사용해도 충분합니다. 그러나 이런 모델은 좀 더 일반적인 상황에 활용할 수 있을만큼 강력하진 않습니다. 기계를 학습시켜서 사람과 여러 주제에 대해 의미 있는 대화를 하게끔 하는 것은 아직 해결되지 않은 연구 주제입니다. 그러나 최근에 딥러닝이 유행하면서 여러 가지의 강력한 생성 모델이 등장했습니다. 그러한 모델의 한 예인 구글의 신경 대화 모델(Neural Conversational Model) 은 다중 도메인 대화 생성 모델(multi-domain generative conversational models) 분야에 있어서 큰 진전을 이루었습니다. 우리는 이 튜토리얼을 통해 이러한 모델을 PyTorch로 구현해보려 합니다.

bot
> hello? (안녕하세요?)
Bot: hello . (안녕하세요.)
> where am I? (여긴 어디죠?)
Bot: you re in a hospital . (병원입니다.)
> who are you? (당신은 누구시죠?)
Bot: i m a lawyer . (변호사입니다.)
> how are you doing? (어떻게 지내세요?)
Bot: i m fine . (잘 지냅니다.)
> are you my friend? (당신은 제 친구인가요?)
Bot: no . (아뇨.)
> you're under arrest (당신을 체포하겠습니다)
Bot: i m trying to help you ! (난 당신을 도우려 하는 겁니다!)
> i'm just kidding (농담이었어요)
Bot: i m sorry . (미안하네요.)
> where are you from? (어디서 오셨어요?)
Bot: san francisco . (샌프란시스코요.)
> it's time for me to leave (전 이제 가봐야겠네요)
Bot: i know . (알겠습니다.)
> goodbye (안녕히 계세요)
Bot: goodbye . (안녕히 가세요.)

이 튜토리얼의 핵심 내용

감사의 글

이 튜토리얼은 다음 자료의 도움을 받아 작성하였습니다.

  1. Yuan-Kuei Wu의 pytorch-chatbot 구현체: https://github.com/ywk991112/pytorch-chatbot

  2. Sean Robertson의 practical-pytorch seq2seq-translation 예제: https://github.com/spro/practical-pytorch/tree/master/seq2seq-translation

  3. FloydHub의 코넬 대학교의 영화 말뭉치 데이터 전처리 코드: https://github.com/floydhub/textutil-preprocess-cornell-movie-corpus

준비 단계

시작에 앞서, 여기 에서 ZIP 파일 형태의 데이터를 내려받고, 현재 디렉토리 아래에 data/ 라는 디렉토리를 만들어서 내려받은 데이터를 옮겨두시기 바랍니다.

그 다음에는, 몇 가지 필요한 도구들을 import 하겠습니다.

from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
from __future__ import unicode_literals

import torch
from torch.jit import script, trace
import torch.nn as nn
from torch import optim
import torch.nn.functional as F
import csv
import random
import re
import os
import unicodedata
import codecs
from io import open
import itertools
import math


USE_CUDA = torch.cuda.is_available()
device = torch.device("cuda" if USE_CUDA else "cpu")

데이터 읽기 & 전처리하기

다음 단계는 데이터 파일의 형식을 재조정한 후, 우리가 작업하기 편한 구조로 읽어들이는 것입니다.

코넬 대학교의 영화 속 대화 말뭉치 데이터셋 은 영화 속 등장 인물의 대화가 풍부하게 포함된 데이터셋입니다.

  • 영화 속 등장 인물 10,292 쌍이 대화를 220,579번 주고받습니다

  • 영화 617개의 등장 인물 9,035명이 나옵니다

  • 총 발화(utterance) 수는 304,713번입니다

이 데이터셋은 규모도 크고 내용도 다양하며, 격식체와 비격식체, 여러 시간대, 여러 감정 상태 등이 두루 포함되어 있습니다. 우리의 바람은 이러한 다양성으로 인해 모델이 견고해지는, 즉 모델이 여러 종류의 입력 및 질의에 잘 대응할 수 있게 되는 것입니다.

우선은 원본 데이터 파일을 몇 줄 살펴보면서 형식이 어떻게 되어있는지 살펴 보겠습니다.

corpus_name = "cornell movie-dialogs corpus"
corpus = os.path.join("data", corpus_name)

def printLines(file, n=10):
    with open(file, 'rb') as datafile:
        lines = datafile.readlines()
    for line in lines[:n]:
        print(line)

printLines(os.path.join(corpus, "movie_lines.txt"))

Out:

b'L1045 +++$+++ u0 +++$+++ m0 +++$+++ BIANCA +++$+++ They do not!\n'
b'L1044 +++$+++ u2 +++$+++ m0 +++$+++ CAMERON +++$+++ They do to!\n'
b'L985 +++$+++ u0 +++$+++ m0 +++$+++ BIANCA +++$+++ I hope so.\n'
b'L984 +++$+++ u2 +++$+++ m0 +++$+++ CAMERON +++$+++ She okay?\n'
b"L925 +++$+++ u0 +++$+++ m0 +++$+++ BIANCA +++$+++ Let's go.\n"
b'L924 +++$+++ u2 +++$+++ m0 +++$+++ CAMERON +++$+++ Wow\n'
b"L872 +++$+++ u0 +++$+++ m0 +++$+++ BIANCA +++$+++ Okay -- you're gonna need to learn how to lie.\n"
b'L871 +++$+++ u2 +++$+++ m0 +++$+++ CAMERON +++$+++ No\n'
b'L870 +++$+++ u0 +++$+++ m0 +++$+++ BIANCA +++$+++ I\'m kidding.  You know how sometimes you just become this "persona"?  And you don\'t know how to quit?\n'
b'L869 +++$+++ u0 +++$+++ m0 +++$+++ BIANCA +++$+++ Like my fear of wearing pastels?\n'

원하는 형식의 데이터 파일로 만들기

편의를 위해 데이터의 형식을 원하는 형태로 만들려고 합니다. 각 줄에 질의 문장응답 문장 의 쌍이 탭으로 구분되어 있게끔 하는 것입니다.

다음의 함수를 통해 movie_lines.txt 원본 데이터 파일을 파싱하려 합니다.

  • loadLines 는 파일에 포함된 대사를 변환하여 항목(대사 ID lineID, 인물 ID characterID, 영화 ID movieID, 인물 character, 대사 내용 text)에 대한 사전 형태로 변환합니다

  • loadConversationsloadLines 를 통해 읽어들인 대사(lines)의 항목(fields)를 movie_conversations.txt 에 나와 있는 내용에 맞춰 대화 형태로 묶습니다

  • extractSentencePairs 는 대화(conversations)에서 문장 쌍을 추출합니다

# 파일에 포함된 대사를 쪼개서 항목에 대한 사전(``dict``) 형태로 변환합니다
def loadLines(fileName, fields):
    lines = {}
    with open(fileName, 'r', encoding='iso-8859-1') as f:
        for line in f:
            values = line.split(" +++$+++ ")
            # 항목을 추출합니다
            lineObj = {}
            for i, field in enumerate(fields):
                lineObj[field] = values[i]
            lines[lineObj['lineID']] = lineObj
    return lines


# 대사의 항목을 *movie_conversations.txt* 를 참고하여 대화 형태로 묶습니다
def loadConversations(fileName, lines, fields):
    conversations = []
    with open(fileName, 'r', encoding='iso-8859-1') as f:
        for line in f:
            values = line.split(" +++$+++ ")
            # 항목을 추출합니다
            convObj = {}
            for i, field in enumerate(fields):
                convObj[field] = values[i]
            # 문자열을 리스트로 변환합니다(convObj["utteranceIDs"] == "['L598485', 'L598486', ...]")
            utterance_id_pattern = re.compile('L[0-9]+')
            lineIds = utterance_id_pattern.findall(convObj["utteranceIDs"])
            # 대사를 재구성합니다
            convObj["lines"] = []
            for lineId in lineIds:
                convObj["lines"].append(lines[lineId])
            conversations.append(convObj)
    return conversations


# conversations에서 문장 쌍을 추출합니다
def extractSentencePairs(conversations):
    qa_pairs = []
    for conversation in conversations:
        # 대화를 이루는 각 대사에 대해 반복문을 수행합니다
        # 대화의 마지막 대사는 (그에 대한 응답이 없으므로) 무시합니다
        for i in range(len(conversation["lines"]) - 1):
            inputLine = conversation["lines"][i]["text"].strip()
            targetLine = conversation["lines"][i+1]["text"].strip()
            # 잘못된 샘플은 제거합니다(리스트가 하나라도 비어 있는 경우)
            if inputLine and targetLine:
                qa_pairs.append([inputLine, targetLine])
    return qa_pairs

이제 이 함수들을 호출하여 새로운 파일인 formatted_movie_lines.txt 를 만듭니다.

# 새 파일에 대한 경로를 정의합니다
datafile = os.path.join(corpus, "formatted_movie_lines.txt")

delimiter = '\t'
# 구분자에 대해 unescape 함수를 호출합니다
delimiter = str(codecs.decode(delimiter, "unicode_escape"))

# 대사 사전(dict), 대화 리스트(list), 그리고 각 항목의 이름을 초기화합니다
lines = {}
conversations = []
MOVIE_LINES_FIELDS = ["lineID", "characterID", "movieID", "character", "text"]
MOVIE_CONVERSATIONS_FIELDS = ["character1ID", "character2ID", "movieID", "utteranceIDs"]

# 대사(lines)를 읽어들여 대화(conversations)로 재구성합니다
print("\nProcessing corpus...")
lines = loadLines(os.path.join(corpus, "movie_lines.txt"), MOVIE_LINES_FIELDS)
print("\nLoading conversations...")
conversations = loadConversations(os.path.join(corpus, "movie_conversations.txt"),
                                  lines, MOVIE_CONVERSATIONS_FIELDS)

# 결과를 새로운 csv 파일로 저장합니다
print("\nWriting newly formatted file...")
with open(datafile, 'w', encoding='utf-8') as outputfile:
    writer = csv.writer(outputfile, delimiter=delimiter, lineterminator='\n')
    for pair in extractSentencePairs(conversations):
        writer.writerow(pair)

# 몇 줄을 예제 삼아 출력해 봅니다
print("\nSample lines from file:")
printLines(datafile)

Out:

Processing corpus...

Loading conversations...

Writing newly formatted file...

Sample lines from file:
b"Can we make this quick?  Roxanne Korrine and Andrew Barrett are having an incredibly horrendous public break- up on the quad.  Again.\tWell, I thought we'd start with pronunciation, if that's okay with you.\n"
b"Well, I thought we'd start with pronunciation, if that's okay with you.\tNot the hacking and gagging and spitting part.  Please.\n"
b"Not the hacking and gagging and spitting part.  Please.\tOkay... then how 'bout we try out some French cuisine.  Saturday?  Night?\n"
b"You're asking me out.  That's so cute. What's your name again?\tForget it.\n"
b"No, no, it's my fault -- we didn't have a proper introduction ---\tCameron.\n"
b"Cameron.\tThe thing is, Cameron -- I'm at the mercy of a particularly hideous breed of loser.  My sister.  I can't date until she does.\n"
b"The thing is, Cameron -- I'm at the mercy of a particularly hideous breed of loser.  My sister.  I can't date until she does.\tSeems like she could get a date easy enough...\n"
b'Why?\tUnsolved mystery.  She used to be really popular when she started high school, then it was just like she got sick of it or something.\n'
b"Unsolved mystery.  She used to be really popular when she started high school, then it was just like she got sick of it or something.\tThat's a shame.\n"
b'Gosh, if only we could find Kat a boyfriend...\tLet me see what I can do.\n'

데이터 읽고 정리하기

다음에 해야 할 일은 어휘집을 만들고, 질의/응답 문장 쌍을 메모리로 읽어들이는 것입니다.

우리가 다루는 대상은 일련의 단어 들이며, 따라서 이들을 이산 공간 상의 수치(discrete numerical space)로 자연스럽게 대응시키기 어렵다는 점에 유의하시기 바랍니다. 따라서 우리는 데이터셋 안에 들어 있는 단어를 인덱스 값으로 변환하는 매핑을 따로 만들어야 합니다.

이를 위해 우리는 Voc 라는 클래스를 만들어 단어에서 인덱스로의 매핑, 인덱스에서 단어로의 역 매핑, 각 단어의 등장 횟수, 전체 단어 수 등을 관리하려 합니다. 이 클래스는 어휘집에 새로운 단어를 추가하는 메서드(addWord), 문장에 등장하는 모든 단어를 추가하는 메서드(addSentence), 그리고 자주 등장하지 않는 단어를 정리하는 메서드(trim)를 제공합니다. 단어를 정리하는 내용에 대해서는 뒤에서 좀 더 자세히 살펴보겠습니다.

# 기본 단어 토큰 값
PAD_token = 0  # 짧은 문장을 채울(패딩, PADding) 때 사용할 제로 토큰
SOS_token = 1  # 문장의 시작(SOS, Start Of Sentence)을 나타내는 토큰
EOS_token = 2  # 문장의 끝(EOS, End Of Sentence)을 나태는 토큰

class Voc:
    def __init__(self, name):
        self.name = name
        self.trimmed = False
        self.word2index = {}
        self.word2count = {}
        self.index2word = {PAD_token: "PAD", SOS_token: "SOS", EOS_token: "EOS"}
        self.num_words = 3  # SOS, EOS, PAD를 센 것

    def addSentence(self, sentence):
        for word in sentence.split(' '):
            self.addWord(word)

    def addWord(self, word):
        if word not in self.word2index:
            self.word2index[word] = self.num_words
            self.word2count[word] = 1
            self.index2word[self.num_words] = word
            self.num_words += 1
        else:
            self.word2count[word] += 1

    # 등장 횟수가 기준 이하인 단어를 정리합니다
    def trim(self, min_count):
        if self.trimmed:
            return
        self.trimmed = True

        keep_words = []

        for k, v in self.word2count.items():
            if v >= min_count:
                keep_words.append(k)

        print('keep_words {} / {} = {:.4f}'.format(
            len(keep_words), len(self.word2index), len(keep_words) / len(self.word2index)
        ))

        # 사전을 다시 초기화힙니다
        self.word2index = {}
        self.word2count = {}
        self.index2word = {PAD_token: "PAD", SOS_token: "SOS", EOS_token: "EOS"}
        self.num_words = 3 # 기본 토큰을 센 것

        for word in keep_words:
            self.addWord(word)

이제 어휘집과 질의/응답 문장 쌍을 재구성하려 합니다. 그러한 데이터를 사용하려면 그 전에 약간의 전처리 작업을 수행해야 합니다.

우선, unicodeToAscii 를 이용하여 유니코드 문자열을 아스키로 변환해야 합니다. 다음에는 모든 글자를 소문자로 변환하고, 알파벳도 아니고 기본적인 문장 부호도 아닌 글자는 제거합니다(정규화, normalizeString). 마지막으로는 학습할 때의 편의성을 위해서, 길이가 일정 기준을 초과하는, 즉 MAX_LENGTH 보다 긴 문장을 제거합니다(filterPairs).

MAX_LENGTH = 10  # 고려할 문장의 최대 길이

# 유니코드 문자열을 아스키로 변환합니다
# https://stackoverflow.com/a/518232/2809427 참고
def unicodeToAscii(s):
    return ''.join(
        c for c in unicodedata.normalize('NFD', s)
        if unicodedata.category(c) != 'Mn'
    )

# 소문자로 만들고, 공백을 넣고, 알파벳 외의 글자를 제거합니다
def normalizeString(s):
    s = unicodeToAscii(s.lower().strip())
    s = re.sub(r"([.!?])", r" \1", s)
    s = re.sub(r"[^a-zA-Z.!?]+", r" ", s)
    s = re.sub(r"\s+", r" ", s).strip()
    return s

# 질의/응답 쌍을 읽어서 voc 객체를 반환합니다
def readVocs(datafile, corpus_name):
    print("Reading lines...")
    # 파일을 읽고, 쪼개어 lines에 저장합니다
    lines = open(datafile, encoding='utf-8').\
        read().strip().split('\n')
    # 각 줄을 쪼개어 pairs에 저장하고 정규화합니다
    pairs = [[normalizeString(s) for s in l.split('\t')] for l in lines]
    voc = Voc(corpus_name)
    return voc, pairs

# 문장의 쌍 'p'에 포함된 두 문장이 모두 MAX_LENGTH라는 기준보다 짧은지를 반환합니다
def filterPair(p):
    # EOS 토큰을 위해 입력 시퀀스의 마지막 단어를 보존해야 합니다
    return len(p[0].split(' ')) < MAX_LENGTH and len(p[1].split(' ')) < MAX_LENGTH

# 조건식 filterPair에 따라 pairs를 필터링합니다
def filterPairs(pairs):
    return [pair for pair in pairs if filterPair(pair)]

# 앞에서 정의한 함수를 이용하여 만든 voc 객체와 리스트 pairs를 반환합니다
def loadPrepareData(corpus, corpus_name, datafile, save_dir):
    print("Start preparing training data ...")
    voc, pairs = readVocs(datafile, corpus_name)
    print("Read {!s} sentence pairs".format(len(pairs)))
    pairs = filterPairs(pairs)
    print("Trimmed to {!s} sentence pairs".format(len(pairs)))
    print("Counting words...")
    for pair in pairs:
        voc.addSentence(pair[0])
        voc.addSentence(pair[1])
    print("Counted words:", voc.num_words)
    return voc, pairs


# voc와 pairs를 읽고 재구성합니다
save_dir = os.path.join("data", "save")
voc, pairs = loadPrepareData(corpus, corpus_name, datafile, save_dir)
# 검증을 위해 pairs의 일부 내용을 출력해 봅니다
print("\npairs:")
for pair in pairs[:10]:
    print(pair)

Out:

Start preparing training data ...
Reading lines...
Read 221282 sentence pairs
Trimmed to 64271 sentence pairs
Counting words...
Counted words: 18008

pairs:
['there .', 'where ?']
['you have my word . as a gentleman', 'you re sweet .']
['hi .', 'looks like things worked out tonight huh ?']
['you know chastity ?', 'i believe we share an art instructor']
['have fun tonight ?', 'tons']
['well no . . .', 'then that s all you had to say .']
['then that s all you had to say .', 'but']
['but', 'you always been this selfish ?']
['do you listen to this crap ?', 'what crap ?']
['what good stuff ?', 'the real you .']

학습 단계가 빨리 수렴하도록 하는 또 다른 전략은 자주 쓰이지 않는 단어를 어휘집에서 제거하는 것입니다. 피처 공간의 크기를 줄이면 모델이 학습을 통해 근사하려는 함수의 난이도를 낮추는 효과도 있습니다. 우리는 이를 두 단계로 나눠 진행하려 합니다.

  1. voc.trim 함수를 이용하여 MIN_COUNT 라는 기준 이하의 단어를 제거합니다.

  2. 제거하기로 한 단어를 포함하는 경우를 pairs에서 제외합니다

MIN_COUNT = 3    # 제외할 단어의 기준이 되는 등장 횟수

def trimRareWords(voc, pairs, MIN_COUNT):
    # MIN_COUNT 미만으로 사용된 단어는 voc에서 제외합니다
    voc.trim(MIN_COUNT)
    # 제외할 단어가 포함된 경우를 pairs에서도 제외합니다
    keep_pairs = []
    for pair in pairs:
        input_sentence = pair[0]
        output_sentence = pair[1]
        keep_input = True
        keep_output = True
        # 입력 문장을 검사합니다
        for word in input_sentence.split(' '):
            if word not in voc.word2index:
                keep_input = False
                break
        # 출력 문장을 검사합니다
        for word in output_sentence.split(' '):
            if word not in voc.word2index:
                keep_output = False
                break

        # 입출력 문장에 제외하기로 한 단어를 포함하지 않는 경우만을 남겨둡니다
        if keep_input and keep_output:
            keep_pairs.append(pair)

    print("Trimmed from {} pairs to {}, {:.4f} of total".format(len(pairs), len(keep_pairs), len(keep_pairs) / len(pairs)))
    return keep_pairs


# voc와 pairs를 정돈합니다
pairs = trimRareWords(voc, pairs, MIN_COUNT)

Out:

keep_words 7823 / 18005 = 0.4345
Trimmed from 64271 pairs to 53165, 0.8272 of total

모델을 위한 데이터 준비하기

상당한 노력을 기울여 데이터를 전처리하고, 잘 정리하여 어휘집 객체와 문장 쌍의 리스트 형태로 만들어두긴 했지만, 결국 우리가 만들 모델에서 사용하는 입력은 수치 값으로 이루어진 torch 텐서입니다. 처리한 데이터를 모델에 맞는 형태로 준비하는 방법의 하나가 seq2seq 변환 튜토리얼 에 나와 있습니다. 이 튜토리얼에서는 배치 크기로 1을 사용하며, 이는 즉 문장에 등장하는 단어를 어휘집에서의 인덱스로 변환하여 모델에 제공하기만 하면 된다는 의미입니다.

그래도 여러분이 학습 속도나 GPU 병렬 처리 용량을 향상하고 싶다면 미니배치를 이용하여 학습해야 할 것입니다.

미니배치를 사용한다는 것은 배치에 포함된 문장 길이가 달라질 수 있다는 점에 유의해야 한다는 것을 뜻합니다. 같은 배치 안에서 크기가 다른 문장을 처리하기 위해서는 배치용 입력 텐서의 모양을 (max_length, batch_size) 로 맞춰야 합니다. 이때 max_length 보다 짧은 문장에 대해서는 EOS 토큰 뒤에 제로 토큰을 덧붙이면 됩니다.

영어로 된 문장을 텐서로 변환하기 위해 단순히 그에 대응하는 인덱스를 사용하고(indexesFromSentence) 제로 토큰을 패딩한다고 해봅시다. 그러면 텐서의 모양이 (batch_size, max_length) 이 되고, 첫 번째 차원에 대해 인덱싱을 수행하면 모든 시간대별 문장이 전부 반환될 것입니다. 그러나 우리는 배치를 시간에 따라, 그리고 배치에 포함된 모든 문장에 대해 인덱싱할 수도 있어야 합니다. 따라서 우리는 입력 배치의 모양을 뒤집어서 (max_length, batch_size) 형태로 만들 것입니다. 그러고 난 후에 첫 번째 차원에 대해 인덱싱하면 배치에 포함된 모든 문장을 시간에 대해 인덱싱한 결과를 반환하게 됩니다. 우리는 이 뒤집기 작업을 zeroPadding 함수를 이용하여 묵시적으로 수행할 것입니다.

batches

inputVar 함수는 문장을 텐서로 변환하는, 그리고 궁극적으로는 제로 패딩하여 올바른 모양으로 맞춘 텐서를 만드는 작업을 수행합니다. 이 함수는 각 배치에 포함된 시퀀스의 길이(lengths)로 구성된 텐서도 같이 반환합니다. 그리고 우리는 이를 나중에 디코더로 넘겨줄 것입니다.

outputVar 함수는 inputVar 와 비슷한 작업을 수행하지만, lengths 텐서를 반환하는 대신에 이진 마스크로 구성된 텐서와 목표 문장의 최대 길이를 같이 반환합니다. 이진 마스크 텐서는 출력에 해당하는 목표 텐서와 그 모양이 같지만, 패딩 토큰(PAD_token)에 해당하는 경우에는 값이 0이며 나머지 경우의 값은 1입니다.

batch2TrainData 는 단순히 여러 쌍을 입력으로 받아서, 앞서 설명한 함수를 이용하여 입력 및 목표 텐서를 구하여 반환합니다.

def indexesFromSentence(voc, sentence):
    return [voc.word2index[word] for word in sentence.split(' ')] + [EOS_token]


def zeroPadding(l, fillvalue=PAD_token):
    return list(itertools.zip_longest(*l, fillvalue=fillvalue))

def binaryMatrix(l, value=PAD_token):
    m = []
    for i, seq in enumerate(l):
        m.append([])
        for token in seq:
            if token == PAD_token:
                m[i].append(0)
            else:
                m[i].append(1)
    return m

# 입력 시퀀스 텐서에 패딩한 결과와 lengths를 반환합니다
def inputVar(l, voc):
    indexes_batch = [indexesFromSentence(voc, sentence) for sentence in l]
    lengths = torch.tensor([len(indexes) for indexes in indexes_batch])
    padList = zeroPadding(indexes_batch)
    padVar = torch.LongTensor(padList)
    return padVar, lengths

# 패딩한 목표 시퀀스 텐서, 패딩 마스크, 그리고 최대 목표 길이를 반환합니다
def outputVar(l, voc):
    indexes_batch = [indexesFromSentence(voc, sentence) for sentence in l]
    max_target_len = max([len(indexes) for indexes in indexes_batch])
    padList = zeroPadding(indexes_batch)
    mask = binaryMatrix(padList)
    mask = torch.ByteTensor(mask)
    padVar = torch.LongTensor(padList)
    return padVar, mask, max_target_len

# 입력 배치를 이루는 쌍에 대한 모든 아이템을 반환합니다
def batch2TrainData(voc, pair_batch):
    pair_batch.sort(key=lambda x: len(x[0].split(" ")), reverse=True)
    input_batch, output_batch = [], []
    for pair in pair_batch:
        input_batch.append(pair[0])
        output_batch.append(pair[1])
    inp, lengths = inputVar(input_batch, voc)
    output, mask, max_target_len = outputVar(output_batch, voc)
    return inp, lengths, output, mask, max_target_len


# 검증용 예시
small_batch_size = 5
batches = batch2TrainData(voc, [random.choice(pairs) for _ in range(small_batch_size)])
input_variable, lengths, target_variable, mask, max_target_len = batches

print("input_variable:", input_variable)
print("lengths:", lengths)
print("target_variable:", target_variable)
print("mask:", mask)
print("max_target_len:", max_target_len)

Out:

input_variable: tensor([[  25,   94,  101,  536,  318],
        [ 200,   25,   37,  709,    4],
        [  64,  359, 1471, 3381,    2],
        [1216,    7,    4,    2,    0],
        [1199,  637,    2,    0,    0],
        [   4,    6,    0,    0,    0],
        [   2,    2,    0,    0,    0]])
lengths: tensor([7, 7, 5, 4, 3])
target_variable: tensor([[  25, 2387,   43,    7,  218],
        [ 348,   67,    6,   18,  188],
        [  45,    4,    2,   36,  242],
        [ 115,    2,    0,    6,  188],
        [  12,    0,    0,    2,   76],
        [ 403,    0,    0,    0,    4],
        [  98,    0,    0,    0,    2],
        [   7,    0,    0,    0,    0],
        [   4,    0,    0,    0,    0],
        [   2,    0,    0,    0,    0]])
mask: tensor([[1, 1, 1, 1, 1],
        [1, 1, 1, 1, 1],
        [1, 1, 1, 1, 1],
        [1, 1, 0, 1, 1],
        [1, 0, 0, 1, 1],
        [1, 0, 0, 0, 1],
        [1, 0, 0, 0, 1],
        [1, 0, 0, 0, 0],
        [1, 0, 0, 0, 0],
        [1, 0, 0, 0, 0]], dtype=torch.uint8)
max_target_len: 10

모델 정의하기

Seq2Seq 모델

우리 챗봇의 두뇌에 해당하는 모델은 sequence-to-sequence (seq2seq) 모델입니다. seq2seq 모델의 목표는 가변 길이 시퀀스를 입력으로 받고, 크기가 고정된 모델을 이용하여, 가변 길이 시퀀스를 출력으로 반환하는 것입니다.

Sutskever 등 은 두 개의 독립된 순환 신경망을 같이 이용하여 이러한 목적을 달성할 수 있음을 발견했습니다. RNN 하나는 인코더 로, 가변 길이 입력 시퀀스를 고정된 길이의 문맥 벡터(context vector)로 인코딩합니다. 이론상 문맥 벡터(RNN의 마지막 은닉 레이어)는 봇에게 입력으로 주어지는 질의 문장에 대한 의미론적 정보를 담고 있을 것입니다. 두 번째 RNN은 디코더 입니다. 디코더는 단어 하나와 문맥 벡터를 입력으로 받고, 시퀀스의 다음 단어가 무엇일지를 추론하여 반환하며, 다음 단계에서 사용할 은닉 상태도 같이 반환합니다.

model

그림 출처: https://jeddy92.github.io/JEddy92.github.io/ts_seq2seq_intro/

인코더

인코더 RNN은 입력 시퀀스를 토큰 단위로(예를 들어, 단어 단위로) 한번에 하나씩 살펴보며 진행합니다. 그리고 각 단계마다 “출력” 벡터와 “은닉 상태” 벡터를 반환합니다. 은닉 상태 벡터는 다음 단계를 진행할 때 같이 사용되며, 출력 벡터는 차례대로 기록됩니다. 인코더는 시퀀스의 각 지점에 대해 파악한 문맥을 고차원 공간에 있는 점들의 집합으로 변환합니다. 나중에 디코더는 이를 이용하여 주어진 문제에 대해 의미 있는 출력을 구할 것입니다.

인코더의 핵심 부분에는 다중 레이어 게이트 순환 유닛(multi-layered Gated Recurrent Unit)이 있습니다. 이는 Cho 등 이 2014년에 고안한 것입니다. 우리는 GRU를 양방향으로 변환한 형태를 사용할 것이며, 이는 본질적으로 두 개의 독립된 RNN이 존재한다는 의미입니다. 하나는 입력 시퀀스를 원래 시퀀스에서의 순서로 처리하며, 다른 하나는 입력 시퀀스를 역순으로 처리합니다. 단계마다 각 네트워크의 출력을 합산합니다. 양방향 GRU를 사용하면 과거와 미래의 문맥을 함께 인코딩할 수 있다는 장점이 있습니다.

양방향 RNN:

rnn_bidir

그림 출처: https://colah.github.io/posts/2015-09-NN-Types-FP/

embedding 레이어가 단어 인덱스를 임의 크기의 피처 공간으로 인코딩하는 데 사용되었음에 유의하기 바랍니다. 우리의 모델에서는 이 레이어가 각 단어를 크기가 hidden_size 인 피처 공간으로 매핑할 것입니다. 학습을 거치면 서로 뜻이 유사한 단어는 의미적으로 유사하게 인코딩될 것입니다.

마지막으로, RNN 모듈에 패딩된 배치를 보내려면 RNN과 연결된 부분에서 패킹 및 언패킹하는 작업을 수행해야 합니다. 각각은 nn.utils.rnn.pack_padded_sequencenn.utils.rnn.pad_packed_sequence 를 통해 수행할 수 있습니다.

계산 그래프:

  1. 단어 인덱스를 임베딩으로 변환합니다.

  2. RNN 모듈을 위한 패딩된 배치 시퀀스를 패킹합니다.

  3. GRU로 포워드 패스를 수행합니다.

  4. 패딩을 언패킹합니다.

  5. 양방향 GRU의 출력을 합산합니다.

  6. 출력과 마지막 은닉 상태를 반환합니다.

입력:

  • input_seq: 입력 시퀀스 배치. shape=(max_length, batch_size)

  • input_lengths: 배치에 포함된 각 문장의 길이로 구성된 리스트. shape=(batch_size)

  • hidden: 은닉 상태. shape=(n_layers x num_directions, batch_size, hidden_size)

출력:

  • outputs: GRU의 마지막 은닉 레이어에 대한 출력 피처 값(양방향 (출력을 합산한 것). shape=(max_length, batch_size, hidden_size)

  • hidden: GRU의 최종 은닉 상태. shape=(n_layers x num_directions, batch_size, hidden_size)

class EncoderRNN(nn.Module):
    def __init__(self, hidden_size, embedding, n_layers=1, dropout=0):
        super(EncoderRNN, self).__init__()
        self.n_layers = n_layers
        self.hidden_size = hidden_size
        self.embedding = embedding

        # GRU를 초기화합니다. input_size와 hidden_size 패러미터는 둘 다 'hidden_size'로
        # 둡니다. 이는 우리 입력의 크기가 hideen_size 만큼의 피처를 갖는 단어 임베딩이기
        # 때문입니다.
        self.gru = nn.GRU(hidden_size, hidden_size, n_layers,
                          dropout=(0 if n_layers == 1 else dropout), bidirectional=True)

    def forward(self, input_seq, input_lengths, hidden=None):
        # 단어 인덱스를 임베딩으로 변환합니다
        embedded = self.embedding(input_seq)
        # RNN 모듈을 위한 패딩된 배치 시퀀스를 패킹합니다
        packed = nn.utils.rnn.pack_padded_sequence(embedded, input_lengths)
        # GRU로 포워드 패스를 수행합니다
        outputs, hidden = self.gru(packed, hidden)
        # 패딩을 언패킹합니다
        outputs, _ = nn.utils.rnn.pad_packed_sequence(outputs)
        # 양방향 GRU의 출력을 합산합니다
        outputs = outputs[:, :, :self.hidden_size] + outputs[:, : ,self.hidden_size:]
        # 출력과 마지막 은닉 상태를 반환합니다
        return outputs, hidden

디코더

디코더 RNN은 토큰 단위로 응답 문장을 생성하는 역할을 수행합니다. 이때 인코더의 문백 벡터를 사용하며, 내부 은닉 상태에 따라 시퀀스의 다음 단어를 생성하게 됩니다. 디코더는 EOS_token, 즉 문장의 끝을 나타내는 토큰을 출력할 때까지 계속 단어를 생성합니다. 원래의 seq2seq 디코더에는 알려진 문제점이 있습니다. 만약 우리가 입력 시퀀스의 의미를 인코딩할 때 문맥 벡터에만 전적으로 의존한다면, 그 과정 중에 정보 손실이 일어날 가능성이 높다는 것입니다. 이는 특히 입력 시퀀스의 길이가 길 때 그러하며, 이 때문에 디코더의 기능이 크게 제한될 수 있습니다.

이를 해결하기 위한 방편으로, Bahdanau 등 은 ‘어텐션 메커니즘’을 고안했습니다. 이는 디코더가 매 단계에 대해 고정된 문맥을 계속 사용하는 것이 아니라, 입력 시퀀스의 특정 부분에 집중하게 하는 방식입니다.

높은 차원에서 이야기 하자면, 어텐션은 디코더의 현재 은닉 상태와 인코더의 출력을 바탕으로 계산됩니다. 출력되는 어텐션 가중치는 입력 시퀀스와 동일한 모양을 가집니다. 따라서 이를 인코더의 출력과 곱할 수 있고, 그 결과로 얻게 되는 가중치 합은 인코더의 출력에서 어느 부분에 집중해야 할지를 알려줍니다. Sean Robertson 의 그림에 이러한 내용이 잘 설명되어 있습니다.

attn2

Luong 등 은 Bahdanau의 기초 연구를 더욱 발전시킨 ‘전역(global) 어텐션’을 제안했습니다. ‘전역 어텐션’의 핵심적인 차이점은 인코더의 은닉 상태를 모두 고려한다는 점입니다. 이는 Bahdanau 등의 ‘지역(local) 어텐션’ 방식이 현재 시점에 대한 인코더의 은닉 상태만을 고려한다는 점과 다른 부분입니다. ‘전역 어텐션’의 또 다른 차이점은 어텐션에 대한 가중치, 혹은 에너지를 계산할 때 현재 시점에 대한 디코더의 은닉 상태만을 사용한다는 점입니다. Bahdanau 등은 어텐션을 계산할 때 디코더의 이전 단계 상태에 대한 정보를 활용합니다. 또한 Luong 등의 방법에서는 인코더의 출력과 디코더의 출력에 대한 어텐션 에너지를 계산하는 방법을 제공하며, 이를 ‘점수 함수(score function)’라 부릅니다.

scores

이때 \(h_t\) 는 목표 디코더의 현재 상태를, \(\bar{h}_s\) 는 인코더의 모든 상태를 뜻합니다.

종합해 보면, 전역 어텐션 메커니즘을 다음 그림과 같이 요약할 수 있을 것입니다. 우리가 ‘어텐션 레이어’를 Attn 라는 독립적인 nn.Module 로 구현할 것임에 유의하기 바랍니다. 이 모듈의 출력은 모양이 (batch_size, 1, max_length) 인 정규화된 softmax 가중치 텐서입니다.

global_attn
# Luong 어텐션 레이어
class Attn(nn.Module):
    def __init__(self, method, hidden_size):
        super(Attn, self).__init__()
        self.method = method
        if self.method not in ['dot', 'general', 'concat']:
            raise ValueError(self.method, "is not an appropriate attention method.")
        self.hidden_size = hidden_size
        if self.method == 'general':
            self.attn = nn.Linear(self.hidden_size, hidden_size)
        elif self.method == 'concat':
            self.attn = nn.Linear(self.hidden_size * 2, hidden_size)
            self.v = nn.Parameter(torch.FloatTensor(hidden_size))

    def dot_score(self, hidden, encoder_output):
        return torch.sum(hidden * encoder_output, dim=2)

    def general_score(self, hidden, encoder_output):
        energy = self.attn(encoder_output)
        return torch.sum(hidden * energy, dim=2)

    def concat_score(self, hidden, encoder_output):
        energy = self.attn(torch.cat((hidden.expand(encoder_output.size(0), -1, -1), encoder_output), 2)).tanh()
        return torch.sum(self.v * energy, dim=2)

    def forward(self, hidden, encoder_outputs):
        # Attention 가중치(에너지)를 제안된 방법에 따라 계산합니다
        if self.method == 'general':
            attn_energies = self.general_score(hidden, encoder_outputs)
        elif self.method == 'concat':
            attn_energies = self.concat_score(hidden, encoder_outputs)
        elif self.method == 'dot':
            attn_energies = self.dot_score(hidden, encoder_outputs)

        # max_length와 batch_size의 차원을 뒤집습니다
        attn_energies = attn_energies.t()

        # 정규화된 softmax 확률 점수를 반환합니다 (차원을 늘려서)
        return F.softmax(attn_energies, dim=1).unsqueeze(1)

이처럼 어텐션 서브모듈을 정의하고 나면 실제 디코더 모델을 구현할 수 있게 됩니다. 디코더에 대해서는 매 시간마다 배치를 하나씩 수동으로 제공하려 합니다. 이는 임베딩된 단어 텐서와 GRU 출력의 모양이 둘 다 (1, batch_size, hidden_size) 라는 의미입니다.

계산 그래프:

  1. 현재의 입력 단어에 대한 임베딩을 구합니다.

  2. 무방향 GRU로 포워드 패스를 수행합니다.

  3. (2)에서 구한 현재의 GRU 출력을 바탕으로 어텐션 가중치를 계산합니다.

  4. 인코더 출력에 어텐션을 곱하여 새로운 “가중치 합” 문백 벡터를 구합니다.

  5. Luong의 논문에 나온 식 5를 이용하여 가중치 문백 벡터와 GRU 출력을 결합합니다.

  6. Luong의 논문에 나온 식 6을 이용하여(softmax 없이) 다음 단어를 예측합니다.

  7. 출력과 마지막 은닉 상태를 반환합니다.

입력:

  • input_step: 입력 시퀀스 배치에 대한 한 단위 시간(한 단어). shape=(1, batch_size)

  • last_hidden: GRU의 마지막 은닉 레이어. shape=(n_layers x num_directions, batch_size, hidden_size)

  • encoder_outputs: 인코더 모델의 출력. shape=(max_length, batch_size, hidden_size)

출력:

  • output: 각 단어가 디코딩된 시퀀스에서 다음 단어로 사용되었을 때 적합할 확률을 나타내는 정규화된 softmax 텐서. shape=(batch_size, voc.num_words)

  • hidden: GRU의 마지막 은닉 상태. shape=(n_layers x num_directions, batch_size, hidden_size)

class LuongAttnDecoderRNN(nn.Module):
    def __init__(self, attn_model, embedding, hidden_size, output_size, n_layers=1, dropout=0.1):
        super(LuongAttnDecoderRNN, self).__init__()

        # 참조를 보존해 둡니다
        self.attn_model = attn_model
        self.hidden_size = hidden_size
        self.output_size = output_size
        self.n_layers = n_layers
        self.dropout = dropout

        # 레이어를 정의합니다
        self.embedding = embedding
        self.embedding_dropout = nn.Dropout(dropout)
        self.gru = nn.GRU(hidden_size, hidden_size, n_layers, dropout=(0 if n_layers == 1 else dropout))
        self.concat = nn.Linear(hidden_size * 2, hidden_size)
        self.out = nn.Linear(hidden_size, output_size)

        self.attn = Attn(attn_model, hidden_size)

    def forward(self, input_step, last_hidden, encoder_outputs):
        # 주의: 한 단위 시간에 대해 한 단계(단어)만을 수행합니다
        # 현재의 입력 단어에 대한 임베딩을 구합니다
        embedded = self.embedding(input_step)
        embedded = self.embedding_dropout(embedded)
        # 무방향 GRU로 포워드 패스를 수행합니다
        rnn_output, hidden = self.gru(embedded, last_hidden)
        # 현재의 GRU 출력을 바탕으로 어텐션 가중치를 계산합니다
        attn_weights = self.attn(rnn_output, encoder_outputs)
        # 인코더 출력에 어텐션을 곱하여 새로운 "가중치 합" 문백 벡터를 구합니다
        context = attn_weights.bmm(encoder_outputs.transpose(0, 1))
        # Luong의 논문에 나온 식 5를 이용하여 가중치 문백 벡터와 GRU 출력을 결합합니다
        rnn_output = rnn_output.squeeze(0)
        context = context.squeeze(1)
        concat_input = torch.cat((rnn_output, context), 1)
        concat_output = torch.tanh(self.concat(concat_input))
        # Luong의 논문에 나온 식 6을 이용하여 다음 단어를 예측합니다
        output = self.out(concat_output)
        output = F.softmax(output, dim=1)
        # 출력과 마지막 은닉 상태를 반환합니다
        return output, hidden

학습 프로시저 정의하기

Masked loss

우리는 패딩된 시퀀스 배치를 다루기 때문에 손실을 계산할 때 단순히 텐서의 모든 원소를 고려할 수는 없습니다. 우리는 maskNLLLoss 를 정의하여 디코더의 출력 텐서, 목표 텐서, 이진 마스크 텐서를 바탕으로 손실을 계산하려 합니다. 이 손실 함수에서는 마스크 텐서의 1 에 대응하는 원소에 대한 음의 로그 우도 값의 평균을 계산합니다.

def maskNLLLoss(inp, target, mask):
    nTotal = mask.sum()
    crossEntropy = -torch.log(torch.gather(inp, 1, target.view(-1, 1)).squeeze(1))
    loss = crossEntropy.masked_select(mask).mean()
    loss = loss.to(device)
    return loss, nTotal.item()

한 번의 학습 단계

train 함수에 학습을 한 단계(입력 배치 한 개에 대한) 진행하는 알고리즘이 나와 있습니다.

우리는 수렴이 잘 되도록 몇 가지 영리한 전략을 사용해보려 합니다.

  • 첫 번째 전략은 teacher forcing 을 사용하는 것입니다. 이는 teacher_forcing_ratio 로 정의된 확률에 따라, 디코더의 이번 단계 예측값 대신에 현재의 목표 단어를 디코더의 다음 입력 값으로 활용하는 것입니다. 이 기법은 디코더의 보조 바퀴처럼 작용하여 효율적으로 학습될 수 있게 도와 줍니다. 하지만 teacher forcing 기법은 추론 과정에서 모델이 불안정 해지도록 할 수도 있는데, 이는 디코더가 학습 과정에서 자신의 출력 시퀀스를 직접 만들어 볼 기회를 충분히 제공받지 못할 수 있기 때문입니다. 따라서 우리는 teacher_forcing_ratio 를 어떻게 설정해 두었는지에 주의를 기울여야 하며, 수렴이 빨리 되었다고 속아 넘어가서는 안 됩니다.

  • 우리가 구현한 두 번째 전략은 gradient clipping 입니다. 이는 소위 ‘그라디언트 폭발’ 문제를 해결하기 위해 널리 사용되는 기법입니다. 핵심은 그라디언트를 클리핑 하거나 임계값을 둠으로써, 그라디언트가 지수 함수적으로 증가하거나 오버플로를 일으키는(NaN) 경우를 막고, 비용 함수의 급격한 경사를 피하겠다는 것입니다.

grad_clip

그림 출처: Goodfellow 등 저. Deep Learning. 2016. https://www.deeplearningbook.org/

작업 절차:

  1. 전체 입력 배치에 대하여 인코더로 포워드 패스를 수행합니다.

  2. 디코더의 입력을 SOS_token로, 은닉 상태를 인코더의 마지막 은닉 상태로 초기화합니다.

  3. 입력 배치 시퀀스를 한 번에 하나씩 디코더로 포워드 패스합니다.

  4. Teacher forcing을 사용하는 경우, 디코더의 다음 입력을 현재의 목표로 둡니다. 그렇지 않으면 디코더의 다음 입력을 현재 디코더의 출력으로 둡니다.

  5. 손실을 계산하고 누적합니다.

  6. 역전파를 수행합니다.

  7. 그라디언트를 클리핑 합니다.

  8. 인코더 및 디코더 모델의 패러미터를 갱신합니다.

Warning

PyTorch의 RNN 모듈(RNN, LSTM, GRU)은 전체 입력 시퀀스(또는 시퀀스의 배치)를 단순히 넣어주기만 하면 다른 비순환 레이어처럼 사용할 수 있습니다. 우리는 encoder 에서 GRU 레이어를 이런 식으로 사용합니다. 그 안이 실제로 어떻게 되어 있는지를 살펴보면, 매 시간 단계마다 은닉 상태를 계산하는 반복 프로세스가 존재합니다. 또 다른 방법은, 이 모듈을 매번 한 단위 시간만큼 수행할 수도 있습니다. 그 경우에는 우리가 decoder 모델을 다룰 때처럼, 학습 과정에서 수동으로 시퀀스에 대해 반복 작업을 수행해 주어야 합니다. 이 모듈에 대해 모델의 개념을 확실히 갖고만 있다면, 순차 모델을 구현하는 것도 매우 단순할 것입니다.

def train(input_variable, lengths, target_variable, mask, max_target_len, encoder, decoder, embedding,
          encoder_optimizer, decoder_optimizer, batch_size, clip, max_length=MAX_LENGTH):

    # 제로 그라디언트
    encoder_optimizer.zero_grad()
    decoder_optimizer.zero_grad()

    # device 옵션을 설정합니다
    input_variable = input_variable.to(device)
    lengths = lengths.to(device)
    target_variable = target_variable.to(device)
    mask = mask.to(device)

    # 변수를 초기화합니다
    loss = 0
    print_losses = []
    n_totals = 0

    # 인코더로 포워드 패스를 수행합니다
    encoder_outputs, encoder_hidden = encoder(input_variable, lengths)

    # 초기 디코더 입력을 생성합니다(각 문장을 SOS 도큰으로 시작합니다)
    decoder_input = torch.LongTensor([[SOS_token for _ in range(batch_size)]])
    decoder_input = decoder_input.to(device)

    # 디코더의 초기 은닉 상태를 인코더의 마지막 은닉 상태로 둡니다
    decoder_hidden = encoder_hidden[:decoder.n_layers]

    # 이번 반복에서 teacher forcing을 사용할지를 결정합니다
    use_teacher_forcing = True if random.random() < teacher_forcing_ratio else False

    # 배치 시퀀스를 한 번에 하나씩 디코더로 포워드 패스합니다
    if use_teacher_forcing:
        for t in range(max_target_len):
            decoder_output, decoder_hidden = decoder(
                decoder_input, decoder_hidden, encoder_outputs
            )
            # Teacher forcing 사용: 다음 입력을 현재의 목표로 둡니다
            decoder_input = target_variable[t].view(1, -1)
            # 손실을 계산하고 누적합니다
            mask_loss, nTotal = maskNLLLoss(decoder_output, target_variable[t], mask[t])
            loss += mask_loss
            print_losses.append(mask_loss.item() * nTotal)
            n_totals += nTotal
    else:
        for t in range(max_target_len):
            decoder_output, decoder_hidden = decoder(
                decoder_input, decoder_hidden, encoder_outputs
            )
            # Teacher forcing 미사용: 다음 입력을 디코더의 출력으로 둡니다
            _, topi = decoder_output.topk(1)
            decoder_input = torch.LongTensor([[topi[i][0] for i in range(batch_size)]])
            decoder_input = decoder_input.to(device)
            # 손실을 계산하고 누적합니다
            mask_loss, nTotal = maskNLLLoss(decoder_output, target_variable[t], mask[t])
            loss += mask_loss
            print_losses.append(mask_loss.item() * nTotal)
            n_totals += nTotal

    # 역전파를 수행합니다
    loss.backward()

    # 그라디언트 클리핑: 그라디언트를 제자리에서 수정합니다
    _ = nn.utils.clip_grad_norm_(encoder.parameters(), clip)
    _ = nn.utils.clip_grad_norm_(decoder.parameters(), clip)

    # 모델의 가중치를 수정합니다
    encoder_optimizer.step()
    decoder_optimizer.step()

    return sum(print_losses) / n_totals

학습 단계

이제 마지막으로 전체 학습 프로시저와 데이터를 하나로 엮을 때가 되었습니다. trainIters 함수는 주어진 모델, optimizer, 데이터 등을 토대로 학습을 n_iterations 번의 단계만큼 진행하는 역할을 담당합니다. 이 함수는 자기 자신을 살 설명하고 있는 편인데, 무거운 작업을 train 함수에 옮겨 놓았기 때문입니다.

한 가지 주의할 점은 우리가 모델을 저장하려 할 때, 인코더와 디코더의 state_dicts (패러미터), optimizer의 state_dicts, 손실, 진행 단계 수 등을 tarball로 만들어 저장한다는 점입니다. 모델을 이러한 방식으로 저장하면 checkpoint에 대해 아주 높은 수준의 유연성을 확보할 수 있게 됩니다. Checkpoint를 불러오고 나면, 우리는 모델 패러미터를 이용하여 예측을 진행할 수도 있고, 이전에 멈췄던 부분부터 학습을 계속 진행할 수도 있게 됩니다.

def trainIters(model_name, voc, pairs, encoder, decoder, encoder_optimizer, decoder_optimizer, embedding, encoder_n_layers, decoder_n_layers, save_dir, n_iteration, batch_size, print_every, save_every, clip, corpus_name, loadFilename):

    # 각 단계에 대한 배치를 읽어옵니다
    training_batches = [batch2TrainData(voc, [random.choice(pairs) for _ in range(batch_size)])
                      for _ in range(n_iteration)]

    # 초기화
    print('Initializing ...')
    start_iteration = 1
    print_loss = 0
    if loadFilename:
        start_iteration = checkpoint['iteration'] + 1

    # 학습 루프
    print("Training...")
    for iteration in range(start_iteration, n_iteration + 1):
        training_batch = training_batches[iteration - 1]
        # 배치에서 각 필드를 읽어옵니다
        input_variable, lengths, target_variable, mask, max_target_len = training_batch

        # 배치에 대해 학습을 한 단계 진행합니다
        loss = train(input_variable, lengths, target_variable, mask, max_target_len, encoder,
                     decoder, embedding, encoder_optimizer, decoder_optimizer, batch_size, clip)
        print_loss += loss

        # 경과를 출력합니다
        if iteration % print_every == 0:
            print_loss_avg = print_loss / print_every
            print("Iteration: {}; Percent complete: {:.1f}%; Average loss: {:.4f}".format(iteration, iteration / n_iteration * 100, print_loss_avg))
            print_loss = 0

        # Checkpoint를 저장합니다
        if (iteration % save_every == 0):
            directory = os.path.join(save_dir, model_name, corpus_name, '{}-{}_{}'.format(encoder_n_layers, decoder_n_layers, hidden_size))
            if not os.path.exists(directory):
                os.makedirs(directory)
            torch.save({
                'iteration': iteration,
                'en': encoder.state_dict(),
                'de': decoder.state_dict(),
                'en_opt': encoder_optimizer.state_dict(),
                'de_opt': decoder_optimizer.state_dict(),
                'loss': loss,
                'voc_dict': voc.__dict__,
                'embedding': embedding.state_dict()
            }, os.path.join(directory, '{}_{}.tar'.format(iteration, 'checkpoint')))

평가 정의하기

모델을 학습시키고 나면 직접 봇과 대화를 나눠보고 싶어질 것입니다. 그러려면 먼저 모델이 인코딩된 입력을 어떻게 디코딩할지를 정의해줘야 합니다.

탐욕적 디코딩

탐욕적 디코딩(Greedy decoding)은 우리가 학습 단계에서 teacher forcing을 적용하지 않았을 때 사용한 디코딩 방법입니다. 달리 말하면, 각 단계에 대해 단순히 decoder_output 에서 가장 높은 softmax값을 갖는 단어를 선택하는 방식입니다. 이 디코딩 방법은 한 번의 단계에 대해서는 최적입니다.

우리는 탐욕적 디코딩 연산을 수행할 수 있도록 GreedySearchDecoder 클래스를 만들었습니다. 수행 과정에서 이 클래스의 인스턴스는 모양이 (input_seq length, 1) 인 입력 시퀀스(input_seq), 조종할 입력 길이(input_length) 텐서, 그리고 응답 문장 길이의 제한을 나타내는 max_length 를 입력으로 받습니다. 입력 시퀀서는 다음과 같은 계산 그래프에 의해 평가됩니다.

계산 그래프:

  1. 인코더 모델로 입력을 포워드 패스합니다.

  2. 인코더의 마지막 은닉 레이어가 디코더의 첫 번째 은닉 레이어의 입력이 되도록 준비합니다.

  3. 디코더의 첫 번째 입력을 SOS_token으로 초기화합니다.

  4. 디코더가 단어를 덧붙여 나갈 텐서를 초기화합니다.

  5. 반복적으로 각 단계마다 하나의 단어 토큰을 디코딩합니다.
    1. 디코더로의 포워드 패스를 수행합니다.

    2. 가장 가능성 높은 단어 토큰과 그 softmax 점수를 구합니다.

    3. 토큰과 점수를 기록합니다.

    4. 현재의 토큰을 디코더의 다음 입력으로 준비시킵니다.

  6. 단어 토큰과 점수를 모아서 반환합니다.

class GreedySearchDecoder(nn.Module):
    def __init__(self, encoder, decoder):
        super(GreedySearchDecoder, self).__init__()
        self.encoder = encoder
        self.decoder = decoder

    def forward(self, input_seq, input_length, max_length):
        # 인코더 모델로 입력을 포워드 패스합니다
        encoder_outputs, encoder_hidden = self.encoder(input_seq, input_length)
        # 인코더의 마지막 은닉 레이어가 디코더의 첫 번째 은닉 레이어의 입력이 되도록 준비합니다
        decoder_hidden = encoder_hidden[:decoder.n_layers]
        # 디코더의 첫 번째 입력을 SOS_token으로 초기화합니다
        decoder_input = torch.ones(1, 1, device=device, dtype=torch.long) * SOS_token
        # 디코더가 단어를 덧붙여 나갈 텐서를 초기화합니다
        all_tokens = torch.zeros([0], device=device, dtype=torch.long)
        all_scores = torch.zeros([0], device=device)
        # 반복적으로 각 단계마다 하나의 단어 토큰을 디코딩합니다
        for _ in range(max_length):
            # 디코더로의 포워드 패스를 수행합니다
            decoder_output, decoder_hidden = self.decoder(decoder_input, decoder_hidden, encoder_outputs)
            # 가장 가능성 높은 단어 토큰과 그 softmax 점수를 구합니다
            decoder_scores, decoder_input = torch.max(decoder_output, dim=1)
            # 토큰과 점수를 기록합니다
            all_tokens = torch.cat((all_tokens, decoder_input), dim=0)
            all_scores = torch.cat((all_scores, decoder_scores), dim=0)
            # 현재의 토큰을 디코더의 다음 입력으로 준비시킵니다(차원을 증가시켜서)
            decoder_input = torch.unsqueeze(decoder_input, 0)
        # 단어 토큰과 점수를 모아서 반환합니다
        return all_tokens, all_scores

내 텍스트 평가하기

이제 디코딩 모델을 정의했으니, 문자열로 된 입력 시퀀스를 평가하는 함수를 작성해볼 수 있을 것입니다. evaluate 함수에 입력 시퀀스를 낮은 레벨에서 어떻게 처리할지가 나와 있습니다. 우리는 먼저 문장을 batch_size==1 이고 단어 인덱스로 구성된 입력 배치 형태로 만듭니다. 이를 위해 문장의 각 단어를 그에 대응하는 인덱스로 변환하고, 차원을 뒤집어서 모델에 맞는 입력 형태로 변환합니다. 우리는 입력 시퀀스의 길이를 저장하고 있는 lengths 텐서도 만듭니다. 이 경우에는 lengths 가 스칼라 값이 되는데, 우리는 한 번에 한 문장만 평가하기 때문입니다(batch_size==1). 다음으로는 GreedySearchDecoder 의 객체(searcher)를 이용하여 응답 문장 텐서를 디코딩합니다. 마지막으로, 응답 인덱스를 단어로 변환하고 디코딩된 단어의 리스트를 반환합니다.

evaluateInput 은 우리의 챗봇에 대한 인터페이스 역할을 수행합니다. 이를 호출하면 입력 텍스트 필드가 생성되는데, 거기에 우리의 질의 문장을 입력해볼 수 있습니다. 입력 문장을 타이핑하고 엔터 를 누르면, 입력한 텍스트가 학습 데이터와 같은 방식으로 정규화되고, 최종적으로는 evaluate 함수에 입력으로 제공되어 디코딩된 출력 문장을 구하게 됩니다. 우리는 이러한 과정을 계속 반복하며, 이를 통해 ‘q’나 ‘quit’를 입력하기 전까지는 계속 채팅할 수 있습니다.

마지막으로, 만약 어휘집에 포함되어 있지 않은 단어를 포함하고 있는 문장이 입력되더라도 이를 예의 바르게 처리합니다. 즉 에러 메시지를 출력하고 사용자에게 새로운 문장을 입력해달라고 요구청합니다.

def evaluate(encoder, decoder, searcher, voc, sentence, max_length=MAX_LENGTH):
    ### 입력 시퀀스를 배치 형태로 만듭니다
    # 단어 -> 인덱스
    indexes_batch = [indexesFromSentence(voc, sentence)]
    # lengths 텐서를 만듭니다
    lengths = torch.tensor([len(indexes) for indexes in indexes_batch])
    # 배치의 차원을 뒤집어서 모델이 사용하는 형태로 만듭니다
    input_batch = torch.LongTensor(indexes_batch).transpose(0, 1)
    # 적절한 디바이스를 사용합니다
    input_batch = input_batch.to(device)
    lengths = lengths.to(device)
    # searcher를 이용하여 문장을 디코딩합니다
    tokens, scores = searcher(input_batch, lengths, max_length)
    # 인덱스 -> 단어
    decoded_words = [voc.index2word[token.item()] for token in tokens]
    return decoded_words


def evaluateInput(encoder, decoder, searcher, voc):
    input_sentence = ''
    while(1):
        try:
            # 입력 문장을 받아옵니다
            input_sentence = input('> ')
            # 종료 조건인지 검사합니다
            if input_sentence == 'q' or input_sentence == 'quit': break
            # 문장을 정규화합니다
            input_sentence = normalizeString(input_sentence)
            # 문장을 평가합니다
            output_words = evaluate(encoder, decoder, searcher, voc, input_sentence)
            # 응답 문장을 형식에 맞춰 출력합니다
            output_words[:] = [x for x in output_words if not (x == 'EOS' or x == 'PAD')]
            print('Bot:', ' '.join(output_words))

        except KeyError:
            print("Error: Encountered unknown word.")

모델 수행하기

마지막으로, 우리의 모델을 수행해 볼 시간입니다!

우리가 챗봇 모델을 학습할 때든 테스트할 때든, 우리는 각각의 인코더 및 디코더 모델을 초기화해줘야 합니다. 다음 블록에서는 우리가 원하는대로 설정을 맞추고, 처음부터 시작할지, 아니면 checkpoint를 불러올지 정하고, 모델을 빌드하고 초기화합니다. 성능을 최적화하기 위해서는 모델 설정을 여러가지로 바꿔 보면서 테스트해보기 바랍니다.

# 모델을 설정합니다
model_name = 'cb_model'
attn_model = 'dot'
#attn_model = 'general'
#attn_model = 'concat'
hidden_size = 500
encoder_n_layers = 2
decoder_n_layers = 2
dropout = 0.1
batch_size = 64

# 불러올 checkpoint를 설정합니다. 처음부터 시작할 때는 None으로 둡니다.
loadFilename = None
checkpoint_iter = 4000
#loadFilename = os.path.join(save_dir, model_name, corpus_name,
#                            '{}-{}_{}'.format(encoder_n_layers, decoder_n_layers, hidden_size),
#                            '{}_checkpoint.tar'.format(checkpoint_iter))


# loadFilename이 제공되는 경우에는 모델을 불러옵니다
if loadFilename:
    # 모델을 학습할 때와 같은 기기에서 불러오는 경우
    checkpoint = torch.load(loadFilename)
    # GPU에서 학습한 모델을 CPU로 불러오는 경우
    #checkpoint = torch.load(loadFilename, map_location=torch.device('cpu'))
    encoder_sd = checkpoint['en']
    decoder_sd = checkpoint['de']
    encoder_optimizer_sd = checkpoint['en_opt']
    decoder_optimizer_sd = checkpoint['de_opt']
    embedding_sd = checkpoint['embedding']
    voc.__dict__ = checkpoint['voc_dict']


print('Building encoder and decoder ...')
# 단어 임베딩을 초기화합니다
embedding = nn.Embedding(voc.num_words, hidden_size)
if loadFilename:
    embedding.load_state_dict(embedding_sd)
# 인코더 및 디코더 모델을 초기화합니다
encoder = EncoderRNN(hidden_size, embedding, encoder_n_layers, dropout)
decoder = LuongAttnDecoderRNN(attn_model, embedding, hidden_size, voc.num_words, decoder_n_layers, dropout)
if loadFilename:
    encoder.load_state_dict(encoder_sd)
    decoder.load_state_dict(decoder_sd)
# 적절한 디바이스를 사용합니다
encoder = encoder.to(device)
decoder = decoder.to(device)
print('Models built and ready to go!')

Out:

Building encoder and decoder ...
Models built and ready to go!

학습 수행하기

모델을 학습해보고 싶다면 다음 블록을 수행하면 됩니다.

먼저 학습 패러미터를 설정하고, optimizer를 초기화한 뒤, 마지막으로 trainIters 함수를 호출하여 학습 단계를 진행합니다.

# 학습 및 최적화 설정
clip = 50.0
teacher_forcing_ratio = 1.0
learning_rate = 0.0001
decoder_learning_ratio = 5.0
n_iteration = 4000
print_every = 1
save_every = 500

# Dropout 레이어를 학습 모드로 둡니다
encoder.train()
decoder.train()

# Optimizer를 초기화합니다
print('Building optimizers ...')
encoder_optimizer = optim.Adam(encoder.parameters(), lr=learning_rate)
decoder_optimizer = optim.Adam(decoder.parameters(), lr=learning_rate * decoder_learning_ratio)
if loadFilename:
    encoder_optimizer.load_state_dict(encoder_optimizer_sd)
    decoder_optimizer.load_state_dict(decoder_optimizer_sd)

# cuda가 있다면 cuda를 설정합니다
for state in encoder_optimizer.state.values():
    for k, v in state.items():
        if isinstance(v, torch.Tensor):
            state[k] = v.cuda()

for state in decoder_optimizer.state.values():
    for k, v in state.items():
        if isinstance(v, torch.Tensor):
            state[k] = v.cuda()

# 학습 단계를 수행합니다
print("Starting Training!")
trainIters(model_name, voc, pairs, encoder, decoder, encoder_optimizer, decoder_optimizer,
           embedding, encoder_n_layers, decoder_n_layers, save_dir, n_iteration, batch_size,
           print_every, save_every, clip, corpus_name, loadFilename)

Out:

Building optimizers ...
Starting Training!
Initializing ...
Training...
Iteration: 1; Percent complete: 0.0%; Average loss: 8.9615
Iteration: 2; Percent complete: 0.1%; Average loss: 8.8544
Iteration: 3; Percent complete: 0.1%; Average loss: 8.7023
Iteration: 4; Percent complete: 0.1%; Average loss: 8.4168
Iteration: 5; Percent complete: 0.1%; Average loss: 7.8623
Iteration: 6; Percent complete: 0.1%; Average loss: 7.4846
Iteration: 7; Percent complete: 0.2%; Average loss: 7.0624
Iteration: 8; Percent complete: 0.2%; Average loss: 7.0901
Iteration: 9; Percent complete: 0.2%; Average loss: 7.2066
Iteration: 10; Percent complete: 0.2%; Average loss: 7.0991
Iteration: 11; Percent complete: 0.3%; Average loss: 6.5571
Iteration: 12; Percent complete: 0.3%; Average loss: 5.9773
Iteration: 13; Percent complete: 0.3%; Average loss: 5.4542
Iteration: 14; Percent complete: 0.4%; Average loss: 5.6626
Iteration: 15; Percent complete: 0.4%; Average loss: 5.5726
Iteration: 16; Percent complete: 0.4%; Average loss: 5.5811
Iteration: 17; Percent complete: 0.4%; Average loss: 5.4539
Iteration: 18; Percent complete: 0.4%; Average loss: 5.3156
Iteration: 19; Percent complete: 0.5%; Average loss: 4.8697
Iteration: 20; Percent complete: 0.5%; Average loss: 5.0321
Iteration: 21; Percent complete: 0.5%; Average loss: 5.1108
Iteration: 22; Percent complete: 0.5%; Average loss: 5.2024
Iteration: 23; Percent complete: 0.6%; Average loss: 4.8962
Iteration: 24; Percent complete: 0.6%; Average loss: 4.8467
Iteration: 25; Percent complete: 0.6%; Average loss: 4.9299
Iteration: 26; Percent complete: 0.7%; Average loss: 4.9358
Iteration: 27; Percent complete: 0.7%; Average loss: 4.8575
Iteration: 28; Percent complete: 0.7%; Average loss: 4.8051
Iteration: 29; Percent complete: 0.7%; Average loss: 4.8730
Iteration: 30; Percent complete: 0.8%; Average loss: 4.7126
Iteration: 31; Percent complete: 0.8%; Average loss: 4.7846
Iteration: 32; Percent complete: 0.8%; Average loss: 4.5240
Iteration: 33; Percent complete: 0.8%; Average loss: 4.7871
Iteration: 34; Percent complete: 0.9%; Average loss: 4.8435
Iteration: 35; Percent complete: 0.9%; Average loss: 4.9138
Iteration: 36; Percent complete: 0.9%; Average loss: 4.5602
Iteration: 37; Percent complete: 0.9%; Average loss: 4.5048
Iteration: 38; Percent complete: 0.9%; Average loss: 4.7657
Iteration: 39; Percent complete: 1.0%; Average loss: 4.6938
Iteration: 40; Percent complete: 1.0%; Average loss: 4.6905
Iteration: 41; Percent complete: 1.0%; Average loss: 4.6822
Iteration: 42; Percent complete: 1.1%; Average loss: 4.6829
Iteration: 43; Percent complete: 1.1%; Average loss: 4.9798
Iteration: 44; Percent complete: 1.1%; Average loss: 4.7588
Iteration: 45; Percent complete: 1.1%; Average loss: 4.6366
Iteration: 46; Percent complete: 1.1%; Average loss: 4.7790
Iteration: 47; Percent complete: 1.2%; Average loss: 4.5686
Iteration: 48; Percent complete: 1.2%; Average loss: 4.7021
Iteration: 49; Percent complete: 1.2%; Average loss: 4.6842
Iteration: 50; Percent complete: 1.2%; Average loss: 4.6113
Iteration: 51; Percent complete: 1.3%; Average loss: 4.8767
Iteration: 52; Percent complete: 1.3%; Average loss: 4.3779
Iteration: 53; Percent complete: 1.3%; Average loss: 4.5862
Iteration: 54; Percent complete: 1.4%; Average loss: 4.3230
Iteration: 55; Percent complete: 1.4%; Average loss: 4.6308
Iteration: 56; Percent complete: 1.4%; Average loss: 4.5964
Iteration: 57; Percent complete: 1.4%; Average loss: 4.5151
Iteration: 58; Percent complete: 1.5%; Average loss: 4.5563
Iteration: 59; Percent complete: 1.5%; Average loss: 4.5992
Iteration: 60; Percent complete: 1.5%; Average loss: 4.5209
Iteration: 61; Percent complete: 1.5%; Average loss: 4.3537
Iteration: 62; Percent complete: 1.6%; Average loss: 4.5533
Iteration: 63; Percent complete: 1.6%; Average loss: 4.5850
Iteration: 64; Percent complete: 1.6%; Average loss: 4.5646
Iteration: 65; Percent complete: 1.6%; Average loss: 4.5251
Iteration: 66; Percent complete: 1.7%; Average loss: 4.7983
Iteration: 67; Percent complete: 1.7%; Average loss: 4.5491
Iteration: 68; Percent complete: 1.7%; Average loss: 4.4872
Iteration: 69; Percent complete: 1.7%; Average loss: 4.6204
Iteration: 70; Percent complete: 1.8%; Average loss: 4.6458
Iteration: 71; Percent complete: 1.8%; Average loss: 4.4539
Iteration: 72; Percent complete: 1.8%; Average loss: 4.4942
Iteration: 73; Percent complete: 1.8%; Average loss: 4.4024
Iteration: 74; Percent complete: 1.8%; Average loss: 4.8446
Iteration: 75; Percent complete: 1.9%; Average loss: 4.5835
Iteration: 76; Percent complete: 1.9%; Average loss: 4.4976
Iteration: 77; Percent complete: 1.9%; Average loss: 4.6159
Iteration: 78; Percent complete: 1.9%; Average loss: 4.3079
Iteration: 79; Percent complete: 2.0%; Average loss: 4.4867
Iteration: 80; Percent complete: 2.0%; Average loss: 4.3033
Iteration: 81; Percent complete: 2.0%; Average loss: 4.2497
Iteration: 82; Percent complete: 2.1%; Average loss: 4.3319
Iteration: 83; Percent complete: 2.1%; Average loss: 4.6114
Iteration: 84; Percent complete: 2.1%; Average loss: 4.4444
Iteration: 85; Percent complete: 2.1%; Average loss: 4.2973
Iteration: 86; Percent complete: 2.1%; Average loss: 4.2523
Iteration: 87; Percent complete: 2.2%; Average loss: 4.7035
Iteration: 88; Percent complete: 2.2%; Average loss: 4.5964
Iteration: 89; Percent complete: 2.2%; Average loss: 4.3912
Iteration: 90; Percent complete: 2.2%; Average loss: 4.3945
Iteration: 91; Percent complete: 2.3%; Average loss: 4.4530
Iteration: 92; Percent complete: 2.3%; Average loss: 4.3078
Iteration: 93; Percent complete: 2.3%; Average loss: 4.3691
Iteration: 94; Percent complete: 2.4%; Average loss: 4.5729
Iteration: 95; Percent complete: 2.4%; Average loss: 4.3206
Iteration: 96; Percent complete: 2.4%; Average loss: 4.4443
Iteration: 97; Percent complete: 2.4%; Average loss: 4.3996
Iteration: 98; Percent complete: 2.5%; Average loss: 4.5063
Iteration: 99; Percent complete: 2.5%; Average loss: 4.5920
Iteration: 100; Percent complete: 2.5%; Average loss: 4.5813
Iteration: 101; Percent complete: 2.5%; Average loss: 4.3174
Iteration: 102; Percent complete: 2.5%; Average loss: 4.4041
Iteration: 103; Percent complete: 2.6%; Average loss: 4.5551
Iteration: 104; Percent complete: 2.6%; Average loss: 4.4138
Iteration: 105; Percent complete: 2.6%; Average loss: 4.6544
Iteration: 106; Percent complete: 2.6%; Average loss: 4.0751
Iteration: 107; Percent complete: 2.7%; Average loss: 4.3358
Iteration: 108; Percent complete: 2.7%; Average loss: 4.3980
Iteration: 109; Percent complete: 2.7%; Average loss: 4.4049
Iteration: 110; Percent complete: 2.8%; Average loss: 4.2932
Iteration: 111; Percent complete: 2.8%; Average loss: 4.3918
Iteration: 112; Percent complete: 2.8%; Average loss: 4.2454
Iteration: 113; Percent complete: 2.8%; Average loss: 4.4025
Iteration: 114; Percent complete: 2.9%; Average loss: 4.2536
Iteration: 115; Percent complete: 2.9%; Average loss: 4.5109
Iteration: 116; Percent complete: 2.9%; Average loss: 4.3479
Iteration: 117; Percent complete: 2.9%; Average loss: 4.4633
Iteration: 118; Percent complete: 2.9%; Average loss: 4.5232
Iteration: 119; Percent complete: 3.0%; Average loss: 4.5811
Iteration: 120; Percent complete: 3.0%; Average loss: 4.4596
Iteration: 121; Percent complete: 3.0%; Average loss: 4.4024
Iteration: 122; Percent complete: 3.0%; Average loss: 4.3763
Iteration: 123; Percent complete: 3.1%; Average loss: 4.4115
Iteration: 124; Percent complete: 3.1%; Average loss: 4.2496
Iteration: 125; Percent complete: 3.1%; Average loss: 4.1680
Iteration: 126; Percent complete: 3.1%; Average loss: 4.3440
Iteration: 127; Percent complete: 3.2%; Average loss: 4.5335
Iteration: 128; Percent complete: 3.2%; Average loss: 4.2567
Iteration: 129; Percent complete: 3.2%; Average loss: 4.2014
Iteration: 130; Percent complete: 3.2%; Average loss: 4.2073
Iteration: 131; Percent complete: 3.3%; Average loss: 4.1413
Iteration: 132; Percent complete: 3.3%; Average loss: 4.4261
Iteration: 133; Percent complete: 3.3%; Average loss: 4.1026
Iteration: 134; Percent complete: 3.4%; Average loss: 4.1035
Iteration: 135; Percent complete: 3.4%; Average loss: 4.3772
Iteration: 136; Percent complete: 3.4%; Average loss: 4.1588
Iteration: 137; Percent complete: 3.4%; Average loss: 4.0663
Iteration: 138; Percent complete: 3.5%; Average loss: 4.0826
Iteration: 139; Percent complete: 3.5%; Average loss: 4.1710
Iteration: 140; Percent complete: 3.5%; Average loss: 4.4940
Iteration: 141; Percent complete: 3.5%; Average loss: 4.3392
Iteration: 142; Percent complete: 3.5%; Average loss: 4.2300
Iteration: 143; Percent complete: 3.6%; Average loss: 4.4431
Iteration: 144; Percent complete: 3.6%; Average loss: 4.1758
Iteration: 145; Percent complete: 3.6%; Average loss: 4.2632
Iteration: 146; Percent complete: 3.6%; Average loss: 4.4169
Iteration: 147; Percent complete: 3.7%; Average loss: 4.3127
Iteration: 148; Percent complete: 3.7%; Average loss: 4.2589
Iteration: 149; Percent complete: 3.7%; Average loss: 4.1161
Iteration: 150; Percent complete: 3.8%; Average loss: 4.3664
Iteration: 151; Percent complete: 3.8%; Average loss: 4.2785
Iteration: 152; Percent complete: 3.8%; Average loss: 4.4078
Iteration: 153; Percent complete: 3.8%; Average loss: 4.2294
Iteration: 154; Percent complete: 3.9%; Average loss: 4.5065
Iteration: 155; Percent complete: 3.9%; Average loss: 4.2910
Iteration: 156; Percent complete: 3.9%; Average loss: 4.2238
Iteration: 157; Percent complete: 3.9%; Average loss: 4.1167
Iteration: 158; Percent complete: 4.0%; Average loss: 4.1026
Iteration: 159; Percent complete: 4.0%; Average loss: 3.9282
Iteration: 160; Percent complete: 4.0%; Average loss: 4.0791
Iteration: 161; Percent complete: 4.0%; Average loss: 4.0380
Iteration: 162; Percent complete: 4.0%; Average loss: 4.3805
Iteration: 163; Percent complete: 4.1%; Average loss: 4.1877
Iteration: 164; Percent complete: 4.1%; Average loss: 3.9613
Iteration: 165; Percent complete: 4.1%; Average loss: 4.1304
Iteration: 166; Percent complete: 4.2%; Average loss: 4.1484
Iteration: 167; Percent complete: 4.2%; Average loss: 4.1402
Iteration: 168; Percent complete: 4.2%; Average loss: 3.8270
Iteration: 169; Percent complete: 4.2%; Average loss: 3.9564
Iteration: 170; Percent complete: 4.2%; Average loss: 4.0151
Iteration: 171; Percent complete: 4.3%; Average loss: 4.0313
Iteration: 172; Percent complete: 4.3%; Average loss: 4.3585
Iteration: 173; Percent complete: 4.3%; Average loss: 4.0126
Iteration: 174; Percent complete: 4.3%; Average loss: 4.1853
Iteration: 175; Percent complete: 4.4%; Average loss: 4.2608
Iteration: 176; Percent complete: 4.4%; Average loss: 4.1220
Iteration: 177; Percent complete: 4.4%; Average loss: 3.9971
Iteration: 178; Percent complete: 4.5%; Average loss: 3.9325
Iteration: 179; Percent complete: 4.5%; Average loss: 4.3832
Iteration: 180; Percent complete: 4.5%; Average loss: 4.2323
Iteration: 181; Percent complete: 4.5%; Average loss: 4.1136
Iteration: 182; Percent complete: 4.5%; Average loss: 4.3793
Iteration: 183; Percent complete: 4.6%; Average loss: 4.3142
Iteration: 184; Percent complete: 4.6%; Average loss: 4.0898
Iteration: 185; Percent complete: 4.6%; Average loss: 3.8518
Iteration: 186; Percent complete: 4.7%; Average loss: 4.2654
Iteration: 187; Percent complete: 4.7%; Average loss: 4.2673
Iteration: 188; Percent complete: 4.7%; Average loss: 4.3268
Iteration: 189; Percent complete: 4.7%; Average loss: 4.2281
Iteration: 190; Percent complete: 4.8%; Average loss: 4.0706
Iteration: 191; Percent complete: 4.8%; Average loss: 3.9326
Iteration: 192; Percent complete: 4.8%; Average loss: 3.9325
Iteration: 193; Percent complete: 4.8%; Average loss: 4.0566
Iteration: 194; Percent complete: 4.9%; Average loss: 4.1284
Iteration: 195; Percent complete: 4.9%; Average loss: 4.2452
Iteration: 196; Percent complete: 4.9%; Average loss: 3.8395
Iteration: 197; Percent complete: 4.9%; Average loss: 4.1760
Iteration: 198; Percent complete: 5.0%; Average loss: 4.0335
Iteration: 199; Percent complete: 5.0%; Average loss: 4.0224
Iteration: 200; Percent complete: 5.0%; Average loss: 4.1291
Iteration: 201; Percent complete: 5.0%; Average loss: 3.8727
Iteration: 202; Percent complete: 5.1%; Average loss: 3.7859
Iteration: 203; Percent complete: 5.1%; Average loss: 3.9728
Iteration: 204; Percent complete: 5.1%; Average loss: 3.8737
Iteration: 205; Percent complete: 5.1%; Average loss: 4.1234
Iteration: 206; Percent complete: 5.1%; Average loss: 3.9053
Iteration: 207; Percent complete: 5.2%; Average loss: 4.1023
Iteration: 208; Percent complete: 5.2%; Average loss: 3.8172
Iteration: 209; Percent complete: 5.2%; Average loss: 3.6450
Iteration: 210; Percent complete: 5.2%; Average loss: 3.9507
Iteration: 211; Percent complete: 5.3%; Average loss: 4.0751
Iteration: 212; Percent complete: 5.3%; Average loss: 3.8407
Iteration: 213; Percent complete: 5.3%; Average loss: 4.2302
Iteration: 214; Percent complete: 5.3%; Average loss: 3.9174
Iteration: 215; Percent complete: 5.4%; Average loss: 4.2745
Iteration: 216; Percent complete: 5.4%; Average loss: 4.1539
Iteration: 217; Percent complete: 5.4%; Average loss: 3.8783
Iteration: 218; Percent complete: 5.5%; Average loss: 3.9238
Iteration: 219; Percent complete: 5.5%; Average loss: 3.8801
Iteration: 220; Percent complete: 5.5%; Average loss: 3.7361
Iteration: 221; Percent complete: 5.5%; Average loss: 4.1488
Iteration: 222; Percent complete: 5.5%; Average loss: 3.8536
Iteration: 223; Percent complete: 5.6%; Average loss: 3.9278
Iteration: 224; Percent complete: 5.6%; Average loss: 3.9966
Iteration: 225; Percent complete: 5.6%; Average loss: 3.8874
Iteration: 226; Percent complete: 5.7%; Average loss: 3.8185
Iteration: 227; Percent complete: 5.7%; Average loss: 4.3508
Iteration: 228; Percent complete: 5.7%; Average loss: 3.9767
Iteration: 229; Percent complete: 5.7%; Average loss: 4.2247
Iteration: 230; Percent complete: 5.8%; Average loss: 4.1535
Iteration: 231; Percent complete: 5.8%; Average loss: 4.0966
Iteration: 232; Percent complete: 5.8%; Average loss: 4.2128
Iteration: 233; Percent complete: 5.8%; Average loss: 4.1807
Iteration: 234; Percent complete: 5.9%; Average loss: 4.0511
Iteration: 235; Percent complete: 5.9%; Average loss: 3.7826
Iteration: 236; Percent complete: 5.9%; Average loss: 3.9780
Iteration: 237; Percent complete: 5.9%; Average loss: 4.0901
Iteration: 238; Percent complete: 5.9%; Average loss: 4.0702
Iteration: 239; Percent complete: 6.0%; Average loss: 4.0398
Iteration: 240; Percent complete: 6.0%; Average loss: 4.0696
Iteration: 241; Percent complete: 6.0%; Average loss: 3.9724
Iteration: 242; Percent complete: 6.0%; Average loss: 3.9524
Iteration: 243; Percent complete: 6.1%; Average loss: 3.9951
Iteration: 244; Percent complete: 6.1%; Average loss: 3.8764
Iteration: 245; Percent complete: 6.1%; Average loss: 4.0802
Iteration: 246; Percent complete: 6.2%; Average loss: 4.0172
Iteration: 247; Percent complete: 6.2%; Average loss: 3.9162
Iteration: 248; Percent complete: 6.2%; Average loss: 4.2023
Iteration: 249; Percent complete: 6.2%; Average loss: 3.9118
Iteration: 250; Percent complete: 6.2%; Average loss: 3.7076
Iteration: 251; Percent complete: 6.3%; Average loss: 4.0613
Iteration: 252; Percent complete: 6.3%; Average loss: 4.0756
Iteration: 253; Percent complete: 6.3%; Average loss: 4.1648
Iteration: 254; Percent complete: 6.3%; Average loss: 3.7464
Iteration: 255; Percent complete: 6.4%; Average loss: 3.9821
Iteration: 256; Percent complete: 6.4%; Average loss: 4.0684
Iteration: 257; Percent complete: 6.4%; Average loss: 3.8457
Iteration: 258; Percent complete: 6.5%; Average loss: 3.9183
Iteration: 259; Percent complete: 6.5%; Average loss: 3.7322
Iteration: 260; Percent complete: 6.5%; Average loss: 4.1314
Iteration: 261; Percent complete: 6.5%; Average loss: 4.0131
Iteration: 262; Percent complete: 6.6%; Average loss: 4.2059
Iteration: 263; Percent complete: 6.6%; Average loss: 3.9609
Iteration: 264; Percent complete: 6.6%; Average loss: 4.1061
Iteration: 265; Percent complete: 6.6%; Average loss: 4.0956
Iteration: 266; Percent complete: 6.7%; Average loss: 3.8368
Iteration: 267; Percent complete: 6.7%; Average loss: 4.0683
Iteration: 268; Percent complete: 6.7%; Average loss: 3.7500
Iteration: 269; Percent complete: 6.7%; Average loss: 3.9279
Iteration: 270; Percent complete: 6.8%; Average loss: 3.9782
Iteration: 271; Percent complete: 6.8%; Average loss: 4.0433
Iteration: 272; Percent complete: 6.8%; Average loss: 4.1085
Iteration: 273; Percent complete: 6.8%; Average loss: 3.9472
Iteration: 274; Percent complete: 6.9%; Average loss: 4.0844
Iteration: 275; Percent complete: 6.9%; Average loss: 3.7877
Iteration: 276; Percent complete: 6.9%; Average loss: 3.8374
Iteration: 277; Percent complete: 6.9%; Average loss: 3.8091
Iteration: 278; Percent complete: 7.0%; Average loss: 3.9510
Iteration: 279; Percent complete: 7.0%; Average loss: 3.8122
Iteration: 280; Percent complete: 7.0%; Average loss: 4.0648
Iteration: 281; Percent complete: 7.0%; Average loss: 3.9356
Iteration: 282; Percent complete: 7.0%; Average loss: 3.9104
Iteration: 283; Percent complete: 7.1%; Average loss: 4.0243
Iteration: 284; Percent complete: 7.1%; Average loss: 3.8137
Iteration: 285; Percent complete: 7.1%; Average loss: 3.8440
Iteration: 286; Percent complete: 7.1%; Average loss: 4.0849
Iteration: 287; Percent complete: 7.2%; Average loss: 3.7771
Iteration: 288; Percent complete: 7.2%; Average loss: 3.7856
Iteration: 289; Percent complete: 7.2%; Average loss: 3.7117
Iteration: 290; Percent complete: 7.2%; Average loss: 3.6706
Iteration: 291; Percent complete: 7.3%; Average loss: 3.9394
Iteration: 292; Percent complete: 7.3%; Average loss: 3.9786
Iteration: 293; Percent complete: 7.3%; Average loss: 3.9295
Iteration: 294; Percent complete: 7.3%; Average loss: 3.9633
Iteration: 295; Percent complete: 7.4%; Average loss: 4.0543
Iteration: 296; Percent complete: 7.4%; Average loss: 4.1497
Iteration: 297; Percent complete: 7.4%; Average loss: 3.7968
Iteration: 298; Percent complete: 7.4%; Average loss: 3.9048
Iteration: 299; Percent complete: 7.5%; Average loss: 3.9103
Iteration: 300; Percent complete: 7.5%; Average loss: 3.8296
Iteration: 301; Percent complete: 7.5%; Average loss: 3.9528
Iteration: 302; Percent complete: 7.5%; Average loss: 3.8796
Iteration: 303; Percent complete: 7.6%; Average loss: 3.7341
Iteration: 304; Percent complete: 7.6%; Average loss: 3.9424
Iteration: 305; Percent complete: 7.6%; Average loss: 4.0921
Iteration: 306; Percent complete: 7.6%; Average loss: 3.5980
Iteration: 307; Percent complete: 7.7%; Average loss: 3.9811
Iteration: 308; Percent complete: 7.7%; Average loss: 3.7497
Iteration: 309; Percent complete: 7.7%; Average loss: 3.7984
Iteration: 310; Percent complete: 7.8%; Average loss: 3.8040
Iteration: 311; Percent complete: 7.8%; Average loss: 3.6634
Iteration: 312; Percent complete: 7.8%; Average loss: 4.1464
Iteration: 313; Percent complete: 7.8%; Average loss: 3.6037
Iteration: 314; Percent complete: 7.8%; Average loss: 3.8201
Iteration: 315; Percent complete: 7.9%; Average loss: 3.7938
Iteration: 316; Percent complete: 7.9%; Average loss: 4.0032
Iteration: 317; Percent complete: 7.9%; Average loss: 3.6759
Iteration: 318; Percent complete: 8.0%; Average loss: 4.0741
Iteration: 319; Percent complete: 8.0%; Average loss: 3.9310
Iteration: 320; Percent complete: 8.0%; Average loss: 3.8196
Iteration: 321; Percent complete: 8.0%; Average loss: 3.7818
Iteration: 322; Percent complete: 8.1%; Average loss: 3.8968
Iteration: 323; Percent complete: 8.1%; Average loss: 4.0029
Iteration: 324; Percent complete: 8.1%; Average loss: 3.8422
Iteration: 325; Percent complete: 8.1%; Average loss: 4.1418
Iteration: 326; Percent complete: 8.2%; Average loss: 3.9272
Iteration: 327; Percent complete: 8.2%; Average loss: 3.7134
Iteration: 328; Percent complete: 8.2%; Average loss: 3.8503
Iteration: 329; Percent complete: 8.2%; Average loss: 3.7786
Iteration: 330; Percent complete: 8.2%; Average loss: 3.9736
Iteration: 331; Percent complete: 8.3%; Average loss: 3.7425
Iteration: 332; Percent complete: 8.3%; Average loss: 3.8481
Iteration: 333; Percent complete: 8.3%; Average loss: 3.7914
Iteration: 334; Percent complete: 8.3%; Average loss: 3.8895
Iteration: 335; Percent complete: 8.4%; Average loss: 3.8418
Iteration: 336; Percent complete: 8.4%; Average loss: 3.7709
Iteration: 337; Percent complete: 8.4%; Average loss: 3.9440
Iteration: 338; Percent complete: 8.5%; Average loss: 3.8181
Iteration: 339; Percent complete: 8.5%; Average loss: 4.1326
Iteration: 340; Percent complete: 8.5%; Average loss: 3.6864
Iteration: 341; Percent complete: 8.5%; Average loss: 3.8144
Iteration: 342; Percent complete: 8.6%; Average loss: 3.7464
Iteration: 343; Percent complete: 8.6%; Average loss: 3.5257
Iteration: 344; Percent complete: 8.6%; Average loss: 3.8670
Iteration: 345; Percent complete: 8.6%; Average loss: 4.0129
Iteration: 346; Percent complete: 8.6%; Average loss: 4.2958
Iteration: 347; Percent complete: 8.7%; Average loss: 4.0808
Iteration: 348; Percent complete: 8.7%; Average loss: 4.0744
Iteration: 349; Percent complete: 8.7%; Average loss: 3.7564
Iteration: 350; Percent complete: 8.8%; Average loss: 3.7386
Iteration: 351; Percent complete: 8.8%; Average loss: 3.8060
Iteration: 352; Percent complete: 8.8%; Average loss: 3.7938
Iteration: 353; Percent complete: 8.8%; Average loss: 3.9657
Iteration: 354; Percent complete: 8.8%; Average loss: 3.9030
Iteration: 355; Percent complete: 8.9%; Average loss: 3.9616
Iteration: 356; Percent complete: 8.9%; Average loss: 4.0559
Iteration: 357; Percent complete: 8.9%; Average loss: 3.9211
Iteration: 358; Percent complete: 8.9%; Average loss: 3.6440
Iteration: 359; Percent complete: 9.0%; Average loss: 4.2290
Iteration: 360; Percent complete: 9.0%; Average loss: 3.9502
Iteration: 361; Percent complete: 9.0%; Average loss: 3.4291
Iteration: 362; Percent complete: 9.0%; Average loss: 3.9088
Iteration: 363; Percent complete: 9.1%; Average loss: 3.6903
Iteration: 364; Percent complete: 9.1%; Average loss: 3.8658
Iteration: 365; Percent complete: 9.1%; Average loss: 3.7694
Iteration: 366; Percent complete: 9.2%; Average loss: 3.9120
Iteration: 367; Percent complete: 9.2%; Average loss: 3.6114
Iteration: 368; Percent complete: 9.2%; Average loss: 4.4909
Iteration: 369; Percent complete: 9.2%; Average loss: 3.5321
Iteration: 370; Percent complete: 9.2%; Average loss: 3.8108
Iteration: 371; Percent complete: 9.3%; Average loss: 3.8931
Iteration: 372; Percent complete: 9.3%; Average loss: 3.6310
Iteration: 373; Percent complete: 9.3%; Average loss: 3.8302
Iteration: 374; Percent complete: 9.3%; Average loss: 3.9251
Iteration: 375; Percent complete: 9.4%; Average loss: 3.9476
Iteration: 376; Percent complete: 9.4%; Average loss: 3.6113
Iteration: 377; Percent complete: 9.4%; Average loss: 3.8990
Iteration: 378; Percent complete: 9.4%; Average loss: 3.9029
Iteration: 379; Percent complete: 9.5%; Average loss: 3.6594
Iteration: 380; Percent complete: 9.5%; Average loss: 3.9886
Iteration: 381; Percent complete: 9.5%; Average loss: 3.7335
Iteration: 382; Percent complete: 9.6%; Average loss: 3.7343
Iteration: 383; Percent complete: 9.6%; Average loss: 4.0274
Iteration: 384; Percent complete: 9.6%; Average loss: 3.6270
Iteration: 385; Percent complete: 9.6%; Average loss: 3.9830
Iteration: 386; Percent complete: 9.7%; Average loss: 3.8342
Iteration: 387; Percent complete: 9.7%; Average loss: 3.7995
Iteration: 388; Percent complete: 9.7%; Average loss: 3.6124
Iteration: 389; Percent complete: 9.7%; Average loss: 3.6068
Iteration: 390; Percent complete: 9.8%; Average loss: 3.9279
Iteration: 391; Percent complete: 9.8%; Average loss: 3.8755
Iteration: 392; Percent complete: 9.8%; Average loss: 3.8161
Iteration: 393; Percent complete: 9.8%; Average loss: 3.7221
Iteration: 394; Percent complete: 9.8%; Average loss: 3.8844
Iteration: 395; Percent complete: 9.9%; Average loss: 3.7936
Iteration: 396; Percent complete: 9.9%; Average loss: 3.8969
Iteration: 397; Percent complete: 9.9%; Average loss: 3.8470
Iteration: 398; Percent complete: 10.0%; Average loss: 3.6635
Iteration: 399; Percent complete: 10.0%; Average loss: 3.6975
Iteration: 400; Percent complete: 10.0%; Average loss: 3.9396
Iteration: 401; Percent complete: 10.0%; Average loss: 3.8376
Iteration: 402; Percent complete: 10.1%; Average loss: 3.9514
Iteration: 403; Percent complete: 10.1%; Average loss: 3.7499
Iteration: 404; Percent complete: 10.1%; Average loss: 3.6822
Iteration: 405; Percent complete: 10.1%; Average loss: 3.8306
Iteration: 406; Percent complete: 10.2%; Average loss: 3.9633
Iteration: 407; Percent complete: 10.2%; Average loss: 4.0970
Iteration: 408; Percent complete: 10.2%; Average loss: 3.8189
Iteration: 409; Percent complete: 10.2%; Average loss: 3.9304
Iteration: 410; Percent complete: 10.2%; Average loss: 3.7470
Iteration: 411; Percent complete: 10.3%; Average loss: 3.7143
Iteration: 412; Percent complete: 10.3%; Average loss: 3.8816
Iteration: 413; Percent complete: 10.3%; Average loss: 3.7113
Iteration: 414; Percent complete: 10.3%; Average loss: 3.8674
Iteration: 415; Percent complete: 10.4%; Average loss: 3.7587
Iteration: 416; Percent complete: 10.4%; Average loss: 3.7021
Iteration: 417; Percent complete: 10.4%; Average loss: 3.7962
Iteration: 418; Percent complete: 10.4%; Average loss: 3.7525
Iteration: 419; Percent complete: 10.5%; Average loss: 3.9385
Iteration: 420; Percent complete: 10.5%; Average loss: 3.6481
Iteration: 421; Percent complete: 10.5%; Average loss: 3.8123
Iteration: 422; Percent complete: 10.5%; Average loss: 3.7909
Iteration: 423; Percent complete: 10.6%; Average loss: 4.0138
Iteration: 424; Percent complete: 10.6%; Average loss: 3.8478
Iteration: 425; Percent complete: 10.6%; Average loss: 3.9835
Iteration: 426; Percent complete: 10.7%; Average loss: 3.6563
Iteration: 427; Percent complete: 10.7%; Average loss: 3.7077
Iteration: 428; Percent complete: 10.7%; Average loss: 3.6959
Iteration: 429; Percent complete: 10.7%; Average loss: 3.7230
Iteration: 430; Percent complete: 10.8%; Average loss: 3.7955
Iteration: 431; Percent complete: 10.8%; Average loss: 3.7149
Iteration: 432; Percent complete: 10.8%; Average loss: 3.6924
Iteration: 433; Percent complete: 10.8%; Average loss: 3.8810
Iteration: 434; Percent complete: 10.8%; Average loss: 3.7947
Iteration: 435; Percent complete: 10.9%; Average loss: 3.7240
Iteration: 436; Percent complete: 10.9%; Average loss: 3.8814
Iteration: 437; Percent complete: 10.9%; Average loss: 3.6983
Iteration: 438; Percent complete: 10.9%; Average loss: 3.7553
Iteration: 439; Percent complete: 11.0%; Average loss: 3.7329
Iteration: 440; Percent complete: 11.0%; Average loss: 3.8750
Iteration: 441; Percent complete: 11.0%; Average loss: 3.7257
Iteration: 442; Percent complete: 11.1%; Average loss: 3.7820
Iteration: 443; Percent complete: 11.1%; Average loss: 3.6424
Iteration: 444; Percent complete: 11.1%; Average loss: 3.7006
Iteration: 445; Percent complete: 11.1%; Average loss: 4.0319
Iteration: 446; Percent complete: 11.2%; Average loss: 3.9674
Iteration: 447; Percent complete: 11.2%; Average loss: 3.7085
Iteration: 448; Percent complete: 11.2%; Average loss: 3.5948
Iteration: 449; Percent complete: 11.2%; Average loss: 3.6223
Iteration: 450; Percent complete: 11.2%; Average loss: 3.8635
Iteration: 451; Percent complete: 11.3%; Average loss: 3.7530
Iteration: 452; Percent complete: 11.3%; Average loss: 3.7367
Iteration: 453; Percent complete: 11.3%; Average loss: 3.8167
Iteration: 454; Percent complete: 11.3%; Average loss: 3.8842
Iteration: 455; Percent complete: 11.4%; Average loss: 3.4287
Iteration: 456; Percent complete: 11.4%; Average loss: 3.6320
Iteration: 457; Percent complete: 11.4%; Average loss: 3.7101
Iteration: 458; Percent complete: 11.5%; Average loss: 3.6251
Iteration: 459; Percent complete: 11.5%; Average loss: 3.8133
Iteration: 460; Percent complete: 11.5%; Average loss: 3.7239
Iteration: 461; Percent complete: 11.5%; Average loss: 3.8015
Iteration: 462; Percent complete: 11.6%; Average loss: 3.6204
Iteration: 463; Percent complete: 11.6%; Average loss: 3.6670
Iteration: 464; Percent complete: 11.6%; Average loss: 3.6587
Iteration: 465; Percent complete: 11.6%; Average loss: 3.7973
Iteration: 466; Percent complete: 11.7%; Average loss: 3.7386
Iteration: 467; Percent complete: 11.7%; Average loss: 3.7102
Iteration: 468; Percent complete: 11.7%; Average loss: 3.8255
Iteration: 469; Percent complete: 11.7%; Average loss: 3.7643
Iteration: 470; Percent complete: 11.8%; Average loss: 3.4146
Iteration: 471; Percent complete: 11.8%; Average loss: 3.7278
Iteration: 472; Percent complete: 11.8%; Average loss: 3.8477
Iteration: 473; Percent complete: 11.8%; Average loss: 3.4801
Iteration: 474; Percent complete: 11.8%; Average loss: 3.9837
Iteration: 475; Percent complete: 11.9%; Average loss: 3.8222
Iteration: 476; Percent complete: 11.9%; Average loss: 3.8676
Iteration: 477; Percent complete: 11.9%; Average loss: 3.7390
Iteration: 478; Percent complete: 11.9%; Average loss: 3.6457
Iteration: 479; Percent complete: 12.0%; Average loss: 3.8751
Iteration: 480; Percent complete: 12.0%; Average loss: 3.8665
Iteration: 481; Percent complete: 12.0%; Average loss: 3.5241
Iteration: 482; Percent complete: 12.0%; Average loss: 3.6860
Iteration: 483; Percent complete: 12.1%; Average loss: 3.6806
Iteration: 484; Percent complete: 12.1%; Average loss: 3.8950
Iteration: 485; Percent complete: 12.1%; Average loss: 4.0179
Iteration: 486; Percent complete: 12.2%; Average loss: 3.5592
Iteration: 487; Percent complete: 12.2%; Average loss: 3.8728
Iteration: 488; Percent complete: 12.2%; Average loss: 3.9579
Iteration: 489; Percent complete: 12.2%; Average loss: 3.5731
Iteration: 490; Percent complete: 12.2%; Average loss: 3.4472
Iteration: 491; Percent complete: 12.3%; Average loss: 3.7771
Iteration: 492; Percent complete: 12.3%; Average loss: 3.7851
Iteration: 493; Percent complete: 12.3%; Average loss: 3.7553
Iteration: 494; Percent complete: 12.3%; Average loss: 3.7610
Iteration: 495; Percent complete: 12.4%; Average loss: 3.7594
Iteration: 496; Percent complete: 12.4%; Average loss: 3.6817
Iteration: 497; Percent complete: 12.4%; Average loss: 3.8286
Iteration: 498; Percent complete: 12.4%; Average loss: 3.6068
Iteration: 499; Percent complete: 12.5%; Average loss: 3.7477
Iteration: 500; Percent complete: 12.5%; Average loss: 3.5540
Iteration: 501; Percent complete: 12.5%; Average loss: 3.5424
Iteration: 502; Percent complete: 12.6%; Average loss: 3.5742
Iteration: 503; Percent complete: 12.6%; Average loss: 3.6445
Iteration: 504; Percent complete: 12.6%; Average loss: 4.0306
Iteration: 505; Percent complete: 12.6%; Average loss: 3.6626
Iteration: 506; Percent complete: 12.7%; Average loss: 3.2654
Iteration: 507; Percent complete: 12.7%; Average loss: 3.9002
Iteration: 508; Percent complete: 12.7%; Average loss: 3.8053
Iteration: 509; Percent complete: 12.7%; Average loss: 3.9239
Iteration: 510; Percent complete: 12.8%; Average loss: 3.7956
Iteration: 511; Percent complete: 12.8%; Average loss: 3.5908
Iteration: 512; Percent complete: 12.8%; Average loss: 3.8485
Iteration: 513; Percent complete: 12.8%; Average loss: 3.8168
Iteration: 514; Percent complete: 12.8%; Average loss: 3.6432
Iteration: 515; Percent complete: 12.9%; Average loss: 3.7146
Iteration: 516; Percent complete: 12.9%; Average loss: 3.8827
Iteration: 517; Percent complete: 12.9%; Average loss: 3.7802
Iteration: 518; Percent complete: 13.0%; Average loss: 3.5551
Iteration: 519; Percent complete: 13.0%; Average loss: 3.5603
Iteration: 520; Percent complete: 13.0%; Average loss: 3.6605
Iteration: 521; Percent complete: 13.0%; Average loss: 3.6889
Iteration: 522; Percent complete: 13.1%; Average loss: 3.7760
Iteration: 523; Percent complete: 13.1%; Average loss: 3.7810
Iteration: 524; Percent complete: 13.1%; Average loss: 3.8142
Iteration: 525; Percent complete: 13.1%; Average loss: 3.5349
Iteration: 526; Percent complete: 13.2%; Average loss: 3.4880
Iteration: 527; Percent complete: 13.2%; Average loss: 3.7787
Iteration: 528; Percent complete: 13.2%; Average loss: 3.8719
Iteration: 529; Percent complete: 13.2%; Average loss: 3.5514
Iteration: 530; Percent complete: 13.2%; Average loss: 3.6839
Iteration: 531; Percent complete: 13.3%; Average loss: 3.7127
Iteration: 532; Percent complete: 13.3%; Average loss: 3.5428
Iteration: 533; Percent complete: 13.3%; Average loss: 3.5508
Iteration: 534; Percent complete: 13.4%; Average loss: 3.5633
Iteration: 535; Percent complete: 13.4%; Average loss: 3.7618
Iteration: 536; Percent complete: 13.4%; Average loss: 3.6619
Iteration: 537; Percent complete: 13.4%; Average loss: 3.6458
Iteration: 538; Percent complete: 13.5%; Average loss: 3.6217
Iteration: 539; Percent complete: 13.5%; Average loss: 3.6412
Iteration: 540; Percent complete: 13.5%; Average loss: 3.7571
Iteration: 541; Percent complete: 13.5%; Average loss: 3.7571
Iteration: 542; Percent complete: 13.6%; Average loss: 3.7658
Iteration: 543; Percent complete: 13.6%; Average loss: 3.7736
Iteration: 544; Percent complete: 13.6%; Average loss: 3.6070
Iteration: 545; Percent complete: 13.6%; Average loss: 3.5834
Iteration: 546; Percent complete: 13.7%; Average loss: 3.7554
Iteration: 547; Percent complete: 13.7%; Average loss: 3.7785
Iteration: 548; Percent complete: 13.7%; Average loss: 3.5477
Iteration: 549; Percent complete: 13.7%; Average loss: 3.7802
Iteration: 550; Percent complete: 13.8%; Average loss: 3.3777
Iteration: 551; Percent complete: 13.8%; Average loss: 3.7040
Iteration: 552; Percent complete: 13.8%; Average loss: 3.7057
Iteration: 553; Percent complete: 13.8%; Average loss: 3.5872
Iteration: 554; Percent complete: 13.9%; Average loss: 3.5950
Iteration: 555; Percent complete: 13.9%; Average loss: 3.7896
Iteration: 556; Percent complete: 13.9%; Average loss: 3.8025
Iteration: 557; Percent complete: 13.9%; Average loss: 3.5710
Iteration: 558; Percent complete: 14.0%; Average loss: 3.5259
Iteration: 559; Percent complete: 14.0%; Average loss: 3.6828
Iteration: 560; Percent complete: 14.0%; Average loss: 3.6629
Iteration: 561; Percent complete: 14.0%; Average loss: 3.7487
Iteration: 562; Percent complete: 14.1%; Average loss: 3.6581
Iteration: 563; Percent complete: 14.1%; Average loss: 3.9435
Iteration: 564; Percent complete: 14.1%; Average loss: 3.7479
Iteration: 565; Percent complete: 14.1%; Average loss: 3.6884
Iteration: 566; Percent complete: 14.1%; Average loss: 3.6916
Iteration: 567; Percent complete: 14.2%; Average loss: 3.6966
Iteration: 568; Percent complete: 14.2%; Average loss: 3.4647
Iteration: 569; Percent complete: 14.2%; Average loss: 3.5477
Iteration: 570; Percent complete: 14.2%; Average loss: 3.8069
Iteration: 571; Percent complete: 14.3%; Average loss: 3.4890
Iteration: 572; Percent complete: 14.3%; Average loss: 3.4913
Iteration: 573; Percent complete: 14.3%; Average loss: 3.7006
Iteration: 574; Percent complete: 14.3%; Average loss: 3.5576
Iteration: 575; Percent complete: 14.4%; Average loss: 3.5206
Iteration: 576; Percent complete: 14.4%; Average loss: 3.6587
Iteration: 577; Percent complete: 14.4%; Average loss: 3.6869
Iteration: 578; Percent complete: 14.4%; Average loss: 3.5169
Iteration: 579; Percent complete: 14.5%; Average loss: 3.8189
Iteration: 580; Percent complete: 14.5%; Average loss: 3.5105
Iteration: 581; Percent complete: 14.5%; Average loss: 3.7479
Iteration: 582; Percent complete: 14.5%; Average loss: 3.4886
Iteration: 583; Percent complete: 14.6%; Average loss: 3.6345
Iteration: 584; Percent complete: 14.6%; Average loss: 4.1084
Iteration: 585; Percent complete: 14.6%; Average loss: 3.5011
Iteration: 586; Percent complete: 14.6%; Average loss: 3.7676
Iteration: 587; Percent complete: 14.7%; Average loss: 3.4538
Iteration: 588; Percent complete: 14.7%; Average loss: 3.5498
Iteration: 589; Percent complete: 14.7%; Average loss: 3.5415
Iteration: 590; Percent complete: 14.8%; Average loss: 3.7191
Iteration: 591; Percent complete: 14.8%; Average loss: 3.7270
Iteration: 592; Percent complete: 14.8%; Average loss: 3.8801
Iteration: 593; Percent complete: 14.8%; Average loss: 3.5397
Iteration: 594; Percent complete: 14.8%; Average loss: 3.9349
Iteration: 595; Percent complete: 14.9%; Average loss: 3.4132
Iteration: 596; Percent complete: 14.9%; Average loss: 3.4447
Iteration: 597; Percent complete: 14.9%; Average loss: 3.7060
Iteration: 598; Percent complete: 14.9%; Average loss: 3.6069
Iteration: 599; Percent complete: 15.0%; Average loss: 3.5731
Iteration: 600; Percent complete: 15.0%; Average loss: 3.5846
Iteration: 601; Percent complete: 15.0%; Average loss: 3.6809
Iteration: 602; Percent complete: 15.0%; Average loss: 3.7390
Iteration: 603; Percent complete: 15.1%; Average loss: 3.4813
Iteration: 604; Percent complete: 15.1%; Average loss: 3.7147
Iteration: 605; Percent complete: 15.1%; Average loss: 3.4229
Iteration: 606; Percent complete: 15.2%; Average loss: 3.6415
Iteration: 607; Percent complete: 15.2%; Average loss: 3.5703
Iteration: 608; Percent complete: 15.2%; Average loss: 3.8003
Iteration: 609; Percent complete: 15.2%; Average loss: 3.6403
Iteration: 610; Percent complete: 15.2%; Average loss: 3.7788
Iteration: 611; Percent complete: 15.3%; Average loss: 3.5974
Iteration: 612; Percent complete: 15.3%; Average loss: 3.8677
Iteration: 613; Percent complete: 15.3%; Average loss: 3.8594
Iteration: 614; Percent complete: 15.3%; Average loss: 3.6327
Iteration: 615; Percent complete: 15.4%; Average loss: 3.3560
Iteration: 616; Percent complete: 15.4%; Average loss: 3.6640
Iteration: 617; Percent complete: 15.4%; Average loss: 3.4037
Iteration: 618; Percent complete: 15.4%; Average loss: 3.7036
Iteration: 619; Percent complete: 15.5%; Average loss: 3.6262
Iteration: 620; Percent complete: 15.5%; Average loss: 3.6174
Iteration: 621; Percent complete: 15.5%; Average loss: 3.6742
Iteration: 622; Percent complete: 15.6%; Average loss: 3.5470
Iteration: 623; Percent complete: 15.6%; Average loss: 3.5525
Iteration: 624; Percent complete: 15.6%; Average loss: 3.9433
Iteration: 625; Percent complete: 15.6%; Average loss: 3.6759
Iteration: 626; Percent complete: 15.7%; Average loss: 3.6171
Iteration: 627; Percent complete: 15.7%; Average loss: 3.5494
Iteration: 628; Percent complete: 15.7%; Average loss: 3.6044
Iteration: 629; Percent complete: 15.7%; Average loss: 3.5592
Iteration: 630; Percent complete: 15.8%; Average loss: 3.6633
Iteration: 631; Percent complete: 15.8%; Average loss: 3.5296
Iteration: 632; Percent complete: 15.8%; Average loss: 3.6589
Iteration: 633; Percent complete: 15.8%; Average loss: 3.6500
Iteration: 634; Percent complete: 15.8%; Average loss: 3.7583
Iteration: 635; Percent complete: 15.9%; Average loss: 3.6974
Iteration: 636; Percent complete: 15.9%; Average loss: 3.7723
Iteration: 637; Percent complete: 15.9%; Average loss: 3.8806
Iteration: 638; Percent complete: 16.0%; Average loss: 3.7303
Iteration: 639; Percent complete: 16.0%; Average loss: 3.7322
Iteration: 640; Percent complete: 16.0%; Average loss: 3.4386
Iteration: 641; Percent complete: 16.0%; Average loss: 3.4954
Iteration: 642; Percent complete: 16.1%; Average loss: 3.7314
Iteration: 643; Percent complete: 16.1%; Average loss: 3.8407
Iteration: 644; Percent complete: 16.1%; Average loss: 3.4257
Iteration: 645; Percent complete: 16.1%; Average loss: 3.3414
Iteration: 646; Percent complete: 16.2%; Average loss: 3.7456
Iteration: 647; Percent complete: 16.2%; Average loss: 3.5244
Iteration: 648; Percent complete: 16.2%; Average loss: 3.4442
Iteration: 649; Percent complete: 16.2%; Average loss: 3.6919
Iteration: 650; Percent complete: 16.2%; Average loss: 3.5138
Iteration: 651; Percent complete: 16.3%; Average loss: 3.5279
Iteration: 652; Percent complete: 16.3%; Average loss: 3.3729
Iteration: 653; Percent complete: 16.3%; Average loss: 3.7924
Iteration: 654; Percent complete: 16.4%; Average loss: 3.7990
Iteration: 655; Percent complete: 16.4%; Average loss: 3.5642
Iteration: 656; Percent complete: 16.4%; Average loss: 3.4457
Iteration: 657; Percent complete: 16.4%; Average loss: 3.6368
Iteration: 658; Percent complete: 16.4%; Average loss: 3.7331
Iteration: 659; Percent complete: 16.5%; Average loss: 3.8896
Iteration: 660; Percent complete: 16.5%; Average loss: 3.6079
Iteration: 661; Percent complete: 16.5%; Average loss: 3.7183
Iteration: 662; Percent complete: 16.6%; Average loss: 3.4988
Iteration: 663; Percent complete: 16.6%; Average loss: 3.5234
Iteration: 664; Percent complete: 16.6%; Average loss: 3.3737
Iteration: 665; Percent complete: 16.6%; Average loss: 3.5568
Iteration: 666; Percent complete: 16.7%; Average loss: 3.6410
Iteration: 667; Percent complete: 16.7%; Average loss: 3.6486
Iteration: 668; Percent complete: 16.7%; Average loss: 3.4472
Iteration: 669; Percent complete: 16.7%; Average loss: 3.5203
Iteration: 670; Percent complete: 16.8%; Average loss: 3.2360
Iteration: 671; Percent complete: 16.8%; Average loss: 3.5432
Iteration: 672; Percent complete: 16.8%; Average loss: 3.6267
Iteration: 673; Percent complete: 16.8%; Average loss: 3.7941
Iteration: 674; Percent complete: 16.9%; Average loss: 3.5570
Iteration: 675; Percent complete: 16.9%; Average loss: 3.2421
Iteration: 676; Percent complete: 16.9%; Average loss: 3.8007
Iteration: 677; Percent complete: 16.9%; Average loss: 3.6752
Iteration: 678; Percent complete: 17.0%; Average loss: 3.6479
Iteration: 679; Percent complete: 17.0%; Average loss: 3.4002
Iteration: 680; Percent complete: 17.0%; Average loss: 3.6918
Iteration: 681; Percent complete: 17.0%; Average loss: 3.6596
Iteration: 682; Percent complete: 17.1%; Average loss: 3.4825
Iteration: 683; Percent complete: 17.1%; Average loss: 3.3791
Iteration: 684; Percent complete: 17.1%; Average loss: 3.4124
Iteration: 685; Percent complete: 17.1%; Average loss: 3.5855
Iteration: 686; Percent complete: 17.2%; Average loss: 3.4367
Iteration: 687; Percent complete: 17.2%; Average loss: 3.4254
Iteration: 688; Percent complete: 17.2%; Average loss: 3.4660
Iteration: 689; Percent complete: 17.2%; Average loss: 3.6211
Iteration: 690; Percent complete: 17.2%; Average loss: 3.6183
Iteration: 691; Percent complete: 17.3%; Average loss: 3.5580
Iteration: 692; Percent complete: 17.3%; Average loss: 3.7624
Iteration: 693; Percent complete: 17.3%; Average loss: 3.7287
Iteration: 694; Percent complete: 17.3%; Average loss: 3.8290
Iteration: 695; Percent complete: 17.4%; Average loss: 3.6595
Iteration: 696; Percent complete: 17.4%; Average loss: 3.5615
Iteration: 697; Percent complete: 17.4%; Average loss: 3.5857
Iteration: 698; Percent complete: 17.4%; Average loss: 3.4030
Iteration: 699; Percent complete: 17.5%; Average loss: 3.6854
Iteration: 700; Percent complete: 17.5%; Average loss: 3.4153
Iteration: 701; Percent complete: 17.5%; Average loss: 3.5569
Iteration: 702; Percent complete: 17.5%; Average loss: 3.4419
Iteration: 703; Percent complete: 17.6%; Average loss: 3.6584
Iteration: 704; Percent complete: 17.6%; Average loss: 3.8675
Iteration: 705; Percent complete: 17.6%; Average loss: 3.6229
Iteration: 706; Percent complete: 17.6%; Average loss: 3.4705
Iteration: 707; Percent complete: 17.7%; Average loss: 3.6767
Iteration: 708; Percent complete: 17.7%; Average loss: 3.6562
Iteration: 709; Percent complete: 17.7%; Average loss: 3.6547
Iteration: 710; Percent complete: 17.8%; Average loss: 3.5242
Iteration: 711; Percent complete: 17.8%; Average loss: 3.5706
Iteration: 712; Percent complete: 17.8%; Average loss: 3.6483
Iteration: 713; Percent complete: 17.8%; Average loss: 3.6797
Iteration: 714; Percent complete: 17.8%; Average loss: 3.7197
Iteration: 715; Percent complete: 17.9%; Average loss: 3.6155
Iteration: 716; Percent complete: 17.9%; Average loss: 3.6701
Iteration: 717; Percent complete: 17.9%; Average loss: 3.4558
Iteration: 718; Percent complete: 17.9%; Average loss: 3.8485
Iteration: 719; Percent complete: 18.0%; Average loss: 3.6603
Iteration: 720; Percent complete: 18.0%; Average loss: 3.6098
Iteration: 721; Percent complete: 18.0%; Average loss: 3.7106
Iteration: 722; Percent complete: 18.1%; Average loss: 3.4725
Iteration: 723; Percent complete: 18.1%; Average loss: 3.3031
Iteration: 724; Percent complete: 18.1%; Average loss: 3.4059
Iteration: 725; Percent complete: 18.1%; Average loss: 3.5610
Iteration: 726; Percent complete: 18.1%; Average loss: 3.5576
Iteration: 727; Percent complete: 18.2%; Average loss: 3.5770
Iteration: 728; Percent complete: 18.2%; Average loss: 3.5770
Iteration: 729; Percent complete: 18.2%; Average loss: 3.8778
Iteration: 730; Percent complete: 18.2%; Average loss: 3.5845
Iteration: 731; Percent complete: 18.3%; Average loss: 3.5845
Iteration: 732; Percent complete: 18.3%; Average loss: 3.6980
Iteration: 733; Percent complete: 18.3%; Average loss: 3.6520
Iteration: 734; Percent complete: 18.4%; Average loss: 3.5085
Iteration: 735; Percent complete: 18.4%; Average loss: 3.5250
Iteration: 736; Percent complete: 18.4%; Average loss: 3.4500
Iteration: 737; Percent complete: 18.4%; Average loss: 3.4592
Iteration: 738; Percent complete: 18.4%; Average loss: 3.4797
Iteration: 739; Percent complete: 18.5%; Average loss: 3.4643
Iteration: 740; Percent complete: 18.5%; Average loss: 3.3503
Iteration: 741; Percent complete: 18.5%; Average loss: 3.6812
Iteration: 742; Percent complete: 18.6%; Average loss: 3.3338
Iteration: 743; Percent complete: 18.6%; Average loss: 3.4910
Iteration: 744; Percent complete: 18.6%; Average loss: 3.7353
Iteration: 745; Percent complete: 18.6%; Average loss: 3.8124
Iteration: 746; Percent complete: 18.6%; Average loss: 3.5286
Iteration: 747; Percent complete: 18.7%; Average loss: 3.5590
Iteration: 748; Percent complete: 18.7%; Average loss: 3.6866
Iteration: 749; Percent complete: 18.7%; Average loss: 3.8010
Iteration: 750; Percent complete: 18.8%; Average loss: 3.5986
Iteration: 751; Percent complete: 18.8%; Average loss: 3.5778
Iteration: 752; Percent complete: 18.8%; Average loss: 3.5727
Iteration: 753; Percent complete: 18.8%; Average loss: 3.6395
Iteration: 754; Percent complete: 18.9%; Average loss: 3.5135
Iteration: 755; Percent complete: 18.9%; Average loss: 3.2496
Iteration: 756; Percent complete: 18.9%; Average loss: 4.0303
Iteration: 757; Percent complete: 18.9%; Average loss: 3.3266
Iteration: 758; Percent complete: 18.9%; Average loss: 3.6960
Iteration: 759; Percent complete: 19.0%; Average loss: 3.5563
Iteration: 760; Percent complete: 19.0%; Average loss: 3.4206
Iteration: 761; Percent complete: 19.0%; Average loss: 3.9044
Iteration: 762; Percent complete: 19.1%; Average loss: 3.4226
Iteration: 763; Percent complete: 19.1%; Average loss: 3.7131
Iteration: 764; Percent complete: 19.1%; Average loss: 3.3724
Iteration: 765; Percent complete: 19.1%; Average loss: 3.6247
Iteration: 766; Percent complete: 19.1%; Average loss: 3.4033
Iteration: 767; Percent complete: 19.2%; Average loss: 3.5745
Iteration: 768; Percent complete: 19.2%; Average loss: 3.3783
Iteration: 769; Percent complete: 19.2%; Average loss: 3.7205
Iteration: 770; Percent complete: 19.2%; Average loss: 3.4740
Iteration: 771; Percent complete: 19.3%; Average loss: 3.4888
Iteration: 772; Percent complete: 19.3%; Average loss: 3.5706
Iteration: 773; Percent complete: 19.3%; Average loss: 3.4109
Iteration: 774; Percent complete: 19.4%; Average loss: 3.5372
Iteration: 775; Percent complete: 19.4%; Average loss: 3.5970
Iteration: 776; Percent complete: 19.4%; Average loss: 3.4313
Iteration: 777; Percent complete: 19.4%; Average loss: 3.3376
Iteration: 778; Percent complete: 19.4%; Average loss: 3.4661
Iteration: 779; Percent complete: 19.5%; Average loss: 3.5705
Iteration: 780; Percent complete: 19.5%; Average loss: 3.5826
Iteration: 781; Percent complete: 19.5%; Average loss: 3.3674
Iteration: 782; Percent complete: 19.6%; Average loss: 3.3919
Iteration: 783; Percent complete: 19.6%; Average loss: 3.5322
Iteration: 784; Percent complete: 19.6%; Average loss: 3.8047
Iteration: 785; Percent complete: 19.6%; Average loss: 3.7231
Iteration: 786; Percent complete: 19.7%; Average loss: 3.5626
Iteration: 787; Percent complete: 19.7%; Average loss: 3.7594
Iteration: 788; Percent complete: 19.7%; Average loss: 3.4842
Iteration: 789; Percent complete: 19.7%; Average loss: 3.7012
Iteration: 790; Percent complete: 19.8%; Average loss: 3.2953
Iteration: 791; Percent complete: 19.8%; Average loss: 3.7217
Iteration: 792; Percent complete: 19.8%; Average loss: 3.3819
Iteration: 793; Percent complete: 19.8%; Average loss: 3.5485
Iteration: 794; Percent complete: 19.9%; Average loss: 3.2370
Iteration: 795; Percent complete: 19.9%; Average loss: 3.4489
Iteration: 796; Percent complete: 19.9%; Average loss: 3.6494
Iteration: 797; Percent complete: 19.9%; Average loss: 3.4667
Iteration: 798; Percent complete: 20.0%; Average loss: 3.5049
Iteration: 799; Percent complete: 20.0%; Average loss: 3.6053
Iteration: 800; Percent complete: 20.0%; Average loss: 3.3439
Iteration: 801; Percent complete: 20.0%; Average loss: 3.5444
Iteration: 802; Percent complete: 20.1%; Average loss: 3.2219
Iteration: 803; Percent complete: 20.1%; Average loss: 3.7114
Iteration: 804; Percent complete: 20.1%; Average loss: 3.3763
Iteration: 805; Percent complete: 20.1%; Average loss: 3.4755
Iteration: 806; Percent complete: 20.2%; Average loss: 3.4348
Iteration: 807; Percent complete: 20.2%; Average loss: 3.6513
Iteration: 808; Percent complete: 20.2%; Average loss: 3.5919
Iteration: 809; Percent complete: 20.2%; Average loss: 3.7649
Iteration: 810; Percent complete: 20.2%; Average loss: 3.4470
Iteration: 811; Percent complete: 20.3%; Average loss: 3.4247
Iteration: 812; Percent complete: 20.3%; Average loss: 3.5424
Iteration: 813; Percent complete: 20.3%; Average loss: 3.2992
Iteration: 814; Percent complete: 20.3%; Average loss: 3.6030
Iteration: 815; Percent complete: 20.4%; Average loss: 3.2973
Iteration: 816; Percent complete: 20.4%; Average loss: 3.6122
Iteration: 817; Percent complete: 20.4%; Average loss: 3.6667
Iteration: 818; Percent complete: 20.4%; Average loss: 3.6945
Iteration: 819; Percent complete: 20.5%; Average loss: 3.3547
Iteration: 820; Percent complete: 20.5%; Average loss: 3.5449
Iteration: 821; Percent complete: 20.5%; Average loss: 3.7515
Iteration: 822; Percent complete: 20.5%; Average loss: 3.9522
Iteration: 823; Percent complete: 20.6%; Average loss: 3.4008
Iteration: 824; Percent complete: 20.6%; Average loss: 3.6022
Iteration: 825; Percent complete: 20.6%; Average loss: 3.6164
Iteration: 826; Percent complete: 20.6%; Average loss: 3.6481
Iteration: 827; Percent complete: 20.7%; Average loss: 3.5612
Iteration: 828; Percent complete: 20.7%; Average loss: 3.5372
Iteration: 829; Percent complete: 20.7%; Average loss: 3.2993
Iteration: 830; Percent complete: 20.8%; Average loss: 3.4860
Iteration: 831; Percent complete: 20.8%; Average loss: 3.6620
Iteration: 832; Percent complete: 20.8%; Average loss: 3.4065
Iteration: 833; Percent complete: 20.8%; Average loss: 3.6343
Iteration: 834; Percent complete: 20.8%; Average loss: 3.6488
Iteration: 835; Percent complete: 20.9%; Average loss: 3.6106
Iteration: 836; Percent complete: 20.9%; Average loss: 3.5725
Iteration: 837; Percent complete: 20.9%; Average loss: 3.4599
Iteration: 838; Percent complete: 20.9%; Average loss: 3.4317
Iteration: 839; Percent complete: 21.0%; Average loss: 3.5843
Iteration: 840; Percent complete: 21.0%; Average loss: 3.4534
Iteration: 841; Percent complete: 21.0%; Average loss: 3.6558
Iteration: 842; Percent complete: 21.1%; Average loss: 3.5939
Iteration: 843; Percent complete: 21.1%; Average loss: 3.5500
Iteration: 844; Percent complete: 21.1%; Average loss: 3.4433
Iteration: 845; Percent complete: 21.1%; Average loss: 3.5728
Iteration: 846; Percent complete: 21.1%; Average loss: 3.5703
Iteration: 847; Percent complete: 21.2%; Average loss: 3.5613
Iteration: 848; Percent complete: 21.2%; Average loss: 3.6046
Iteration: 849; Percent complete: 21.2%; Average loss: 3.6233
Iteration: 850; Percent complete: 21.2%; Average loss: 3.6822
Iteration: 851; Percent complete: 21.3%; Average loss: 3.5426
Iteration: 852; Percent complete: 21.3%; Average loss: 3.4280
Iteration: 853; Percent complete: 21.3%; Average loss: 3.5370
Iteration: 854; Percent complete: 21.3%; Average loss: 3.4705
Iteration: 855; Percent complete: 21.4%; Average loss: 3.5495
Iteration: 856; Percent complete: 21.4%; Average loss: 3.4806
Iteration: 857; Percent complete: 21.4%; Average loss: 3.6012
Iteration: 858; Percent complete: 21.4%; Average loss: 3.5160
Iteration: 859; Percent complete: 21.5%; Average loss: 3.4522
Iteration: 860; Percent complete: 21.5%; Average loss: 3.3452
Iteration: 861; Percent complete: 21.5%; Average loss: 3.4710
Iteration: 862; Percent complete: 21.6%; Average loss: 3.5624
Iteration: 863; Percent complete: 21.6%; Average loss: 3.6644
Iteration: 864; Percent complete: 21.6%; Average loss: 3.5700
Iteration: 865; Percent complete: 21.6%; Average loss: 3.5977
Iteration: 866; Percent complete: 21.6%; Average loss: 3.4797
Iteration: 867; Percent complete: 21.7%; Average loss: 3.4282
Iteration: 868; Percent complete: 21.7%; Average loss: 3.3703
Iteration: 869; Percent complete: 21.7%; Average loss: 3.6710
Iteration: 870; Percent complete: 21.8%; Average loss: 3.5489
Iteration: 871; Percent complete: 21.8%; Average loss: 3.4419
Iteration: 872; Percent complete: 21.8%; Average loss: 3.6455
Iteration: 873; Percent complete: 21.8%; Average loss: 3.3300
Iteration: 874; Percent complete: 21.9%; Average loss: 3.1999
Iteration: 875; Percent complete: 21.9%; Average loss: 3.3690
Iteration: 876; Percent complete: 21.9%; Average loss: 3.5168
Iteration: 877; Percent complete: 21.9%; Average loss: 3.6015
Iteration: 878; Percent complete: 21.9%; Average loss: 3.3561
Iteration: 879; Percent complete: 22.0%; Average loss: 3.6229
Iteration: 880; Percent complete: 22.0%; Average loss: 3.5929
Iteration: 881; Percent complete: 22.0%; Average loss: 3.4219
Iteration: 882; Percent complete: 22.1%; Average loss: 3.6991
Iteration: 883; Percent complete: 22.1%; Average loss: 3.5404
Iteration: 884; Percent complete: 22.1%; Average loss: 3.5088
Iteration: 885; Percent complete: 22.1%; Average loss: 3.5746
Iteration: 886; Percent complete: 22.1%; Average loss: 3.3787
Iteration: 887; Percent complete: 22.2%; Average loss: 3.5193
Iteration: 888; Percent complete: 22.2%; Average loss: 3.4104
Iteration: 889; Percent complete: 22.2%; Average loss: 3.4873
Iteration: 890; Percent complete: 22.2%; Average loss: 3.5644
Iteration: 891; Percent complete: 22.3%; Average loss: 3.5834
Iteration: 892; Percent complete: 22.3%; Average loss: 3.4142
Iteration: 893; Percent complete: 22.3%; Average loss: 3.4441
Iteration: 894; Percent complete: 22.4%; Average loss: 3.6236
Iteration: 895; Percent complete: 22.4%; Average loss: 3.6148
Iteration: 896; Percent complete: 22.4%; Average loss: 3.7322
Iteration: 897; Percent complete: 22.4%; Average loss: 3.2684
Iteration: 898; Percent complete: 22.4%; Average loss: 3.4719
Iteration: 899; Percent complete: 22.5%; Average loss: 3.5524
Iteration: 900; Percent complete: 22.5%; Average loss: 3.6612
Iteration: 901; Percent complete: 22.5%; Average loss: 3.4268
Iteration: 902; Percent complete: 22.6%; Average loss: 3.3148
Iteration: 903; Percent complete: 22.6%; Average loss: 3.3289
Iteration: 904; Percent complete: 22.6%; Average loss: 3.2685
Iteration: 905; Percent complete: 22.6%; Average loss: 3.3271
Iteration: 906; Percent complete: 22.7%; Average loss: 3.5116
Iteration: 907; Percent complete: 22.7%; Average loss: 3.5838
Iteration: 908; Percent complete: 22.7%; Average loss: 3.8453
Iteration: 909; Percent complete: 22.7%; Average loss: 3.3544
Iteration: 910; Percent complete: 22.8%; Average loss: 3.5066
Iteration: 911; Percent complete: 22.8%; Average loss: 3.7142
Iteration: 912; Percent complete: 22.8%; Average loss: 3.8456
Iteration: 913; Percent complete: 22.8%; Average loss: 3.5377
Iteration: 914; Percent complete: 22.9%; Average loss: 3.3723
Iteration: 915; Percent complete: 22.9%; Average loss: 3.5405
Iteration: 916; Percent complete: 22.9%; Average loss: 3.2091
Iteration: 917; Percent complete: 22.9%; Average loss: 3.5238
Iteration: 918; Percent complete: 22.9%; Average loss: 3.3996
Iteration: 919; Percent complete: 23.0%; Average loss: 3.4327
Iteration: 920; Percent complete: 23.0%; Average loss: 3.4268
Iteration: 921; Percent complete: 23.0%; Average loss: 3.7553
Iteration: 922; Percent complete: 23.1%; Average loss: 3.5238
Iteration: 923; Percent complete: 23.1%; Average loss: 3.3916
Iteration: 924; Percent complete: 23.1%; Average loss: 3.5976
Iteration: 925; Percent complete: 23.1%; Average loss: 3.5849
Iteration: 926; Percent complete: 23.2%; Average loss: 3.6121
Iteration: 927; Percent complete: 23.2%; Average loss: 3.1054
Iteration: 928; Percent complete: 23.2%; Average loss: 3.4243
Iteration: 929; Percent complete: 23.2%; Average loss: 3.6645
Iteration: 930; Percent complete: 23.2%; Average loss: 3.3263
Iteration: 931; Percent complete: 23.3%; Average loss: 3.3157
Iteration: 932; Percent complete: 23.3%; Average loss: 3.4515
Iteration: 933; Percent complete: 23.3%; Average loss: 3.4620
Iteration: 934; Percent complete: 23.4%; Average loss: 3.8871
Iteration: 935; Percent complete: 23.4%; Average loss: 3.4011
Iteration: 936; Percent complete: 23.4%; Average loss: 3.5410
Iteration: 937; Percent complete: 23.4%; Average loss: 3.3580
Iteration: 938; Percent complete: 23.4%; Average loss: 3.4138
Iteration: 939; Percent complete: 23.5%; Average loss: 3.4920
Iteration: 940; Percent complete: 23.5%; Average loss: 3.4267
Iteration: 941; Percent complete: 23.5%; Average loss: 3.7999
Iteration: 942; Percent complete: 23.5%; Average loss: 3.5149
Iteration: 943; Percent complete: 23.6%; Average loss: 3.6015
Iteration: 944; Percent complete: 23.6%; Average loss: 3.5777
Iteration: 945; Percent complete: 23.6%; Average loss: 3.3661
Iteration: 946; Percent complete: 23.6%; Average loss: 3.5295
Iteration: 947; Percent complete: 23.7%; Average loss: 3.5168
Iteration: 948; Percent complete: 23.7%; Average loss: 3.5562
Iteration: 949; Percent complete: 23.7%; Average loss: 3.1832
Iteration: 950; Percent complete: 23.8%; Average loss: 3.4259
Iteration: 951; Percent complete: 23.8%; Average loss: 3.3685
Iteration: 952; Percent complete: 23.8%; Average loss: 3.4030
Iteration: 953; Percent complete: 23.8%; Average loss: 3.4096
Iteration: 954; Percent complete: 23.8%; Average loss: 3.4062
Iteration: 955; Percent complete: 23.9%; Average loss: 3.5720
Iteration: 956; Percent complete: 23.9%; Average loss: 3.4573
Iteration: 957; Percent complete: 23.9%; Average loss: 3.4776
Iteration: 958; Percent complete: 23.9%; Average loss: 3.4873
Iteration: 959; Percent complete: 24.0%; Average loss: 3.3692
Iteration: 960; Percent complete: 24.0%; Average loss: 3.7615
Iteration: 961; Percent complete: 24.0%; Average loss: 3.3545
Iteration: 962; Percent complete: 24.1%; Average loss: 3.4140
Iteration: 963; Percent complete: 24.1%; Average loss: 3.5176
Iteration: 964; Percent complete: 24.1%; Average loss: 3.4684
Iteration: 965; Percent complete: 24.1%; Average loss: 3.4610
Iteration: 966; Percent complete: 24.1%; Average loss: 3.4979
Iteration: 967; Percent complete: 24.2%; Average loss: 3.4007
Iteration: 968; Percent complete: 24.2%; Average loss: 3.3774
Iteration: 969; Percent complete: 24.2%; Average loss: 3.5774
Iteration: 970; Percent complete: 24.2%; Average loss: 3.4227
Iteration: 971; Percent complete: 24.3%; Average loss: 3.6026
Iteration: 972; Percent complete: 24.3%; Average loss: 3.5021
Iteration: 973; Percent complete: 24.3%; Average loss: 3.3754
Iteration: 974; Percent complete: 24.3%; Average loss: 3.8624
Iteration: 975; Percent complete: 24.4%; Average loss: 3.4258
Iteration: 976; Percent complete: 24.4%; Average loss: 3.5934
Iteration: 977; Percent complete: 24.4%; Average loss: 3.6836
Iteration: 978; Percent complete: 24.4%; Average loss: 3.4449
Iteration: 979; Percent complete: 24.5%; Average loss: 3.6334
Iteration: 980; Percent complete: 24.5%; Average loss: 3.3878
Iteration: 981; Percent complete: 24.5%; Average loss: 3.5435
Iteration: 982; Percent complete: 24.6%; Average loss: 3.5398
Iteration: 983; Percent complete: 24.6%; Average loss: 3.3515
Iteration: 984; Percent complete: 24.6%; Average loss: 3.3247
Iteration: 985; Percent complete: 24.6%; Average loss: 3.1774
Iteration: 986; Percent complete: 24.6%; Average loss: 3.3936
Iteration: 987; Percent complete: 24.7%; Average loss: 3.3816
Iteration: 988; Percent complete: 24.7%; Average loss: 3.5853
Iteration: 989; Percent complete: 24.7%; Average loss: 3.4075
Iteration: 990; Percent complete: 24.8%; Average loss: 3.6807
Iteration: 991; Percent complete: 24.8%; Average loss: 3.6206
Iteration: 992; Percent complete: 24.8%; Average loss: 3.4426
Iteration: 993; Percent complete: 24.8%; Average loss: 3.4240
Iteration: 994; Percent complete: 24.9%; Average loss: 3.4741
Iteration: 995; Percent complete: 24.9%; Average loss: 3.4374
Iteration: 996; Percent complete: 24.9%; Average loss: 3.6430
Iteration: 997; Percent complete: 24.9%; Average loss: 3.2452
Iteration: 998; Percent complete: 24.9%; Average loss: 3.4775
Iteration: 999; Percent complete: 25.0%; Average loss: 3.6756
Iteration: 1000; Percent complete: 25.0%; Average loss: 3.5252
Iteration: 1001; Percent complete: 25.0%; Average loss: 3.5552
Iteration: 1002; Percent complete: 25.1%; Average loss: 3.4392
Iteration: 1003; Percent complete: 25.1%; Average loss: 3.5873
Iteration: 1004; Percent complete: 25.1%; Average loss: 3.4826
Iteration: 1005; Percent complete: 25.1%; Average loss: 3.3800
Iteration: 1006; Percent complete: 25.1%; Average loss: 3.5582
Iteration: 1007; Percent complete: 25.2%; Average loss: 3.3951
Iteration: 1008; Percent complete: 25.2%; Average loss: 3.5682
Iteration: 1009; Percent complete: 25.2%; Average loss: 3.4847
Iteration: 1010; Percent complete: 25.2%; Average loss: 3.4489
Iteration: 1011; Percent complete: 25.3%; Average loss: 3.5619
Iteration: 1012; Percent complete: 25.3%; Average loss: 3.3334
Iteration: 1013; Percent complete: 25.3%; Average loss: 3.5441
Iteration: 1014; Percent complete: 25.4%; Average loss: 3.5521
Iteration: 1015; Percent complete: 25.4%; Average loss: 3.5780
Iteration: 1016; Percent complete: 25.4%; Average loss: 3.4754
Iteration: 1017; Percent complete: 25.4%; Average loss: 3.3243
Iteration: 1018; Percent complete: 25.4%; Average loss: 3.1131
Iteration: 1019; Percent complete: 25.5%; Average loss: 3.4801
Iteration: 1020; Percent complete: 25.5%; Average loss: 3.3847
Iteration: 1021; Percent complete: 25.5%; Average loss: 3.6219
Iteration: 1022; Percent complete: 25.6%; Average loss: 3.2721
Iteration: 1023; Percent complete: 25.6%; Average loss: 3.1282
Iteration: 1024; Percent complete: 25.6%; Average loss: 3.3455
Iteration: 1025; Percent complete: 25.6%; Average loss: 3.3884
Iteration: 1026; Percent complete: 25.7%; Average loss: 3.2988
Iteration: 1027; Percent complete: 25.7%; Average loss: 3.6557
Iteration: 1028; Percent complete: 25.7%; Average loss: 3.6749
Iteration: 1029; Percent complete: 25.7%; Average loss: 3.4967
Iteration: 1030; Percent complete: 25.8%; Average loss: 3.6192
Iteration: 1031; Percent complete: 25.8%; Average loss: 3.1370
Iteration: 1032; Percent complete: 25.8%; Average loss: 3.6583
Iteration: 1033; Percent complete: 25.8%; Average loss: 3.4089
Iteration: 1034; Percent complete: 25.9%; Average loss: 3.3167
Iteration: 1035; Percent complete: 25.9%; Average loss: 3.3886
Iteration: 1036; Percent complete: 25.9%; Average loss: 3.5442
Iteration: 1037; Percent complete: 25.9%; Average loss: 3.3864
Iteration: 1038; Percent complete: 25.9%; Average loss: 3.4516
Iteration: 1039; Percent complete: 26.0%; Average loss: 3.3863
Iteration: 1040; Percent complete: 26.0%; Average loss: 3.7195
Iteration: 1041; Percent complete: 26.0%; Average loss: 3.4417
Iteration: 1042; Percent complete: 26.1%; Average loss: 3.4435
Iteration: 1043; Percent complete: 26.1%; Average loss: 3.5658
Iteration: 1044; Percent complete: 26.1%; Average loss: 3.4596
Iteration: 1045; Percent complete: 26.1%; Average loss: 3.5206
Iteration: 1046; Percent complete: 26.2%; Average loss: 3.4596
Iteration: 1047; Percent complete: 26.2%; Average loss: 3.2908
Iteration: 1048; Percent complete: 26.2%; Average loss: 3.2358
Iteration: 1049; Percent complete: 26.2%; Average loss: 3.4016
Iteration: 1050; Percent complete: 26.2%; Average loss: 3.2830
Iteration: 1051; Percent complete: 26.3%; Average loss: 3.3567
Iteration: 1052; Percent complete: 26.3%; Average loss: 3.2701
Iteration: 1053; Percent complete: 26.3%; Average loss: 3.0635
Iteration: 1054; Percent complete: 26.4%; Average loss: 3.7382
Iteration: 1055; Percent complete: 26.4%; Average loss: 3.6166
Iteration: 1056; Percent complete: 26.4%; Average loss: 3.3576
Iteration: 1057; Percent complete: 26.4%; Average loss: 3.2860
Iteration: 1058; Percent complete: 26.5%; Average loss: 3.4651
Iteration: 1059; Percent complete: 26.5%; Average loss: 3.5277
Iteration: 1060; Percent complete: 26.5%; Average loss: 3.4877
Iteration: 1061; Percent complete: 26.5%; Average loss: 3.5839
Iteration: 1062; Percent complete: 26.6%; Average loss: 3.4158
Iteration: 1063; Percent complete: 26.6%; Average loss: 3.3631
Iteration: 1064; Percent complete: 26.6%; Average loss: 3.5939
Iteration: 1065; Percent complete: 26.6%; Average loss: 3.4610
Iteration: 1066; Percent complete: 26.7%; Average loss: 3.3750
Iteration: 1067; Percent complete: 26.7%; Average loss: 3.4177
Iteration: 1068; Percent complete: 26.7%; Average loss: 3.4995
Iteration: 1069; Percent complete: 26.7%; Average loss: 3.5088
Iteration: 1070; Percent complete: 26.8%; Average loss: 3.5276
Iteration: 1071; Percent complete: 26.8%; Average loss: 3.3787
Iteration: 1072; Percent complete: 26.8%; Average loss: 3.4194
Iteration: 1073; Percent complete: 26.8%; Average loss: 3.3943
Iteration: 1074; Percent complete: 26.9%; Average loss: 3.2938
Iteration: 1075; Percent complete: 26.9%; Average loss: 3.3724
Iteration: 1076; Percent complete: 26.9%; Average loss: 3.6737
Iteration: 1077; Percent complete: 26.9%; Average loss: 3.2110
Iteration: 1078; Percent complete: 27.0%; Average loss: 3.4370
Iteration: 1079; Percent complete: 27.0%; Average loss: 3.4526
Iteration: 1080; Percent complete: 27.0%; Average loss: 3.2664
Iteration: 1081; Percent complete: 27.0%; Average loss: 3.3306
Iteration: 1082; Percent complete: 27.1%; Average loss: 3.4225
Iteration: 1083; Percent complete: 27.1%; Average loss: 3.5952
Iteration: 1084; Percent complete: 27.1%; Average loss: 3.3518
Iteration: 1085; Percent complete: 27.1%; Average loss: 3.4083
Iteration: 1086; Percent complete: 27.2%; Average loss: 3.2811
Iteration: 1087; Percent complete: 27.2%; Average loss: 3.1197
Iteration: 1088; Percent complete: 27.2%; Average loss: 3.4320
Iteration: 1089; Percent complete: 27.2%; Average loss: 3.3764
Iteration: 1090; Percent complete: 27.3%; Average loss: 3.4577
Iteration: 1091; Percent complete: 27.3%; Average loss: 3.4980
Iteration: 1092; Percent complete: 27.3%; Average loss: 3.3610
Iteration: 1093; Percent complete: 27.3%; Average loss: 3.3054
Iteration: 1094; Percent complete: 27.4%; Average loss: 3.2865
Iteration: 1095; Percent complete: 27.4%; Average loss: 3.7924
Iteration: 1096; Percent complete: 27.4%; Average loss: 3.4070
Iteration: 1097; Percent complete: 27.4%; Average loss: 3.4998
Iteration: 1098; Percent complete: 27.5%; Average loss: 3.5081
Iteration: 1099; Percent complete: 27.5%; Average loss: 3.3106
Iteration: 1100; Percent complete: 27.5%; Average loss: 3.2894
Iteration: 1101; Percent complete: 27.5%; Average loss: 3.1030
Iteration: 1102; Percent complete: 27.6%; Average loss: 3.3956
Iteration: 1103; Percent complete: 27.6%; Average loss: 3.4057
Iteration: 1104; Percent complete: 27.6%; Average loss: 3.2475
Iteration: 1105; Percent complete: 27.6%; Average loss: 3.4279
Iteration: 1106; Percent complete: 27.7%; Average loss: 3.2682
Iteration: 1107; Percent complete: 27.7%; Average loss: 3.5773
Iteration: 1108; Percent complete: 27.7%; Average loss: 3.3415
Iteration: 1109; Percent complete: 27.7%; Average loss: 3.3604
Iteration: 1110; Percent complete: 27.8%; Average loss: 3.5016
Iteration: 1111; Percent complete: 27.8%; Average loss: 3.5937
Iteration: 1112; Percent complete: 27.8%; Average loss: 3.6953
Iteration: 1113; Percent complete: 27.8%; Average loss: 3.3567
Iteration: 1114; Percent complete: 27.9%; Average loss: 3.3831
Iteration: 1115; Percent complete: 27.9%; Average loss: 3.4637
Iteration: 1116; Percent complete: 27.9%; Average loss: 3.4205
Iteration: 1117; Percent complete: 27.9%; Average loss: 3.4314
Iteration: 1118; Percent complete: 28.0%; Average loss: 3.5158
Iteration: 1119; Percent complete: 28.0%; Average loss: 3.3518
Iteration: 1120; Percent complete: 28.0%; Average loss: 3.4329
Iteration: 1121; Percent complete: 28.0%; Average loss: 3.3238
Iteration: 1122; Percent complete: 28.1%; Average loss: 3.2633
Iteration: 1123; Percent complete: 28.1%; Average loss: 3.4070
Iteration: 1124; Percent complete: 28.1%; Average loss: 3.2811
Iteration: 1125; Percent complete: 28.1%; Average loss: 3.4413
Iteration: 1126; Percent complete: 28.1%; Average loss: 3.3729
Iteration: 1127; Percent complete: 28.2%; Average loss: 3.3609
Iteration: 1128; Percent complete: 28.2%; Average loss: 3.4060
Iteration: 1129; Percent complete: 28.2%; Average loss: 3.4710
Iteration: 1130; Percent complete: 28.2%; Average loss: 3.4691
Iteration: 1131; Percent complete: 28.3%; Average loss: 3.6186
Iteration: 1132; Percent complete: 28.3%; Average loss: 3.4485
Iteration: 1133; Percent complete: 28.3%; Average loss: 3.2781
Iteration: 1134; Percent complete: 28.3%; Average loss: 3.6192
Iteration: 1135; Percent complete: 28.4%; Average loss: 3.2405
Iteration: 1136; Percent complete: 28.4%; Average loss: 3.6868
Iteration: 1137; Percent complete: 28.4%; Average loss: 3.3613
Iteration: 1138; Percent complete: 28.4%; Average loss: 3.4807
Iteration: 1139; Percent complete: 28.5%; Average loss: 3.3723
Iteration: 1140; Percent complete: 28.5%; Average loss: 3.3640
Iteration: 1141; Percent complete: 28.5%; Average loss: 3.5320
Iteration: 1142; Percent complete: 28.5%; Average loss: 3.2797
Iteration: 1143; Percent complete: 28.6%; Average loss: 3.3884
Iteration: 1144; Percent complete: 28.6%; Average loss: 3.3119
Iteration: 1145; Percent complete: 28.6%; Average loss: 3.3318
Iteration: 1146; Percent complete: 28.6%; Average loss: 3.3100
Iteration: 1147; Percent complete: 28.7%; Average loss: 3.5317
Iteration: 1148; Percent complete: 28.7%; Average loss: 3.4275
Iteration: 1149; Percent complete: 28.7%; Average loss: 3.2761
Iteration: 1150; Percent complete: 28.7%; Average loss: 3.3850
Iteration: 1151; Percent complete: 28.8%; Average loss: 3.3434
Iteration: 1152; Percent complete: 28.8%; Average loss: 3.5196
Iteration: 1153; Percent complete: 28.8%; Average loss: 3.3047
Iteration: 1154; Percent complete: 28.8%; Average loss: 3.4016
Iteration: 1155; Percent complete: 28.9%; Average loss: 3.7308
Iteration: 1156; Percent complete: 28.9%; Average loss: 3.2601
Iteration: 1157; Percent complete: 28.9%; Average loss: 3.5090
Iteration: 1158; Percent complete: 28.9%; Average loss: 3.4140
Iteration: 1159; Percent complete: 29.0%; Average loss: 3.2734
Iteration: 1160; Percent complete: 29.0%; Average loss: 3.3192
Iteration: 1161; Percent complete: 29.0%; Average loss: 3.4539
Iteration: 1162; Percent complete: 29.0%; Average loss: 3.2977
Iteration: 1163; Percent complete: 29.1%; Average loss: 3.2384
Iteration: 1164; Percent complete: 29.1%; Average loss: 3.4423
Iteration: 1165; Percent complete: 29.1%; Average loss: 3.2725
Iteration: 1166; Percent complete: 29.1%; Average loss: 3.4562
Iteration: 1167; Percent complete: 29.2%; Average loss: 3.2339
Iteration: 1168; Percent complete: 29.2%; Average loss: 3.4550
Iteration: 1169; Percent complete: 29.2%; Average loss: 3.3785
Iteration: 1170; Percent complete: 29.2%; Average loss: 3.4416
Iteration: 1171; Percent complete: 29.3%; Average loss: 3.1825
Iteration: 1172; Percent complete: 29.3%; Average loss: 3.3797
Iteration: 1173; Percent complete: 29.3%; Average loss: 3.1588
Iteration: 1174; Percent complete: 29.3%; Average loss: 3.5668
Iteration: 1175; Percent complete: 29.4%; Average loss: 3.5934
Iteration: 1176; Percent complete: 29.4%; Average loss: 3.4019
Iteration: 1177; Percent complete: 29.4%; Average loss: 3.2446
Iteration: 1178; Percent complete: 29.4%; Average loss: 3.1922
Iteration: 1179; Percent complete: 29.5%; Average loss: 3.5072
Iteration: 1180; Percent complete: 29.5%; Average loss: 3.5168
Iteration: 1181; Percent complete: 29.5%; Average loss: 3.2152
Iteration: 1182; Percent complete: 29.5%; Average loss: 3.4673
Iteration: 1183; Percent complete: 29.6%; Average loss: 3.3961
Iteration: 1184; Percent complete: 29.6%; Average loss: 3.5292
Iteration: 1185; Percent complete: 29.6%; Average loss: 3.3447
Iteration: 1186; Percent complete: 29.6%; Average loss: 3.2802
Iteration: 1187; Percent complete: 29.7%; Average loss: 3.4760
Iteration: 1188; Percent complete: 29.7%; Average loss: 3.1267
Iteration: 1189; Percent complete: 29.7%; Average loss: 3.5367
Iteration: 1190; Percent complete: 29.8%; Average loss: 3.5028
Iteration: 1191; Percent complete: 29.8%; Average loss: 3.2922
Iteration: 1192; Percent complete: 29.8%; Average loss: 3.3059
Iteration: 1193; Percent complete: 29.8%; Average loss: 3.5494
Iteration: 1194; Percent complete: 29.8%; Average loss: 3.2905
Iteration: 1195; Percent complete: 29.9%; Average loss: 3.2982
Iteration: 1196; Percent complete: 29.9%; Average loss: 3.4343
Iteration: 1197; Percent complete: 29.9%; Average loss: 3.6965
Iteration: 1198; Percent complete: 29.9%; Average loss: 3.3316
Iteration: 1199; Percent complete: 30.0%; Average loss: 3.2620
Iteration: 1200; Percent complete: 30.0%; Average loss: 3.2584
Iteration: 1201; Percent complete: 30.0%; Average loss: 3.2738
Iteration: 1202; Percent complete: 30.0%; Average loss: 3.3941
Iteration: 1203; Percent complete: 30.1%; Average loss: 3.5925
Iteration: 1204; Percent complete: 30.1%; Average loss: 3.3721
Iteration: 1205; Percent complete: 30.1%; Average loss: 3.4024
Iteration: 1206; Percent complete: 30.1%; Average loss: 3.2002
Iteration: 1207; Percent complete: 30.2%; Average loss: 3.5858
Iteration: 1208; Percent complete: 30.2%; Average loss: 3.5270
Iteration: 1209; Percent complete: 30.2%; Average loss: 3.4290
Iteration: 1210; Percent complete: 30.2%; Average loss: 3.3211
Iteration: 1211; Percent complete: 30.3%; Average loss: 3.4768
Iteration: 1212; Percent complete: 30.3%; Average loss: 3.3971
Iteration: 1213; Percent complete: 30.3%; Average loss: 3.2326
Iteration: 1214; Percent complete: 30.3%; Average loss: 3.4409
Iteration: 1215; Percent complete: 30.4%; Average loss: 3.2186
Iteration: 1216; Percent complete: 30.4%; Average loss: 3.3929
Iteration: 1217; Percent complete: 30.4%; Average loss: 3.4702
Iteration: 1218; Percent complete: 30.4%; Average loss: 3.4431
Iteration: 1219; Percent complete: 30.5%; Average loss: 3.6162
Iteration: 1220; Percent complete: 30.5%; Average loss: 3.4571
Iteration: 1221; Percent complete: 30.5%; Average loss: 3.4528
Iteration: 1222; Percent complete: 30.6%; Average loss: 3.5091
Iteration: 1223; Percent complete: 30.6%; Average loss: 3.1546
Iteration: 1224; Percent complete: 30.6%; Average loss: 3.2697
Iteration: 1225; Percent complete: 30.6%; Average loss: 3.6116
Iteration: 1226; Percent complete: 30.6%; Average loss: 3.3688
Iteration: 1227; Percent complete: 30.7%; Average loss: 3.3831
Iteration: 1228; Percent complete: 30.7%; Average loss: 3.4626
Iteration: 1229; Percent complete: 30.7%; Average loss: 3.2755
Iteration: 1230; Percent complete: 30.8%; Average loss: 3.3384
Iteration: 1231; Percent complete: 30.8%; Average loss: 3.2878
Iteration: 1232; Percent complete: 30.8%; Average loss: 3.1967
Iteration: 1233; Percent complete: 30.8%; Average loss: 3.4674
Iteration: 1234; Percent complete: 30.9%; Average loss: 3.4141
Iteration: 1235; Percent complete: 30.9%; Average loss: 3.4440
Iteration: 1236; Percent complete: 30.9%; Average loss: 3.2121
Iteration: 1237; Percent complete: 30.9%; Average loss: 3.5890
Iteration: 1238; Percent complete: 30.9%; Average loss: 3.2600
Iteration: 1239; Percent complete: 31.0%; Average loss: 3.7515
Iteration: 1240; Percent complete: 31.0%; Average loss: 3.3774
Iteration: 1241; Percent complete: 31.0%; Average loss: 3.2315
Iteration: 1242; Percent complete: 31.1%; Average loss: 3.4899
Iteration: 1243; Percent complete: 31.1%; Average loss: 3.3961
Iteration: 1244; Percent complete: 31.1%; Average loss: 3.2580
Iteration: 1245; Percent complete: 31.1%; Average loss: 3.2694
Iteration: 1246; Percent complete: 31.1%; Average loss: 3.3408
Iteration: 1247; Percent complete: 31.2%; Average loss: 3.6161
Iteration: 1248; Percent complete: 31.2%; Average loss: 3.4534
Iteration: 1249; Percent complete: 31.2%; Average loss: 3.2666
Iteration: 1250; Percent complete: 31.2%; Average loss: 3.5520
Iteration: 1251; Percent complete: 31.3%; Average loss: 3.2384
Iteration: 1252; Percent complete: 31.3%; Average loss: 3.4111
Iteration: 1253; Percent complete: 31.3%; Average loss: 3.2999
Iteration: 1254; Percent complete: 31.4%; Average loss: 3.3724
Iteration: 1255; Percent complete: 31.4%; Average loss: 3.2943
Iteration: 1256; Percent complete: 31.4%; Average loss: 3.4953
Iteration: 1257; Percent complete: 31.4%; Average loss: 3.0973
Iteration: 1258; Percent complete: 31.4%; Average loss: 3.7648
Iteration: 1259; Percent complete: 31.5%; Average loss: 3.2212
Iteration: 1260; Percent complete: 31.5%; Average loss: 3.3918
Iteration: 1261; Percent complete: 31.5%; Average loss: 3.2032
Iteration: 1262; Percent complete: 31.6%; Average loss: 3.4899
Iteration: 1263; Percent complete: 31.6%; Average loss: 3.4588
Iteration: 1264; Percent complete: 31.6%; Average loss: 3.4516
Iteration: 1265; Percent complete: 31.6%; Average loss: 3.3621
Iteration: 1266; Percent complete: 31.6%; Average loss: 3.4353
Iteration: 1267; Percent complete: 31.7%; Average loss: 3.3820
Iteration: 1268; Percent complete: 31.7%; Average loss: 3.3053
Iteration: 1269; Percent complete: 31.7%; Average loss: 3.4510
Iteration: 1270; Percent complete: 31.8%; Average loss: 3.4235
Iteration: 1271; Percent complete: 31.8%; Average loss: 3.5053
Iteration: 1272; Percent complete: 31.8%; Average loss: 3.3361
Iteration: 1273; Percent complete: 31.8%; Average loss: 3.3710
Iteration: 1274; Percent complete: 31.9%; Average loss: 3.3132
Iteration: 1275; Percent complete: 31.9%; Average loss: 3.2149
Iteration: 1276; Percent complete: 31.9%; Average loss: 3.1889
Iteration: 1277; Percent complete: 31.9%; Average loss: 3.3413
Iteration: 1278; Percent complete: 31.9%; Average loss: 3.2612
Iteration: 1279; Percent complete: 32.0%; Average loss: 3.4015
Iteration: 1280; Percent complete: 32.0%; Average loss: 3.0486
Iteration: 1281; Percent complete: 32.0%; Average loss: 3.3229
Iteration: 1282; Percent complete: 32.0%; Average loss: 3.2239
Iteration: 1283; Percent complete: 32.1%; Average loss: 3.4364
Iteration: 1284; Percent complete: 32.1%; Average loss: 3.3344
Iteration: 1285; Percent complete: 32.1%; Average loss: 3.5462
Iteration: 1286; Percent complete: 32.1%; Average loss: 3.1881
Iteration: 1287; Percent complete: 32.2%; Average loss: 3.2863
Iteration: 1288; Percent complete: 32.2%; Average loss: 3.5025
Iteration: 1289; Percent complete: 32.2%; Average loss: 3.3897
Iteration: 1290; Percent complete: 32.2%; Average loss: 3.2635
Iteration: 1291; Percent complete: 32.3%; Average loss: 3.2727
Iteration: 1292; Percent complete: 32.3%; Average loss: 3.4709
Iteration: 1293; Percent complete: 32.3%; Average loss: 3.3640
Iteration: 1294; Percent complete: 32.4%; Average loss: 3.4631
Iteration: 1295; Percent complete: 32.4%; Average loss: 3.0155
Iteration: 1296; Percent complete: 32.4%; Average loss: 3.4592
Iteration: 1297; Percent complete: 32.4%; Average loss: 3.2239
Iteration: 1298; Percent complete: 32.5%; Average loss: 3.2913
Iteration: 1299; Percent complete: 32.5%; Average loss: 3.2325
Iteration: 1300; Percent complete: 32.5%; Average loss: 3.1810
Iteration: 1301; Percent complete: 32.5%; Average loss: 3.3211
Iteration: 1302; Percent complete: 32.6%; Average loss: 3.3678
Iteration: 1303; Percent complete: 32.6%; Average loss: 3.0430
Iteration: 1304; Percent complete: 32.6%; Average loss: 3.1564
Iteration: 1305; Percent complete: 32.6%; Average loss: 3.6150
Iteration: 1306; Percent complete: 32.6%; Average loss: 2.9383
Iteration: 1307; Percent complete: 32.7%; Average loss: 3.1376
Iteration: 1308; Percent complete: 32.7%; Average loss: 3.3581
Iteration: 1309; Percent complete: 32.7%; Average loss: 3.2391
Iteration: 1310; Percent complete: 32.8%; Average loss: 3.5338
Iteration: 1311; Percent complete: 32.8%; Average loss: 3.2088
Iteration: 1312; Percent complete: 32.8%; Average loss: 3.6510
Iteration: 1313; Percent complete: 32.8%; Average loss: 3.2113
Iteration: 1314; Percent complete: 32.9%; Average loss: 3.2545
Iteration: 1315; Percent complete: 32.9%; Average loss: 3.3037
Iteration: 1316; Percent complete: 32.9%; Average loss: 3.3957
Iteration: 1317; Percent complete: 32.9%; Average loss: 3.5249
Iteration: 1318; Percent complete: 33.0%; Average loss: 3.1438
Iteration: 1319; Percent complete: 33.0%; Average loss: 3.1916
Iteration: 1320; Percent complete: 33.0%; Average loss: 3.2041
Iteration: 1321; Percent complete: 33.0%; Average loss: 3.1105
Iteration: 1322; Percent complete: 33.1%; Average loss: 3.3201
Iteration: 1323; Percent complete: 33.1%; Average loss: 3.3784
Iteration: 1324; Percent complete: 33.1%; Average loss: 3.3342
Iteration: 1325; Percent complete: 33.1%; Average loss: 3.5009
Iteration: 1326; Percent complete: 33.1%; Average loss: 3.1138
Iteration: 1327; Percent complete: 33.2%; Average loss: 3.6013
Iteration: 1328; Percent complete: 33.2%; Average loss: 3.4086
Iteration: 1329; Percent complete: 33.2%; Average loss: 3.2213
Iteration: 1330; Percent complete: 33.2%; Average loss: 3.1003
Iteration: 1331; Percent complete: 33.3%; Average loss: 3.1431
Iteration: 1332; Percent complete: 33.3%; Average loss: 3.2105
Iteration: 1333; Percent complete: 33.3%; Average loss: 3.0559
Iteration: 1334; Percent complete: 33.4%; Average loss: 3.0970
Iteration: 1335; Percent complete: 33.4%; Average loss: 3.3290
Iteration: 1336; Percent complete: 33.4%; Average loss: 3.5538
Iteration: 1337; Percent complete: 33.4%; Average loss: 3.5263
Iteration: 1338; Percent complete: 33.5%; Average loss: 3.3394
Iteration: 1339; Percent complete: 33.5%; Average loss: 3.4289
Iteration: 1340; Percent complete: 33.5%; Average loss: 3.3616
Iteration: 1341; Percent complete: 33.5%; Average loss: 3.4585
Iteration: 1342; Percent complete: 33.6%; Average loss: 3.3217
Iteration: 1343; Percent complete: 33.6%; Average loss: 3.4893
Iteration: 1344; Percent complete: 33.6%; Average loss: 3.2323
Iteration: 1345; Percent complete: 33.6%; Average loss: 3.3973
Iteration: 1346; Percent complete: 33.7%; Average loss: 3.4223
Iteration: 1347; Percent complete: 33.7%; Average loss: 3.3118
Iteration: 1348; Percent complete: 33.7%; Average loss: 3.1264
Iteration: 1349; Percent complete: 33.7%; Average loss: 3.4788
Iteration: 1350; Percent complete: 33.8%; Average loss: 3.3859
Iteration: 1351; Percent complete: 33.8%; Average loss: 3.4425
Iteration: 1352; Percent complete: 33.8%; Average loss: 3.4185
Iteration: 1353; Percent complete: 33.8%; Average loss: 3.2917
Iteration: 1354; Percent complete: 33.9%; Average loss: 3.2228
Iteration: 1355; Percent complete: 33.9%; Average loss: 3.3170
Iteration: 1356; Percent complete: 33.9%; Average loss: 3.2029
Iteration: 1357; Percent complete: 33.9%; Average loss: 3.4169
Iteration: 1358; Percent complete: 34.0%; Average loss: 3.3344
Iteration: 1359; Percent complete: 34.0%; Average loss: 3.7396
Iteration: 1360; Percent complete: 34.0%; Average loss: 3.3684
Iteration: 1361; Percent complete: 34.0%; Average loss: 3.2243
Iteration: 1362; Percent complete: 34.1%; Average loss: 3.1236
Iteration: 1363; Percent complete: 34.1%; Average loss: 3.2237
Iteration: 1364; Percent complete: 34.1%; Average loss: 3.4954
Iteration: 1365; Percent complete: 34.1%; Average loss: 3.3455
Iteration: 1366; Percent complete: 34.2%; Average loss: 3.3731
Iteration: 1367; Percent complete: 34.2%; Average loss: 3.2063
Iteration: 1368; Percent complete: 34.2%; Average loss: 3.2176
Iteration: 1369; Percent complete: 34.2%; Average loss: 3.2686
Iteration: 1370; Percent complete: 34.2%; Average loss: 3.2150
Iteration: 1371; Percent complete: 34.3%; Average loss: 3.3943
Iteration: 1372; Percent complete: 34.3%; Average loss: 3.4921
Iteration: 1373; Percent complete: 34.3%; Average loss: 3.2185
Iteration: 1374; Percent complete: 34.4%; Average loss: 3.5844
Iteration: 1375; Percent complete: 34.4%; Average loss: 3.4848
Iteration: 1376; Percent complete: 34.4%; Average loss: 3.4283
Iteration: 1377; Percent complete: 34.4%; Average loss: 3.4380
Iteration: 1378; Percent complete: 34.4%; Average loss: 3.2525
Iteration: 1379; Percent complete: 34.5%; Average loss: 3.2289
Iteration: 1380; Percent complete: 34.5%; Average loss: 3.2574
Iteration: 1381; Percent complete: 34.5%; Average loss: 3.4364
Iteration: 1382; Percent complete: 34.5%; Average loss: 3.3846
Iteration: 1383; Percent complete: 34.6%; Average loss: 3.4650
Iteration: 1384; Percent complete: 34.6%; Average loss: 3.3714
Iteration: 1385; Percent complete: 34.6%; Average loss: 3.2930
Iteration: 1386; Percent complete: 34.6%; Average loss: 3.3916
Iteration: 1387; Percent complete: 34.7%; Average loss: 3.1633
Iteration: 1388; Percent complete: 34.7%; Average loss: 3.1068
Iteration: 1389; Percent complete: 34.7%; Average loss: 3.4279
Iteration: 1390; Percent complete: 34.8%; Average loss: 3.4385
Iteration: 1391; Percent complete: 34.8%; Average loss: 3.6867
Iteration: 1392; Percent complete: 34.8%; Average loss: 3.3002
Iteration: 1393; Percent complete: 34.8%; Average loss: 3.0857
Iteration: 1394; Percent complete: 34.8%; Average loss: 3.2900
Iteration: 1395; Percent complete: 34.9%; Average loss: 3.2917
Iteration: 1396; Percent complete: 34.9%; Average loss: 3.5154
Iteration: 1397; Percent complete: 34.9%; Average loss: 3.6103
Iteration: 1398; Percent complete: 34.9%; Average loss: 3.3048
Iteration: 1399; Percent complete: 35.0%; Average loss: 3.1527
Iteration: 1400; Percent complete: 35.0%; Average loss: 3.2903
Iteration: 1401; Percent complete: 35.0%; Average loss: 3.5121
Iteration: 1402; Percent complete: 35.0%; Average loss: 3.2657
Iteration: 1403; Percent complete: 35.1%; Average loss: 3.3174
Iteration: 1404; Percent complete: 35.1%; Average loss: 3.3786
Iteration: 1405; Percent complete: 35.1%; Average loss: 3.3409
Iteration: 1406; Percent complete: 35.1%; Average loss: 3.2533
Iteration: 1407; Percent complete: 35.2%; Average loss: 3.2046
Iteration: 1408; Percent complete: 35.2%; Average loss: 3.3788
Iteration: 1409; Percent complete: 35.2%; Average loss: 3.3496
Iteration: 1410; Percent complete: 35.2%; Average loss: 3.2327
Iteration: 1411; Percent complete: 35.3%; Average loss: 3.3050
Iteration: 1412; Percent complete: 35.3%; Average loss: 3.3089
Iteration: 1413; Percent complete: 35.3%; Average loss: 3.3592
Iteration: 1414; Percent complete: 35.4%; Average loss: 3.2938
Iteration: 1415; Percent complete: 35.4%; Average loss: 3.4826
Iteration: 1416; Percent complete: 35.4%; Average loss: 3.1928
Iteration: 1417; Percent complete: 35.4%; Average loss: 3.4911
Iteration: 1418; Percent complete: 35.4%; Average loss: 3.5369
Iteration: 1419; Percent complete: 35.5%; Average loss: 3.1840
Iteration: 1420; Percent complete: 35.5%; Average loss: 3.2208
Iteration: 1421; Percent complete: 35.5%; Average loss: 3.5164
Iteration: 1422; Percent complete: 35.5%; Average loss: 3.1924
Iteration: 1423; Percent complete: 35.6%; Average loss: 3.5397
Iteration: 1424; Percent complete: 35.6%; Average loss: 3.2621
Iteration: 1425; Percent complete: 35.6%; Average loss: 3.3336
Iteration: 1426; Percent complete: 35.6%; Average loss: 3.1560
Iteration: 1427; Percent complete: 35.7%; Average loss: 3.0699
Iteration: 1428; Percent complete: 35.7%; Average loss: 3.6094
Iteration: 1429; Percent complete: 35.7%; Average loss: 3.3602
Iteration: 1430; Percent complete: 35.8%; Average loss: 3.4574
Iteration: 1431; Percent complete: 35.8%; Average loss: 3.4553
Iteration: 1432; Percent complete: 35.8%; Average loss: 3.3323
Iteration: 1433; Percent complete: 35.8%; Average loss: 3.2409
Iteration: 1434; Percent complete: 35.9%; Average loss: 3.5226
Iteration: 1435; Percent complete: 35.9%; Average loss: 3.0384
Iteration: 1436; Percent complete: 35.9%; Average loss: 3.3577
Iteration: 1437; Percent complete: 35.9%; Average loss: 3.2925
Iteration: 1438; Percent complete: 35.9%; Average loss: 3.2682
Iteration: 1439; Percent complete: 36.0%; Average loss: 3.2131
Iteration: 1440; Percent complete: 36.0%; Average loss: 3.5562
Iteration: 1441; Percent complete: 36.0%; Average loss: 3.1730
Iteration: 1442; Percent complete: 36.0%; Average loss: 3.2897
Iteration: 1443; Percent complete: 36.1%; Average loss: 3.1475
Iteration: 1444; Percent complete: 36.1%; Average loss: 3.3140
Iteration: 1445; Percent complete: 36.1%; Average loss: 3.1793
Iteration: 1446; Percent complete: 36.1%; Average loss: 3.2501
Iteration: 1447; Percent complete: 36.2%; Average loss: 3.2420
Iteration: 1448; Percent complete: 36.2%; Average loss: 3.5231
Iteration: 1449; Percent complete: 36.2%; Average loss: 3.1555
Iteration: 1450; Percent complete: 36.2%; Average loss: 3.4702
Iteration: 1451; Percent complete: 36.3%; Average loss: 3.3822
Iteration: 1452; Percent complete: 36.3%; Average loss: 3.1009
Iteration: 1453; Percent complete: 36.3%; Average loss: 3.4665
Iteration: 1454; Percent complete: 36.4%; Average loss: 3.1903
Iteration: 1455; Percent complete: 36.4%; Average loss: 3.2749
Iteration: 1456; Percent complete: 36.4%; Average loss: 3.3752
Iteration: 1457; Percent complete: 36.4%; Average loss: 3.4369
Iteration: 1458; Percent complete: 36.4%; Average loss: 3.2680
Iteration: 1459; Percent complete: 36.5%; Average loss: 3.5229
Iteration: 1460; Percent complete: 36.5%; Average loss: 3.4973
Iteration: 1461; Percent complete: 36.5%; Average loss: 3.3952
Iteration: 1462; Percent complete: 36.5%; Average loss: 3.3497
Iteration: 1463; Percent complete: 36.6%; Average loss: 3.3665
Iteration: 1464; Percent complete: 36.6%; Average loss: 3.2195
Iteration: 1465; Percent complete: 36.6%; Average loss: 3.4306
Iteration: 1466; Percent complete: 36.6%; Average loss: 3.4151
Iteration: 1467; Percent complete: 36.7%; Average loss: 3.4397
Iteration: 1468; Percent complete: 36.7%; Average loss: 3.0964
Iteration: 1469; Percent complete: 36.7%; Average loss: 3.3827
Iteration: 1470; Percent complete: 36.8%; Average loss: 3.5796
Iteration: 1471; Percent complete: 36.8%; Average loss: 3.4822
Iteration: 1472; Percent complete: 36.8%; Average loss: 3.2677
Iteration: 1473; Percent complete: 36.8%; Average loss: 3.2774
Iteration: 1474; Percent complete: 36.9%; Average loss: 3.3711
Iteration: 1475; Percent complete: 36.9%; Average loss: 3.4179
Iteration: 1476; Percent complete: 36.9%; Average loss: 3.4911
Iteration: 1477; Percent complete: 36.9%; Average loss: 3.3914
Iteration: 1478; Percent complete: 37.0%; Average loss: 3.4804
Iteration: 1479; Percent complete: 37.0%; Average loss: 3.2077
Iteration: 1480; Percent complete: 37.0%; Average loss: 3.1854
Iteration: 1481; Percent complete: 37.0%; Average loss: 3.2180
Iteration: 1482; Percent complete: 37.0%; Average loss: 3.5114
Iteration: 1483; Percent complete: 37.1%; Average loss: 3.1804
Iteration: 1484; Percent complete: 37.1%; Average loss: 3.2421
Iteration: 1485; Percent complete: 37.1%; Average loss: 3.2689
Iteration: 1486; Percent complete: 37.1%; Average loss: 3.0731
Iteration: 1487; Percent complete: 37.2%; Average loss: 3.3455
Iteration: 1488; Percent complete: 37.2%; Average loss: 3.5318
Iteration: 1489; Percent complete: 37.2%; Average loss: 3.2865
Iteration: 1490; Percent complete: 37.2%; Average loss: 3.6195
Iteration: 1491; Percent complete: 37.3%; Average loss: 3.2223
Iteration: 1492; Percent complete: 37.3%; Average loss: 3.3345
Iteration: 1493; Percent complete: 37.3%; Average loss: 3.2657
Iteration: 1494; Percent complete: 37.4%; Average loss: 3.3570
Iteration: 1495; Percent complete: 37.4%; Average loss: 3.1910
Iteration: 1496; Percent complete: 37.4%; Average loss: 3.3047
Iteration: 1497; Percent complete: 37.4%; Average loss: 3.2666
Iteration: 1498; Percent complete: 37.5%; Average loss: 3.0912
Iteration: 1499; Percent complete: 37.5%; Average loss: 3.2766
Iteration: 1500; Percent complete: 37.5%; Average loss: 3.3132
Iteration: 1501; Percent complete: 37.5%; Average loss: 3.0297
Iteration: 1502; Percent complete: 37.5%; Average loss: 3.2398
Iteration: 1503; Percent complete: 37.6%; Average loss: 3.4416
Iteration: 1504; Percent complete: 37.6%; Average loss: 3.2404
Iteration: 1505; Percent complete: 37.6%; Average loss: 3.5543
Iteration: 1506; Percent complete: 37.6%; Average loss: 3.1258
Iteration: 1507; Percent complete: 37.7%; Average loss: 3.1856
Iteration: 1508; Percent complete: 37.7%; Average loss: 3.3235
Iteration: 1509; Percent complete: 37.7%; Average loss: 3.1549
Iteration: 1510; Percent complete: 37.8%; Average loss: 3.4487
Iteration: 1511; Percent complete: 37.8%; Average loss: 3.3700
Iteration: 1512; Percent complete: 37.8%; Average loss: 3.2179
Iteration: 1513; Percent complete: 37.8%; Average loss: 3.0287
Iteration: 1514; Percent complete: 37.9%; Average loss: 3.3520
Iteration: 1515; Percent complete: 37.9%; Average loss: 3.6127
Iteration: 1516; Percent complete: 37.9%; Average loss: 3.4045
Iteration: 1517; Percent complete: 37.9%; Average loss: 3.3005
Iteration: 1518; Percent complete: 38.0%; Average loss: 3.4562
Iteration: 1519; Percent complete: 38.0%; Average loss: 3.2643
Iteration: 1520; Percent complete: 38.0%; Average loss: 3.3690
Iteration: 1521; Percent complete: 38.0%; Average loss: 3.1113
Iteration: 1522; Percent complete: 38.0%; Average loss: 3.2261
Iteration: 1523; Percent complete: 38.1%; Average loss: 3.3970
Iteration: 1524; Percent complete: 38.1%; Average loss: 3.2275
Iteration: 1525; Percent complete: 38.1%; Average loss: 3.2524
Iteration: 1526; Percent complete: 38.1%; Average loss: 3.4294
Iteration: 1527; Percent complete: 38.2%; Average loss: 3.3059
Iteration: 1528; Percent complete: 38.2%; Average loss: 3.1656
Iteration: 1529; Percent complete: 38.2%; Average loss: 3.1103
Iteration: 1530; Percent complete: 38.2%; Average loss: 3.3110
Iteration: 1531; Percent complete: 38.3%; Average loss: 3.3237
Iteration: 1532; Percent complete: 38.3%; Average loss: 3.1737
Iteration: 1533; Percent complete: 38.3%; Average loss: 3.0589
Iteration: 1534; Percent complete: 38.4%; Average loss: 3.5075
Iteration: 1535; Percent complete: 38.4%; Average loss: 3.2442
Iteration: 1536; Percent complete: 38.4%; Average loss: 3.2374
Iteration: 1537; Percent complete: 38.4%; Average loss: 3.3813
Iteration: 1538; Percent complete: 38.5%; Average loss: 3.1288
Iteration: 1539; Percent complete: 38.5%; Average loss: 3.4647
Iteration: 1540; Percent complete: 38.5%; Average loss: 3.1079
Iteration: 1541; Percent complete: 38.5%; Average loss: 3.3827
Iteration: 1542; Percent complete: 38.6%; Average loss: 3.2868
Iteration: 1543; Percent complete: 38.6%; Average loss: 3.1521
Iteration: 1544; Percent complete: 38.6%; Average loss: 3.1204
Iteration: 1545; Percent complete: 38.6%; Average loss: 3.3515
Iteration: 1546; Percent complete: 38.6%; Average loss: 3.3545
Iteration: 1547; Percent complete: 38.7%; Average loss: 3.3675
Iteration: 1548; Percent complete: 38.7%; Average loss: 3.3347
Iteration: 1549; Percent complete: 38.7%; Average loss: 3.3143
Iteration: 1550; Percent complete: 38.8%; Average loss: 3.3410
Iteration: 1551; Percent complete: 38.8%; Average loss: 3.2026
Iteration: 1552; Percent complete: 38.8%; Average loss: 3.4770
Iteration: 1553; Percent complete: 38.8%; Average loss: 3.1074
Iteration: 1554; Percent complete: 38.9%; Average loss: 3.3664
Iteration: 1555; Percent complete: 38.9%; Average loss: 3.2222
Iteration: 1556; Percent complete: 38.9%; Average loss: 3.4965
Iteration: 1557; Percent complete: 38.9%; Average loss: 3.3048
Iteration: 1558; Percent complete: 39.0%; Average loss: 3.1487
Iteration: 1559; Percent complete: 39.0%; Average loss: 3.1296
Iteration: 1560; Percent complete: 39.0%; Average loss: 3.1058
Iteration: 1561; Percent complete: 39.0%; Average loss: 3.3184
Iteration: 1562; Percent complete: 39.1%; Average loss: 3.3378
Iteration: 1563; Percent complete: 39.1%; Average loss: 3.5492
Iteration: 1564; Percent complete: 39.1%; Average loss: 3.3410
Iteration: 1565; Percent complete: 39.1%; Average loss: 3.2160
Iteration: 1566; Percent complete: 39.1%; Average loss: 3.2761
Iteration: 1567; Percent complete: 39.2%; Average loss: 3.3356
Iteration: 1568; Percent complete: 39.2%; Average loss: 3.2693
Iteration: 1569; Percent complete: 39.2%; Average loss: 3.2543
Iteration: 1570; Percent complete: 39.2%; Average loss: 3.2353
Iteration: 1571; Percent complete: 39.3%; Average loss: 3.2235
Iteration: 1572; Percent complete: 39.3%; Average loss: 3.4883
Iteration: 1573; Percent complete: 39.3%; Average loss: 3.2640
Iteration: 1574; Percent complete: 39.4%; Average loss: 3.2472
Iteration: 1575; Percent complete: 39.4%; Average loss: 3.2076
Iteration: 1576; Percent complete: 39.4%; Average loss: 3.2638
Iteration: 1577; Percent complete: 39.4%; Average loss: 3.3838
Iteration: 1578; Percent complete: 39.5%; Average loss: 2.9980
Iteration: 1579; Percent complete: 39.5%; Average loss: 3.3008
Iteration: 1580; Percent complete: 39.5%; Average loss: 3.2447
Iteration: 1581; Percent complete: 39.5%; Average loss: 3.5763
Iteration: 1582; Percent complete: 39.6%; Average loss: 3.3948
Iteration: 1583; Percent complete: 39.6%; Average loss: 3.0937
Iteration: 1584; Percent complete: 39.6%; Average loss: 3.4063
Iteration: 1585; Percent complete: 39.6%; Average loss: 3.1825
Iteration: 1586; Percent complete: 39.6%; Average loss: 3.2559
Iteration: 1587; Percent complete: 39.7%; Average loss: 3.2221
Iteration: 1588; Percent complete: 39.7%; Average loss: 3.2601
Iteration: 1589; Percent complete: 39.7%; Average loss: 3.2461
Iteration: 1590; Percent complete: 39.8%; Average loss: 3.2754
Iteration: 1591; Percent complete: 39.8%; Average loss: 3.0361
Iteration: 1592; Percent complete: 39.8%; Average loss: 3.4335
Iteration: 1593; Percent complete: 39.8%; Average loss: 3.2445
Iteration: 1594; Percent complete: 39.9%; Average loss: 3.3703
Iteration: 1595; Percent complete: 39.9%; Average loss: 3.2574
Iteration: 1596; Percent complete: 39.9%; Average loss: 2.9619
Iteration: 1597; Percent complete: 39.9%; Average loss: 3.2296
Iteration: 1598; Percent complete: 40.0%; Average loss: 3.2613
Iteration: 1599; Percent complete: 40.0%; Average loss: 3.3084
Iteration: 1600; Percent complete: 40.0%; Average loss: 3.1252
Iteration: 1601; Percent complete: 40.0%; Average loss: 3.0962
Iteration: 1602; Percent complete: 40.1%; Average loss: 3.2033
Iteration: 1603; Percent complete: 40.1%; Average loss: 3.1631
Iteration: 1604; Percent complete: 40.1%; Average loss: 3.2349
Iteration: 1605; Percent complete: 40.1%; Average loss: 3.4295
Iteration: 1606; Percent complete: 40.2%; Average loss: 3.1774
Iteration: 1607; Percent complete: 40.2%; Average loss: 3.4225
Iteration: 1608; Percent complete: 40.2%; Average loss: 3.3350
Iteration: 1609; Percent complete: 40.2%; Average loss: 3.2977
Iteration: 1610; Percent complete: 40.2%; Average loss: 3.2733
Iteration: 1611; Percent complete: 40.3%; Average loss: 3.3210
Iteration: 1612; Percent complete: 40.3%; Average loss: 3.1395
Iteration: 1613; Percent complete: 40.3%; Average loss: 3.1367
Iteration: 1614; Percent complete: 40.4%; Average loss: 3.0651
Iteration: 1615; Percent complete: 40.4%; Average loss: 3.2321
Iteration: 1616; Percent complete: 40.4%; Average loss: 3.0706
Iteration: 1617; Percent complete: 40.4%; Average loss: 3.1081
Iteration: 1618; Percent complete: 40.5%; Average loss: 3.3985
Iteration: 1619; Percent complete: 40.5%; Average loss: 3.3856
Iteration: 1620; Percent complete: 40.5%; Average loss: 3.3999
Iteration: 1621; Percent complete: 40.5%; Average loss: 3.2934
Iteration: 1622; Percent complete: 40.6%; Average loss: 3.1696
Iteration: 1623; Percent complete: 40.6%; Average loss: 3.4367
Iteration: 1624; Percent complete: 40.6%; Average loss: 3.1601
Iteration: 1625; Percent complete: 40.6%; Average loss: 3.3650
Iteration: 1626; Percent complete: 40.6%; Average loss: 3.3317
Iteration: 1627; Percent complete: 40.7%; Average loss: 3.3856
Iteration: 1628; Percent complete: 40.7%; Average loss: 3.1963
Iteration: 1629; Percent complete: 40.7%; Average loss: 3.3234
Iteration: 1630; Percent complete: 40.8%; Average loss: 3.3461
Iteration: 1631; Percent complete: 40.8%; Average loss: 3.1775
Iteration: 1632; Percent complete: 40.8%; Average loss: 3.4374
Iteration: 1633; Percent complete: 40.8%; Average loss: 3.1931
Iteration: 1634; Percent complete: 40.8%; Average loss: 3.2892
Iteration: 1635; Percent complete: 40.9%; Average loss: 3.0565
Iteration: 1636; Percent complete: 40.9%; Average loss: 3.3374
Iteration: 1637; Percent complete: 40.9%; Average loss: 3.1254
Iteration: 1638; Percent complete: 40.9%; Average loss: 3.1348
Iteration: 1639; Percent complete: 41.0%; Average loss: 3.3204
Iteration: 1640; Percent complete: 41.0%; Average loss: 3.3174
Iteration: 1641; Percent complete: 41.0%; Average loss: 3.3801
Iteration: 1642; Percent complete: 41.0%; Average loss: 3.1793
Iteration: 1643; Percent complete: 41.1%; Average loss: 3.4916
Iteration: 1644; Percent complete: 41.1%; Average loss: 3.1879
Iteration: 1645; Percent complete: 41.1%; Average loss: 3.1938
Iteration: 1646; Percent complete: 41.1%; Average loss: 3.3064
Iteration: 1647; Percent complete: 41.2%; Average loss: 3.1812
Iteration: 1648; Percent complete: 41.2%; Average loss: 3.3290
Iteration: 1649; Percent complete: 41.2%; Average loss: 3.2133
Iteration: 1650; Percent complete: 41.2%; Average loss: 3.3611
Iteration: 1651; Percent complete: 41.3%; Average loss: 3.2010
Iteration: 1652; Percent complete: 41.3%; Average loss: 2.8829
Iteration: 1653; Percent complete: 41.3%; Average loss: 3.0925
Iteration: 1654; Percent complete: 41.3%; Average loss: 3.1041
Iteration: 1655; Percent complete: 41.4%; Average loss: 3.4432
Iteration: 1656; Percent complete: 41.4%; Average loss: 3.2698
Iteration: 1657; Percent complete: 41.4%; Average loss: 3.3684
Iteration: 1658; Percent complete: 41.4%; Average loss: 3.1845
Iteration: 1659; Percent complete: 41.5%; Average loss: 2.8482
Iteration: 1660; Percent complete: 41.5%; Average loss: 3.2165
Iteration: 1661; Percent complete: 41.5%; Average loss: 3.0607
Iteration: 1662; Percent complete: 41.5%; Average loss: 2.9843
Iteration: 1663; Percent complete: 41.6%; Average loss: 2.8840
Iteration: 1664; Percent complete: 41.6%; Average loss: 3.2049
Iteration: 1665; Percent complete: 41.6%; Average loss: 3.5160
Iteration: 1666; Percent complete: 41.6%; Average loss: 3.2936
Iteration: 1667; Percent complete: 41.7%; Average loss: 3.2197
Iteration: 1668; Percent complete: 41.7%; Average loss: 3.2336
Iteration: 1669; Percent complete: 41.7%; Average loss: 3.3256
Iteration: 1670; Percent complete: 41.8%; Average loss: 3.3742
Iteration: 1671; Percent complete: 41.8%; Average loss: 3.1780
Iteration: 1672; Percent complete: 41.8%; Average loss: 3.3781
Iteration: 1673; Percent complete: 41.8%; Average loss: 3.0303
Iteration: 1674; Percent complete: 41.9%; Average loss: 3.2737
Iteration: 1675; Percent complete: 41.9%; Average loss: 3.3286
Iteration: 1676; Percent complete: 41.9%; Average loss: 3.0709
Iteration: 1677; Percent complete: 41.9%; Average loss: 3.3185
Iteration: 1678; Percent complete: 41.9%; Average loss: 3.0999
Iteration: 1679; Percent complete: 42.0%; Average loss: 3.1819
Iteration: 1680; Percent complete: 42.0%; Average loss: 3.0718
Iteration: 1681; Percent complete: 42.0%; Average loss: 3.2228
Iteration: 1682; Percent complete: 42.0%; Average loss: 3.1574
Iteration: 1683; Percent complete: 42.1%; Average loss: 3.3818
Iteration: 1684; Percent complete: 42.1%; Average loss: 3.1917
Iteration: 1685; Percent complete: 42.1%; Average loss: 3.4617
Iteration: 1686; Percent complete: 42.1%; Average loss: 3.1188
Iteration: 1687; Percent complete: 42.2%; Average loss: 3.1082
Iteration: 1688; Percent complete: 42.2%; Average loss: 3.3794
Iteration: 1689; Percent complete: 42.2%; Average loss: 2.8428
Iteration: 1690; Percent complete: 42.2%; Average loss: 3.4489
Iteration: 1691; Percent complete: 42.3%; Average loss: 3.0640
Iteration: 1692; Percent complete: 42.3%; Average loss: 3.3702
Iteration: 1693; Percent complete: 42.3%; Average loss: 3.2361
Iteration: 1694; Percent complete: 42.4%; Average loss: 3.3851
Iteration: 1695; Percent complete: 42.4%; Average loss: 3.1169
Iteration: 1696; Percent complete: 42.4%; Average loss: 3.2691
Iteration: 1697; Percent complete: 42.4%; Average loss: 3.1288
Iteration: 1698; Percent complete: 42.4%; Average loss: 3.3533
Iteration: 1699; Percent complete: 42.5%; Average loss: 3.3393
Iteration: 1700; Percent complete: 42.5%; Average loss: 3.1659
Iteration: 1701; Percent complete: 42.5%; Average loss: 3.0264
Iteration: 1702; Percent complete: 42.5%; Average loss: 3.4309
Iteration: 1703; Percent complete: 42.6%; Average loss: 3.2574
Iteration: 1704; Percent complete: 42.6%; Average loss: 3.4589
Iteration: 1705; Percent complete: 42.6%; Average loss: 3.4503
Iteration: 1706; Percent complete: 42.6%; Average loss: 3.2049
Iteration: 1707; Percent complete: 42.7%; Average loss: 3.0763
Iteration: 1708; Percent complete: 42.7%; Average loss: 3.4825
Iteration: 1709; Percent complete: 42.7%; Average loss: 2.9439
Iteration: 1710; Percent complete: 42.8%; Average loss: 3.1017
Iteration: 1711; Percent complete: 42.8%; Average loss: 3.3846
Iteration: 1712; Percent complete: 42.8%; Average loss: 3.1429
Iteration: 1713; Percent complete: 42.8%; Average loss: 3.2438
Iteration: 1714; Percent complete: 42.9%; Average loss: 3.1086
Iteration: 1715; Percent complete: 42.9%; Average loss: 2.9488
Iteration: 1716; Percent complete: 42.9%; Average loss: 3.3646
Iteration: 1717; Percent complete: 42.9%; Average loss: 3.2853
Iteration: 1718; Percent complete: 43.0%; Average loss: 3.2786
Iteration: 1719; Percent complete: 43.0%; Average loss: 3.2934
Iteration: 1720; Percent complete: 43.0%; Average loss: 3.0542
Iteration: 1721; Percent complete: 43.0%; Average loss: 3.2280
Iteration: 1722; Percent complete: 43.0%; Average loss: 3.0031
Iteration: 1723; Percent complete: 43.1%; Average loss: 3.2339
Iteration: 1724; Percent complete: 43.1%; Average loss: 2.9292
Iteration: 1725; Percent complete: 43.1%; Average loss: 3.1162
Iteration: 1726; Percent complete: 43.1%; Average loss: 3.1384
Iteration: 1727; Percent complete: 43.2%; Average loss: 3.2412
Iteration: 1728; Percent complete: 43.2%; Average loss: 3.2575
Iteration: 1729; Percent complete: 43.2%; Average loss: 3.2854
Iteration: 1730; Percent complete: 43.2%; Average loss: 2.9479
Iteration: 1731; Percent complete: 43.3%; Average loss: 3.0457
Iteration: 1732; Percent complete: 43.3%; Average loss: 3.1261
Iteration: 1733; Percent complete: 43.3%; Average loss: 3.0949
Iteration: 1734; Percent complete: 43.4%; Average loss: 3.3868
Iteration: 1735; Percent complete: 43.4%; Average loss: 2.8852
Iteration: 1736; Percent complete: 43.4%; Average loss: 3.2738
Iteration: 1737; Percent complete: 43.4%; Average loss: 3.0773
Iteration: 1738; Percent complete: 43.5%; Average loss: 2.9565
Iteration: 1739; Percent complete: 43.5%; Average loss: 3.0866
Iteration: 1740; Percent complete: 43.5%; Average loss: 3.0190
Iteration: 1741; Percent complete: 43.5%; Average loss: 3.1597
Iteration: 1742; Percent complete: 43.5%; Average loss: 3.2779
Iteration: 1743; Percent complete: 43.6%; Average loss: 3.2972
Iteration: 1744; Percent complete: 43.6%; Average loss: 3.3208
Iteration: 1745; Percent complete: 43.6%; Average loss: 3.2184
Iteration: 1746; Percent complete: 43.6%; Average loss: 3.2938
Iteration: 1747; Percent complete: 43.7%; Average loss: 3.0913
Iteration: 1748; Percent complete: 43.7%; Average loss: 3.3900
Iteration: 1749; Percent complete: 43.7%; Average loss: 3.3180
Iteration: 1750; Percent complete: 43.8%; Average loss: 3.1018
Iteration: 1751; Percent complete: 43.8%; Average loss: 3.1956
Iteration: 1752; Percent complete: 43.8%; Average loss: 3.3396
Iteration: 1753; Percent complete: 43.8%; Average loss: 3.2691
Iteration: 1754; Percent complete: 43.9%; Average loss: 3.0785
Iteration: 1755; Percent complete: 43.9%; Average loss: 3.2074
Iteration: 1756; Percent complete: 43.9%; Average loss: 3.3330
Iteration: 1757; Percent complete: 43.9%; Average loss: 3.3616
Iteration: 1758; Percent complete: 44.0%; Average loss: 3.1159
Iteration: 1759; Percent complete: 44.0%; Average loss: 2.9030
Iteration: 1760; Percent complete: 44.0%; Average loss: 3.0230
Iteration: 1761; Percent complete: 44.0%; Average loss: 3.4553
Iteration: 1762; Percent complete: 44.0%; Average loss: 3.1526
Iteration: 1763; Percent complete: 44.1%; Average loss: 3.5376
Iteration: 1764; Percent complete: 44.1%; Average loss: 3.3680
Iteration: 1765; Percent complete: 44.1%; Average loss: 2.9812
Iteration: 1766; Percent complete: 44.1%; Average loss: 3.3543
Iteration: 1767; Percent complete: 44.2%; Average loss: 3.1230
Iteration: 1768; Percent complete: 44.2%; Average loss: 3.0565
Iteration: 1769; Percent complete: 44.2%; Average loss: 3.2822
Iteration: 1770; Percent complete: 44.2%; Average loss: 3.1150
Iteration: 1771; Percent complete: 44.3%; Average loss: 3.0938
Iteration: 1772; Percent complete: 44.3%; Average loss: 3.2051
Iteration: 1773; Percent complete: 44.3%; Average loss: 3.1649
Iteration: 1774; Percent complete: 44.4%; Average loss: 3.0474
Iteration: 1775; Percent complete: 44.4%; Average loss: 2.9854
Iteration: 1776; Percent complete: 44.4%; Average loss: 3.1360
Iteration: 1777; Percent complete: 44.4%; Average loss: 3.3059
Iteration: 1778; Percent complete: 44.5%; Average loss: 2.9364
Iteration: 1779; Percent complete: 44.5%; Average loss: 2.9931
Iteration: 1780; Percent complete: 44.5%; Average loss: 3.0975
Iteration: 1781; Percent complete: 44.5%; Average loss: 3.0169
Iteration: 1782; Percent complete: 44.5%; Average loss: 3.2194
Iteration: 1783; Percent complete: 44.6%; Average loss: 3.2971
Iteration: 1784; Percent complete: 44.6%; Average loss: 3.2764
Iteration: 1785; Percent complete: 44.6%; Average loss: 3.2113
Iteration: 1786; Percent complete: 44.6%; Average loss: 3.3257
Iteration: 1787; Percent complete: 44.7%; Average loss: 3.1959
Iteration: 1788; Percent complete: 44.7%; Average loss: 3.1974
Iteration: 1789; Percent complete: 44.7%; Average loss: 3.0934
Iteration: 1790; Percent complete: 44.8%; Average loss: 3.1481
Iteration: 1791; Percent complete: 44.8%; Average loss: 3.2495
Iteration: 1792; Percent complete: 44.8%; Average loss: 3.1638
Iteration: 1793; Percent complete: 44.8%; Average loss: 3.0856
Iteration: 1794; Percent complete: 44.9%; Average loss: 3.2144
Iteration: 1795; Percent complete: 44.9%; Average loss: 3.0293
Iteration: 1796; Percent complete: 44.9%; Average loss: 3.1602
Iteration: 1797; Percent complete: 44.9%; Average loss: 3.1384
Iteration: 1798; Percent complete: 45.0%; Average loss: 3.2077
Iteration: 1799; Percent complete: 45.0%; Average loss: 3.1686
Iteration: 1800; Percent complete: 45.0%; Average loss: 3.2630
Iteration: 1801; Percent complete: 45.0%; Average loss: 3.5091
Iteration: 1802; Percent complete: 45.1%; Average loss: 3.1045
Iteration: 1803; Percent complete: 45.1%; Average loss: 3.1922
Iteration: 1804; Percent complete: 45.1%; Average loss: 3.2472
Iteration: 1805; Percent complete: 45.1%; Average loss: 3.4418
Iteration: 1806; Percent complete: 45.1%; Average loss: 2.9266
Iteration: 1807; Percent complete: 45.2%; Average loss: 3.1927
Iteration: 1808; Percent complete: 45.2%; Average loss: 3.1889
Iteration: 1809; Percent complete: 45.2%; Average loss: 3.6557
Iteration: 1810; Percent complete: 45.2%; Average loss: 3.1677
Iteration: 1811; Percent complete: 45.3%; Average loss: 2.8958
Iteration: 1812; Percent complete: 45.3%; Average loss: 3.2348
Iteration: 1813; Percent complete: 45.3%; Average loss: 2.9880
Iteration: 1814; Percent complete: 45.4%; Average loss: 3.3093
Iteration: 1815; Percent complete: 45.4%; Average loss: 3.1110
Iteration: 1816; Percent complete: 45.4%; Average loss: 3.2001
Iteration: 1817; Percent complete: 45.4%; Average loss: 3.1760
Iteration: 1818; Percent complete: 45.5%; Average loss: 3.0040
Iteration: 1819; Percent complete: 45.5%; Average loss: 3.3049
Iteration: 1820; Percent complete: 45.5%; Average loss: 2.9996
Iteration: 1821; Percent complete: 45.5%; Average loss: 3.2227
Iteration: 1822; Percent complete: 45.6%; Average loss: 3.2543
Iteration: 1823; Percent complete: 45.6%; Average loss: 3.2409
Iteration: 1824; Percent complete: 45.6%; Average loss: 3.1095
Iteration: 1825; Percent complete: 45.6%; Average loss: 3.1292
Iteration: 1826; Percent complete: 45.6%; Average loss: 3.2265
Iteration: 1827; Percent complete: 45.7%; Average loss: 3.3218
Iteration: 1828; Percent complete: 45.7%; Average loss: 3.2976
Iteration: 1829; Percent complete: 45.7%; Average loss: 3.1308
Iteration: 1830; Percent complete: 45.8%; Average loss: 3.3332
Iteration: 1831; Percent complete: 45.8%; Average loss: 3.2457
Iteration: 1832; Percent complete: 45.8%; Average loss: 3.3316
Iteration: 1833; Percent complete: 45.8%; Average loss: 3.2355
Iteration: 1834; Percent complete: 45.9%; Average loss: 3.2141
Iteration: 1835; Percent complete: 45.9%; Average loss: 3.3310
Iteration: 1836; Percent complete: 45.9%; Average loss: 3.2833
Iteration: 1837; Percent complete: 45.9%; Average loss: 3.0922
Iteration: 1838; Percent complete: 46.0%; Average loss: 3.2554
Iteration: 1839; Percent complete: 46.0%; Average loss: 3.3729
Iteration: 1840; Percent complete: 46.0%; Average loss: 3.1412
Iteration: 1841; Percent complete: 46.0%; Average loss: 2.9586
Iteration: 1842; Percent complete: 46.1%; Average loss: 3.0469
Iteration: 1843; Percent complete: 46.1%; Average loss: 3.5331
Iteration: 1844; Percent complete: 46.1%; Average loss: 2.9713
Iteration: 1845; Percent complete: 46.1%; Average loss: 3.0463
Iteration: 1846; Percent complete: 46.2%; Average loss: 2.6617
Iteration: 1847; Percent complete: 46.2%; Average loss: 3.3489
Iteration: 1848; Percent complete: 46.2%; Average loss: 3.4877
Iteration: 1849; Percent complete: 46.2%; Average loss: 3.1240
Iteration: 1850; Percent complete: 46.2%; Average loss: 3.1709
Iteration: 1851; Percent complete: 46.3%; Average loss: 2.9687
Iteration: 1852; Percent complete: 46.3%; Average loss: 3.1672
Iteration: 1853; Percent complete: 46.3%; Average loss: 3.0292
Iteration: 1854; Percent complete: 46.4%; Average loss: 3.0692
Iteration: 1855; Percent complete: 46.4%; Average loss: 3.3701
Iteration: 1856; Percent complete: 46.4%; Average loss: 3.2800
Iteration: 1857; Percent complete: 46.4%; Average loss: 3.5746
Iteration: 1858; Percent complete: 46.5%; Average loss: 2.8063
Iteration: 1859; Percent complete: 46.5%; Average loss: 3.3134
Iteration: 1860; Percent complete: 46.5%; Average loss: 3.1649
Iteration: 1861; Percent complete: 46.5%; Average loss: 2.8830
Iteration: 1862; Percent complete: 46.6%; Average loss: 2.8473
Iteration: 1863; Percent complete: 46.6%; Average loss: 3.3073
Iteration: 1864; Percent complete: 46.6%; Average loss: 3.1247
Iteration: 1865; Percent complete: 46.6%; Average loss: 3.2353
Iteration: 1866; Percent complete: 46.7%; Average loss: 2.8994
Iteration: 1867; Percent complete: 46.7%; Average loss: 2.9666
Iteration: 1868; Percent complete: 46.7%; Average loss: 3.2133
Iteration: 1869; Percent complete: 46.7%; Average loss: 3.2585
Iteration: 1870; Percent complete: 46.8%; Average loss: 3.1590
Iteration: 1871; Percent complete: 46.8%; Average loss: 2.9655
Iteration: 1872; Percent complete: 46.8%; Average loss: 3.1875
Iteration: 1873; Percent complete: 46.8%; Average loss: 3.1883
Iteration: 1874; Percent complete: 46.9%; Average loss: 3.3180
Iteration: 1875; Percent complete: 46.9%; Average loss: 3.1352
Iteration: 1876; Percent complete: 46.9%; Average loss: 3.1612
Iteration: 1877; Percent complete: 46.9%; Average loss: 2.8480
Iteration: 1878; Percent complete: 46.9%; Average loss: 3.1647
Iteration: 1879; Percent complete: 47.0%; Average loss: 3.1363
Iteration: 1880; Percent complete: 47.0%; Average loss: 2.6916
Iteration: 1881; Percent complete: 47.0%; Average loss: 3.0009
Iteration: 1882; Percent complete: 47.0%; Average loss: 3.2498
Iteration: 1883; Percent complete: 47.1%; Average loss: 3.1374
Iteration: 1884; Percent complete: 47.1%; Average loss: 2.9352
Iteration: 1885; Percent complete: 47.1%; Average loss: 3.2254
Iteration: 1886; Percent complete: 47.1%; Average loss: 3.3018
Iteration: 1887; Percent complete: 47.2%; Average loss: 3.2428
Iteration: 1888; Percent complete: 47.2%; Average loss: 3.4422
Iteration: 1889; Percent complete: 47.2%; Average loss: 3.2030
Iteration: 1890; Percent complete: 47.2%; Average loss: 3.0113
Iteration: 1891; Percent complete: 47.3%; Average loss: 3.3420
Iteration: 1892; Percent complete: 47.3%; Average loss: 2.9998
Iteration: 1893; Percent complete: 47.3%; Average loss: 3.4895
Iteration: 1894; Percent complete: 47.3%; Average loss: 2.8292
Iteration: 1895; Percent complete: 47.4%; Average loss: 3.1511
Iteration: 1896; Percent complete: 47.4%; Average loss: 3.2074
Iteration: 1897; Percent complete: 47.4%; Average loss: 3.4068
Iteration: 1898; Percent complete: 47.4%; Average loss: 3.1261
Iteration: 1899; Percent complete: 47.5%; Average loss: 3.1586
Iteration: 1900; Percent complete: 47.5%; Average loss: 3.1132
Iteration: 1901; Percent complete: 47.5%; Average loss: 3.5344
Iteration: 1902; Percent complete: 47.5%; Average loss: 3.3005
Iteration: 1903; Percent complete: 47.6%; Average loss: 3.0850
Iteration: 1904; Percent complete: 47.6%; Average loss: 3.4315
Iteration: 1905; Percent complete: 47.6%; Average loss: 3.1694
Iteration: 1906; Percent complete: 47.6%; Average loss: 3.0690
Iteration: 1907; Percent complete: 47.7%; Average loss: 3.0589
Iteration: 1908; Percent complete: 47.7%; Average loss: 2.9449
Iteration: 1909; Percent complete: 47.7%; Average loss: 2.9759
Iteration: 1910; Percent complete: 47.8%; Average loss: 3.4877
Iteration: 1911; Percent complete: 47.8%; Average loss: 3.0515
Iteration: 1912; Percent complete: 47.8%; Average loss: 3.3309
Iteration: 1913; Percent complete: 47.8%; Average loss: 3.0580
Iteration: 1914; Percent complete: 47.9%; Average loss: 3.0854
Iteration: 1915; Percent complete: 47.9%; Average loss: 2.9911
Iteration: 1916; Percent complete: 47.9%; Average loss: 3.1237
Iteration: 1917; Percent complete: 47.9%; Average loss: 3.2590
Iteration: 1918; Percent complete: 47.9%; Average loss: 3.3965
Iteration: 1919; Percent complete: 48.0%; Average loss: 3.2523
Iteration: 1920; Percent complete: 48.0%; Average loss: 3.2687
Iteration: 1921; Percent complete: 48.0%; Average loss: 3.3081
Iteration: 1922; Percent complete: 48.0%; Average loss: 3.0966
Iteration: 1923; Percent complete: 48.1%; Average loss: 3.1486
Iteration: 1924; Percent complete: 48.1%; Average loss: 3.1845
Iteration: 1925; Percent complete: 48.1%; Average loss: 3.2392
Iteration: 1926; Percent complete: 48.1%; Average loss: 3.1463
Iteration: 1927; Percent complete: 48.2%; Average loss: 3.3959
Iteration: 1928; Percent complete: 48.2%; Average loss: 3.1687
Iteration: 1929; Percent complete: 48.2%; Average loss: 3.0957
Iteration: 1930; Percent complete: 48.2%; Average loss: 3.3300
Iteration: 1931; Percent complete: 48.3%; Average loss: 3.1811
Iteration: 1932; Percent complete: 48.3%; Average loss: 3.3108
Iteration: 1933; Percent complete: 48.3%; Average loss: 3.2785
Iteration: 1934; Percent complete: 48.4%; Average loss: 3.2360
Iteration: 1935; Percent complete: 48.4%; Average loss: 3.0090
Iteration: 1936; Percent complete: 48.4%; Average loss: 3.1216
Iteration: 1937; Percent complete: 48.4%; Average loss: 3.2024
Iteration: 1938; Percent complete: 48.4%; Average loss: 3.1252
Iteration: 1939; Percent complete: 48.5%; Average loss: 3.0149
Iteration: 1940; Percent complete: 48.5%; Average loss: 3.3443
Iteration: 1941; Percent complete: 48.5%; Average loss: 2.9664
Iteration: 1942; Percent complete: 48.5%; Average loss: 3.0498
Iteration: 1943; Percent complete: 48.6%; Average loss: 3.1425
Iteration: 1944; Percent complete: 48.6%; Average loss: 3.0315
Iteration: 1945; Percent complete: 48.6%; Average loss: 3.0550
Iteration: 1946; Percent complete: 48.6%; Average loss: 3.2380
Iteration: 1947; Percent complete: 48.7%; Average loss: 3.3333
Iteration: 1948; Percent complete: 48.7%; Average loss: 2.9961
Iteration: 1949; Percent complete: 48.7%; Average loss: 3.0301
Iteration: 1950; Percent complete: 48.8%; Average loss: 3.1823
Iteration: 1951; Percent complete: 48.8%; Average loss: 3.0356
Iteration: 1952; Percent complete: 48.8%; Average loss: 2.9389
Iteration: 1953; Percent complete: 48.8%; Average loss: 3.2527
Iteration: 1954; Percent complete: 48.9%; Average loss: 2.9759
Iteration: 1955; Percent complete: 48.9%; Average loss: 2.9044
Iteration: 1956; Percent complete: 48.9%; Average loss: 2.9606
Iteration: 1957; Percent complete: 48.9%; Average loss: 3.2899
Iteration: 1958; Percent complete: 48.9%; Average loss: 3.1045
Iteration: 1959; Percent complete: 49.0%; Average loss: 2.9509
Iteration: 1960; Percent complete: 49.0%; Average loss: 2.8301
Iteration: 1961; Percent complete: 49.0%; Average loss: 3.2573
Iteration: 1962; Percent complete: 49.0%; Average loss: 3.2985
Iteration: 1963; Percent complete: 49.1%; Average loss: 3.3449
Iteration: 1964; Percent complete: 49.1%; Average loss: 2.9919
Iteration: 1965; Percent complete: 49.1%; Average loss: 3.0833
Iteration: 1966; Percent complete: 49.1%; Average loss: 3.1681
Iteration: 1967; Percent complete: 49.2%; Average loss: 2.9267
Iteration: 1968; Percent complete: 49.2%; Average loss: 3.2041
Iteration: 1969; Percent complete: 49.2%; Average loss: 3.1665
Iteration: 1970; Percent complete: 49.2%; Average loss: 3.0089
Iteration: 1971; Percent complete: 49.3%; Average loss: 3.2573
Iteration: 1972; Percent complete: 49.3%; Average loss: 2.9627
Iteration: 1973; Percent complete: 49.3%; Average loss: 3.0754
Iteration: 1974; Percent complete: 49.4%; Average loss: 3.1061
Iteration: 1975; Percent complete: 49.4%; Average loss: 2.9685
Iteration: 1976; Percent complete: 49.4%; Average loss: 3.1507
Iteration: 1977; Percent complete: 49.4%; Average loss: 3.2171
Iteration: 1978; Percent complete: 49.5%; Average loss: 3.1887
Iteration: 1979; Percent complete: 49.5%; Average loss: 3.0440
Iteration: 1980; Percent complete: 49.5%; Average loss: 3.0384
Iteration: 1981; Percent complete: 49.5%; Average loss: 2.9805
Iteration: 1982; Percent complete: 49.5%; Average loss: 2.9117
Iteration: 1983; Percent complete: 49.6%; Average loss: 3.0918
Iteration: 1984; Percent complete: 49.6%; Average loss: 3.1743
Iteration: 1985; Percent complete: 49.6%; Average loss: 3.0662
Iteration: 1986; Percent complete: 49.6%; Average loss: 3.1978
Iteration: 1987; Percent complete: 49.7%; Average loss: 3.1935
Iteration: 1988; Percent complete: 49.7%; Average loss: 3.0161
Iteration: 1989; Percent complete: 49.7%; Average loss: 3.1584
Iteration: 1990; Percent complete: 49.8%; Average loss: 3.2628
Iteration: 1991; Percent complete: 49.8%; Average loss: 2.9824
Iteration: 1992; Percent complete: 49.8%; Average loss: 3.1996
Iteration: 1993; Percent complete: 49.8%; Average loss: 3.1210
Iteration: 1994; Percent complete: 49.9%; Average loss: 2.8125
Iteration: 1995; Percent complete: 49.9%; Average loss: 3.4895
Iteration: 1996; Percent complete: 49.9%; Average loss: 3.5853
Iteration: 1997; Percent complete: 49.9%; Average loss: 3.2666
Iteration: 1998; Percent complete: 50.0%; Average loss: 3.2491
Iteration: 1999; Percent complete: 50.0%; Average loss: 3.2959
Iteration: 2000; Percent complete: 50.0%; Average loss: 3.1935
Iteration: 2001; Percent complete: 50.0%; Average loss: 3.2390
Iteration: 2002; Percent complete: 50.0%; Average loss: 3.0517
Iteration: 2003; Percent complete: 50.1%; Average loss: 3.2818
Iteration: 2004; Percent complete: 50.1%; Average loss: 3.1472
Iteration: 2005; Percent complete: 50.1%; Average loss: 2.9875
Iteration: 2006; Percent complete: 50.1%; Average loss: 3.2030
Iteration: 2007; Percent complete: 50.2%; Average loss: 2.8667
Iteration: 2008; Percent complete: 50.2%; Average loss: 2.9447
Iteration: 2009; Percent complete: 50.2%; Average loss: 3.1676
Iteration: 2010; Percent complete: 50.2%; Average loss: 3.0352
Iteration: 2011; Percent complete: 50.3%; Average loss: 3.3011
Iteration: 2012; Percent complete: 50.3%; Average loss: 3.0748
Iteration: 2013; Percent complete: 50.3%; Average loss: 3.4567
Iteration: 2014; Percent complete: 50.3%; Average loss: 2.8200
Iteration: 2015; Percent complete: 50.4%; Average loss: 3.3221
Iteration: 2016; Percent complete: 50.4%; Average loss: 3.0619
Iteration: 2017; Percent complete: 50.4%; Average loss: 3.0240
Iteration: 2018; Percent complete: 50.4%; Average loss: 3.1904
Iteration: 2019; Percent complete: 50.5%; Average loss: 2.9324
Iteration: 2020; Percent complete: 50.5%; Average loss: 3.4005
Iteration: 2021; Percent complete: 50.5%; Average loss: 3.2098
Iteration: 2022; Percent complete: 50.5%; Average loss: 3.4255
Iteration: 2023; Percent complete: 50.6%; Average loss: 3.3084
Iteration: 2024; Percent complete: 50.6%; Average loss: 2.8181
Iteration: 2025; Percent complete: 50.6%; Average loss: 2.9653
Iteration: 2026; Percent complete: 50.6%; Average loss: 3.2956
Iteration: 2027; Percent complete: 50.7%; Average loss: 3.1183
Iteration: 2028; Percent complete: 50.7%; Average loss: 3.2134
Iteration: 2029; Percent complete: 50.7%; Average loss: 3.3761
Iteration: 2030; Percent complete: 50.7%; Average loss: 3.0059
Iteration: 2031; Percent complete: 50.8%; Average loss: 3.0791
Iteration: 2032; Percent complete: 50.8%; Average loss: 3.1098
Iteration: 2033; Percent complete: 50.8%; Average loss: 3.0521
Iteration: 2034; Percent complete: 50.8%; Average loss: 3.3106
Iteration: 2035; Percent complete: 50.9%; Average loss: 2.9889
Iteration: 2036; Percent complete: 50.9%; Average loss: 3.2285
Iteration: 2037; Percent complete: 50.9%; Average loss: 3.5288
Iteration: 2038; Percent complete: 50.9%; Average loss: 3.2799
Iteration: 2039; Percent complete: 51.0%; Average loss: 3.1438
Iteration: 2040; Percent complete: 51.0%; Average loss: 3.2551
Iteration: 2041; Percent complete: 51.0%; Average loss: 3.2980
Iteration: 2042; Percent complete: 51.0%; Average loss: 3.1708
Iteration: 2043; Percent complete: 51.1%; Average loss: 3.2091
Iteration: 2044; Percent complete: 51.1%; Average loss: 3.0754
Iteration: 2045; Percent complete: 51.1%; Average loss: 3.1647
Iteration: 2046; Percent complete: 51.1%; Average loss: 3.0655
Iteration: 2047; Percent complete: 51.2%; Average loss: 3.0300
Iteration: 2048; Percent complete: 51.2%; Average loss: 2.9691
Iteration: 2049; Percent complete: 51.2%; Average loss: 2.9364
Iteration: 2050; Percent complete: 51.2%; Average loss: 3.1561
Iteration: 2051; Percent complete: 51.3%; Average loss: 3.0956
Iteration: 2052; Percent complete: 51.3%; Average loss: 3.1083
Iteration: 2053; Percent complete: 51.3%; Average loss: 3.2162
Iteration: 2054; Percent complete: 51.3%; Average loss: 3.0175
Iteration: 2055; Percent complete: 51.4%; Average loss: 3.1706
Iteration: 2056; Percent complete: 51.4%; Average loss: 3.0551
Iteration: 2057; Percent complete: 51.4%; Average loss: 3.1483
Iteration: 2058; Percent complete: 51.4%; Average loss: 3.0995
Iteration: 2059; Percent complete: 51.5%; Average loss: 2.9176
Iteration: 2060; Percent complete: 51.5%; Average loss: 3.0863
Iteration: 2061; Percent complete: 51.5%; Average loss: 3.1744
Iteration: 2062; Percent complete: 51.5%; Average loss: 3.3291
Iteration: 2063; Percent complete: 51.6%; Average loss: 3.1222
Iteration: 2064; Percent complete: 51.6%; Average loss: 3.2315
Iteration: 2065; Percent complete: 51.6%; Average loss: 3.0968
Iteration: 2066; Percent complete: 51.6%; Average loss: 3.1884
Iteration: 2067; Percent complete: 51.7%; Average loss: 2.9984
Iteration: 2068; Percent complete: 51.7%; Average loss: 3.0101
Iteration: 2069; Percent complete: 51.7%; Average loss: 2.9685
Iteration: 2070; Percent complete: 51.7%; Average loss: 3.1869
Iteration: 2071; Percent complete: 51.8%; Average loss: 3.1238
Iteration: 2072; Percent complete: 51.8%; Average loss: 3.1776
Iteration: 2073; Percent complete: 51.8%; Average loss: 3.2449
Iteration: 2074; Percent complete: 51.8%; Average loss: 3.1899
Iteration: 2075; Percent complete: 51.9%; Average loss: 2.9354
Iteration: 2076; Percent complete: 51.9%; Average loss: 2.9848
Iteration: 2077; Percent complete: 51.9%; Average loss: 3.1377
Iteration: 2078; Percent complete: 51.9%; Average loss: 3.0492
Iteration: 2079; Percent complete: 52.0%; Average loss: 3.1656
Iteration: 2080; Percent complete: 52.0%; Average loss: 3.1917
Iteration: 2081; Percent complete: 52.0%; Average loss: 3.2569
Iteration: 2082; Percent complete: 52.0%; Average loss: 3.1361
Iteration: 2083; Percent complete: 52.1%; Average loss: 3.1703
Iteration: 2084; Percent complete: 52.1%; Average loss: 3.1122
Iteration: 2085; Percent complete: 52.1%; Average loss: 3.0684
Iteration: 2086; Percent complete: 52.1%; Average loss: 3.0683
Iteration: 2087; Percent complete: 52.2%; Average loss: 3.1287
Iteration: 2088; Percent complete: 52.2%; Average loss: 2.9018
Iteration: 2089; Percent complete: 52.2%; Average loss: 3.1624
Iteration: 2090; Percent complete: 52.2%; Average loss: 2.9893
Iteration: 2091; Percent complete: 52.3%; Average loss: 3.0377
Iteration: 2092; Percent complete: 52.3%; Average loss: 2.9899
Iteration: 2093; Percent complete: 52.3%; Average loss: 3.1500
Iteration: 2094; Percent complete: 52.3%; Average loss: 3.1007
Iteration: 2095; Percent complete: 52.4%; Average loss: 3.2037
Iteration: 2096; Percent complete: 52.4%; Average loss: 2.8919
Iteration: 2097; Percent complete: 52.4%; Average loss: 3.0132
Iteration: 2098; Percent complete: 52.4%; Average loss: 3.0228
Iteration: 2099; Percent complete: 52.5%; Average loss: 3.2635
Iteration: 2100; Percent complete: 52.5%; Average loss: 3.0952
Iteration: 2101; Percent complete: 52.5%; Average loss: 3.0300
Iteration: 2102; Percent complete: 52.5%; Average loss: 3.1560
Iteration: 2103; Percent complete: 52.6%; Average loss: 3.2692
Iteration: 2104; Percent complete: 52.6%; Average loss: 2.9519
Iteration: 2105; Percent complete: 52.6%; Average loss: 3.3291
Iteration: 2106; Percent complete: 52.6%; Average loss: 3.0723
Iteration: 2107; Percent complete: 52.7%; Average loss: 3.1264
Iteration: 2108; Percent complete: 52.7%; Average loss: 3.3589
Iteration: 2109; Percent complete: 52.7%; Average loss: 3.1166
Iteration: 2110; Percent complete: 52.8%; Average loss: 2.9100
Iteration: 2111; Percent complete: 52.8%; Average loss: 3.2329
Iteration: 2112; Percent complete: 52.8%; Average loss: 3.1854
Iteration: 2113; Percent complete: 52.8%; Average loss: 3.1223
Iteration: 2114; Percent complete: 52.8%; Average loss: 2.9751
Iteration: 2115; Percent complete: 52.9%; Average loss: 3.1458
Iteration: 2116; Percent complete: 52.9%; Average loss: 3.3074
Iteration: 2117; Percent complete: 52.9%; Average loss: 2.9771
Iteration: 2118; Percent complete: 52.9%; Average loss: 3.1491
Iteration: 2119; Percent complete: 53.0%; Average loss: 3.1373
Iteration: 2120; Percent complete: 53.0%; Average loss: 3.1664
Iteration: 2121; Percent complete: 53.0%; Average loss: 3.2920
Iteration: 2122; Percent complete: 53.0%; Average loss: 3.1545
Iteration: 2123; Percent complete: 53.1%; Average loss: 3.0286
Iteration: 2124; Percent complete: 53.1%; Average loss: 3.2196
Iteration: 2125; Percent complete: 53.1%; Average loss: 3.1046
Iteration: 2126; Percent complete: 53.1%; Average loss: 3.1328
Iteration: 2127; Percent complete: 53.2%; Average loss: 3.0361
Iteration: 2128; Percent complete: 53.2%; Average loss: 3.0444
Iteration: 2129; Percent complete: 53.2%; Average loss: 3.0541
Iteration: 2130; Percent complete: 53.2%; Average loss: 3.1002
Iteration: 2131; Percent complete: 53.3%; Average loss: 3.1948
Iteration: 2132; Percent complete: 53.3%; Average loss: 2.9962
Iteration: 2133; Percent complete: 53.3%; Average loss: 3.2452
Iteration: 2134; Percent complete: 53.3%; Average loss: 3.2911
Iteration: 2135; Percent complete: 53.4%; Average loss: 3.2516
Iteration: 2136; Percent complete: 53.4%; Average loss: 3.1726
Iteration: 2137; Percent complete: 53.4%; Average loss: 3.0235
Iteration: 2138; Percent complete: 53.4%; Average loss: 3.2360
Iteration: 2139; Percent complete: 53.5%; Average loss: 3.2077
Iteration: 2140; Percent complete: 53.5%; Average loss: 3.0560
Iteration: 2141; Percent complete: 53.5%; Average loss: 3.2676
Iteration: 2142; Percent complete: 53.5%; Average loss: 2.7735
Iteration: 2143; Percent complete: 53.6%; Average loss: 3.1470
Iteration: 2144; Percent complete: 53.6%; Average loss: 3.0707
Iteration: 2145; Percent complete: 53.6%; Average loss: 3.1245
Iteration: 2146; Percent complete: 53.6%; Average loss: 2.8571
Iteration: 2147; Percent complete: 53.7%; Average loss: 3.0965
Iteration: 2148; Percent complete: 53.7%; Average loss: 2.9936
Iteration: 2149; Percent complete: 53.7%; Average loss: 3.1575
Iteration: 2150; Percent complete: 53.8%; Average loss: 3.3618
Iteration: 2151; Percent complete: 53.8%; Average loss: 3.0752
Iteration: 2152; Percent complete: 53.8%; Average loss: 2.8127
Iteration: 2153; Percent complete: 53.8%; Average loss: 3.2822
Iteration: 2154; Percent complete: 53.8%; Average loss: 3.1318
Iteration: 2155; Percent complete: 53.9%; Average loss: 3.3781
Iteration: 2156; Percent complete: 53.9%; Average loss: 2.9978
Iteration: 2157; Percent complete: 53.9%; Average loss: 3.1111
Iteration: 2158; Percent complete: 53.9%; Average loss: 2.8886
Iteration: 2159; Percent complete: 54.0%; Average loss: 2.9581
Iteration: 2160; Percent complete: 54.0%; Average loss: 3.0893
Iteration: 2161; Percent complete: 54.0%; Average loss: 3.2024
Iteration: 2162; Percent complete: 54.0%; Average loss: 3.1864
Iteration: 2163; Percent complete: 54.1%; Average loss: 2.9798
Iteration: 2164; Percent complete: 54.1%; Average loss: 3.1429
Iteration: 2165; Percent complete: 54.1%; Average loss: 3.0931
Iteration: 2166; Percent complete: 54.1%; Average loss: 3.0426
Iteration: 2167; Percent complete: 54.2%; Average loss: 2.9082
Iteration: 2168; Percent complete: 54.2%; Average loss: 2.9899
Iteration: 2169; Percent complete: 54.2%; Average loss: 2.9139
Iteration: 2170; Percent complete: 54.2%; Average loss: 3.2987
Iteration: 2171; Percent complete: 54.3%; Average loss: 3.1464
Iteration: 2172; Percent complete: 54.3%; Average loss: 3.1410
Iteration: 2173; Percent complete: 54.3%; Average loss: 3.1261
Iteration: 2174; Percent complete: 54.4%; Average loss: 3.0499
Iteration: 2175; Percent complete: 54.4%; Average loss: 3.1631
Iteration: 2176; Percent complete: 54.4%; Average loss: 3.1343
Iteration: 2177; Percent complete: 54.4%; Average loss: 3.2206
Iteration: 2178; Percent complete: 54.4%; Average loss: 2.8479
Iteration: 2179; Percent complete: 54.5%; Average loss: 3.1551
Iteration: 2180; Percent complete: 54.5%; Average loss: 3.0937
Iteration: 2181; Percent complete: 54.5%; Average loss: 3.0091
Iteration: 2182; Percent complete: 54.5%; Average loss: 3.2379
Iteration: 2183; Percent complete: 54.6%; Average loss: 3.1686
Iteration: 2184; Percent complete: 54.6%; Average loss: 3.0810
Iteration: 2185; Percent complete: 54.6%; Average loss: 3.1354
Iteration: 2186; Percent complete: 54.6%; Average loss: 3.1121
Iteration: 2187; Percent complete: 54.7%; Average loss: 3.2200
Iteration: 2188; Percent complete: 54.7%; Average loss: 3.3040
Iteration: 2189; Percent complete: 54.7%; Average loss: 3.2033
Iteration: 2190; Percent complete: 54.8%; Average loss: 3.0813
Iteration: 2191; Percent complete: 54.8%; Average loss: 3.0042
Iteration: 2192; Percent complete: 54.8%; Average loss: 3.3673
Iteration: 2193; Percent complete: 54.8%; Average loss: 2.8570
Iteration: 2194; Percent complete: 54.9%; Average loss: 2.8192
Iteration: 2195; Percent complete: 54.9%; Average loss: 2.8975
Iteration: 2196; Percent complete: 54.9%; Average loss: 3.2275
Iteration: 2197; Percent complete: 54.9%; Average loss: 2.9508
Iteration: 2198; Percent complete: 54.9%; Average loss: 3.2330
Iteration: 2199; Percent complete: 55.0%; Average loss: 3.0871
Iteration: 2200; Percent complete: 55.0%; Average loss: 3.1968
Iteration: 2201; Percent complete: 55.0%; Average loss: 3.1843
Iteration: 2202; Percent complete: 55.0%; Average loss: 3.0140
Iteration: 2203; Percent complete: 55.1%; Average loss: 3.2980
Iteration: 2204; Percent complete: 55.1%; Average loss: 3.0815
Iteration: 2205; Percent complete: 55.1%; Average loss: 2.9163
Iteration: 2206; Percent complete: 55.1%; Average loss: 3.2418
Iteration: 2207; Percent complete: 55.2%; Average loss: 2.9247
Iteration: 2208; Percent complete: 55.2%; Average loss: 3.0062
Iteration: 2209; Percent complete: 55.2%; Average loss: 3.1353
Iteration: 2210; Percent complete: 55.2%; Average loss: 3.2786
Iteration: 2211; Percent complete: 55.3%; Average loss: 3.2580
Iteration: 2212; Percent complete: 55.3%; Average loss: 3.1529
Iteration: 2213; Percent complete: 55.3%; Average loss: 3.0897
Iteration: 2214; Percent complete: 55.4%; Average loss: 3.0356
Iteration: 2215; Percent complete: 55.4%; Average loss: 3.1873
Iteration: 2216; Percent complete: 55.4%; Average loss: 2.7810
Iteration: 2217; Percent complete: 55.4%; Average loss: 3.0332
Iteration: 2218; Percent complete: 55.5%; Average loss: 3.1166
Iteration: 2219; Percent complete: 55.5%; Average loss: 3.2752
Iteration: 2220; Percent complete: 55.5%; Average loss: 3.1755
Iteration: 2221; Percent complete: 55.5%; Average loss: 3.1592
Iteration: 2222; Percent complete: 55.5%; Average loss: 2.9294
Iteration: 2223; Percent complete: 55.6%; Average loss: 2.8551
Iteration: 2224; Percent complete: 55.6%; Average loss: 3.0750
Iteration: 2225; Percent complete: 55.6%; Average loss: 3.2390
Iteration: 2226; Percent complete: 55.6%; Average loss: 2.9460
Iteration: 2227; Percent complete: 55.7%; Average loss: 3.1236
Iteration: 2228; Percent complete: 55.7%; Average loss: 3.0741
Iteration: 2229; Percent complete: 55.7%; Average loss: 3.1995
Iteration: 2230; Percent complete: 55.8%; Average loss: 3.0414
Iteration: 2231; Percent complete: 55.8%; Average loss: 2.8165
Iteration: 2232; Percent complete: 55.8%; Average loss: 2.8906
Iteration: 2233; Percent complete: 55.8%; Average loss: 3.0341
Iteration: 2234; Percent complete: 55.9%; Average loss: 3.4220
Iteration: 2235; Percent complete: 55.9%; Average loss: 3.3461
Iteration: 2236; Percent complete: 55.9%; Average loss: 3.3830
Iteration: 2237; Percent complete: 55.9%; Average loss: 3.2834
Iteration: 2238; Percent complete: 56.0%; Average loss: 3.0923
Iteration: 2239; Percent complete: 56.0%; Average loss: 3.1008
Iteration: 2240; Percent complete: 56.0%; Average loss: 3.0201
Iteration: 2241; Percent complete: 56.0%; Average loss: 2.9765
Iteration: 2242; Percent complete: 56.0%; Average loss: 3.1148
Iteration: 2243; Percent complete: 56.1%; Average loss: 3.0573
Iteration: 2244; Percent complete: 56.1%; Average loss: 3.2911
Iteration: 2245; Percent complete: 56.1%; Average loss: 3.0093
Iteration: 2246; Percent complete: 56.1%; Average loss: 2.8787
Iteration: 2247; Percent complete: 56.2%; Average loss: 2.9812
Iteration: 2248; Percent complete: 56.2%; Average loss: 2.9851
Iteration: 2249; Percent complete: 56.2%; Average loss: 2.8041
Iteration: 2250; Percent complete: 56.2%; Average loss: 3.0974
Iteration: 2251; Percent complete: 56.3%; Average loss: 3.0519
Iteration: 2252; Percent complete: 56.3%; Average loss: 3.2681
Iteration: 2253; Percent complete: 56.3%; Average loss: 2.8992
Iteration: 2254; Percent complete: 56.4%; Average loss: 3.0762
Iteration: 2255; Percent complete: 56.4%; Average loss: 2.9714
Iteration: 2256; Percent complete: 56.4%; Average loss: 3.0830
Iteration: 2257; Percent complete: 56.4%; Average loss: 3.3636
Iteration: 2258; Percent complete: 56.5%; Average loss: 3.0969
Iteration: 2259; Percent complete: 56.5%; Average loss: 3.2529
Iteration: 2260; Percent complete: 56.5%; Average loss: 2.8651
Iteration: 2261; Percent complete: 56.5%; Average loss: 3.1538
Iteration: 2262; Percent complete: 56.5%; Average loss: 2.9728
Iteration: 2263; Percent complete: 56.6%; Average loss: 3.0451
Iteration: 2264; Percent complete: 56.6%; Average loss: 2.9624
Iteration: 2265; Percent complete: 56.6%; Average loss: 3.0105
Iteration: 2266; Percent complete: 56.6%; Average loss: 3.0399
Iteration: 2267; Percent complete: 56.7%; Average loss: 2.9156
Iteration: 2268; Percent complete: 56.7%; Average loss: 3.1719
Iteration: 2269; Percent complete: 56.7%; Average loss: 3.0578
Iteration: 2270; Percent complete: 56.8%; Average loss: 2.9535
Iteration: 2271; Percent complete: 56.8%; Average loss: 3.1736
Iteration: 2272; Percent complete: 56.8%; Average loss: 2.6764
Iteration: 2273; Percent complete: 56.8%; Average loss: 3.2682
Iteration: 2274; Percent complete: 56.9%; Average loss: 3.0921
Iteration: 2275; Percent complete: 56.9%; Average loss: 3.3249
Iteration: 2276; Percent complete: 56.9%; Average loss: 2.9899
Iteration: 2277; Percent complete: 56.9%; Average loss: 3.1136
Iteration: 2278; Percent complete: 57.0%; Average loss: 3.0725
Iteration: 2279; Percent complete: 57.0%; Average loss: 3.0101
Iteration: 2280; Percent complete: 57.0%; Average loss: 3.1222
Iteration: 2281; Percent complete: 57.0%; Average loss: 2.9737
Iteration: 2282; Percent complete: 57.0%; Average loss: 3.1435
Iteration: 2283; Percent complete: 57.1%; Average loss: 3.1456
Iteration: 2284; Percent complete: 57.1%; Average loss: 3.2922
Iteration: 2285; Percent complete: 57.1%; Average loss: 3.2141
Iteration: 2286; Percent complete: 57.1%; Average loss: 3.0503
Iteration: 2287; Percent complete: 57.2%; Average loss: 2.8228
Iteration: 2288; Percent complete: 57.2%; Average loss: 3.1460
Iteration: 2289; Percent complete: 57.2%; Average loss: 3.1118
Iteration: 2290; Percent complete: 57.2%; Average loss: 2.8556
Iteration: 2291; Percent complete: 57.3%; Average loss: 2.9363
Iteration: 2292; Percent complete: 57.3%; Average loss: 3.1308
Iteration: 2293; Percent complete: 57.3%; Average loss: 3.1195
Iteration: 2294; Percent complete: 57.4%; Average loss: 2.6924
Iteration: 2295; Percent complete: 57.4%; Average loss: 3.2686
Iteration: 2296; Percent complete: 57.4%; Average loss: 3.1080
Iteration: 2297; Percent complete: 57.4%; Average loss: 2.8869
Iteration: 2298; Percent complete: 57.5%; Average loss: 3.0376
Iteration: 2299; Percent complete: 57.5%; Average loss: 3.0348
Iteration: 2300; Percent complete: 57.5%; Average loss: 3.0230
Iteration: 2301; Percent complete: 57.5%; Average loss: 2.9704
Iteration: 2302; Percent complete: 57.6%; Average loss: 3.1963
Iteration: 2303; Percent complete: 57.6%; Average loss: 3.2366
Iteration: 2304; Percent complete: 57.6%; Average loss: 2.9668
Iteration: 2305; Percent complete: 57.6%; Average loss: 2.9136
Iteration: 2306; Percent complete: 57.6%; Average loss: 3.0209
Iteration: 2307; Percent complete: 57.7%; Average loss: 2.9594
Iteration: 2308; Percent complete: 57.7%; Average loss: 2.9982
Iteration: 2309; Percent complete: 57.7%; Average loss: 3.0388
Iteration: 2310; Percent complete: 57.8%; Average loss: 3.0614
Iteration: 2311; Percent complete: 57.8%; Average loss: 3.1609
Iteration: 2312; Percent complete: 57.8%; Average loss: 3.2051
Iteration: 2313; Percent complete: 57.8%; Average loss: 3.0914
Iteration: 2314; Percent complete: 57.9%; Average loss: 3.2319
Iteration: 2315; Percent complete: 57.9%; Average loss: 2.8648
Iteration: 2316; Percent complete: 57.9%; Average loss: 3.0980
Iteration: 2317; Percent complete: 57.9%; Average loss: 3.0873
Iteration: 2318; Percent complete: 58.0%; Average loss: 3.3539
Iteration: 2319; Percent complete: 58.0%; Average loss: 3.1714
Iteration: 2320; Percent complete: 58.0%; Average loss: 3.0903
Iteration: 2321; Percent complete: 58.0%; Average loss: 3.1368
Iteration: 2322; Percent complete: 58.1%; Average loss: 3.1111
Iteration: 2323; Percent complete: 58.1%; Average loss: 2.8683
Iteration: 2324; Percent complete: 58.1%; Average loss: 3.0974
Iteration: 2325; Percent complete: 58.1%; Average loss: 2.9891
Iteration: 2326; Percent complete: 58.1%; Average loss: 3.1875
Iteration: 2327; Percent complete: 58.2%; Average loss: 2.9840
Iteration: 2328; Percent complete: 58.2%; Average loss: 3.1375
Iteration: 2329; Percent complete: 58.2%; Average loss: 2.9892
Iteration: 2330; Percent complete: 58.2%; Average loss: 3.2345
Iteration: 2331; Percent complete: 58.3%; Average loss: 2.9688
Iteration: 2332; Percent complete: 58.3%; Average loss: 3.0076
Iteration: 2333; Percent complete: 58.3%; Average loss: 2.9936
Iteration: 2334; Percent complete: 58.4%; Average loss: 3.0936
Iteration: 2335; Percent complete: 58.4%; Average loss: 2.9405
Iteration: 2336; Percent complete: 58.4%; Average loss: 3.0913
Iteration: 2337; Percent complete: 58.4%; Average loss: 3.1598
Iteration: 2338; Percent complete: 58.5%; Average loss: 3.0535
Iteration: 2339; Percent complete: 58.5%; Average loss: 3.1484
Iteration: 2340; Percent complete: 58.5%; Average loss: 3.0517
Iteration: 2341; Percent complete: 58.5%; Average loss: 3.0861
Iteration: 2342; Percent complete: 58.6%; Average loss: 3.0017
Iteration: 2343; Percent complete: 58.6%; Average loss: 3.0647
Iteration: 2344; Percent complete: 58.6%; Average loss: 3.3036
Iteration: 2345; Percent complete: 58.6%; Average loss: 3.2234
Iteration: 2346; Percent complete: 58.7%; Average loss: 2.7075
Iteration: 2347; Percent complete: 58.7%; Average loss: 2.9046
Iteration: 2348; Percent complete: 58.7%; Average loss: 3.3044
Iteration: 2349; Percent complete: 58.7%; Average loss: 2.8807
Iteration: 2350; Percent complete: 58.8%; Average loss: 2.8234
Iteration: 2351; Percent complete: 58.8%; Average loss: 3.0493
Iteration: 2352; Percent complete: 58.8%; Average loss: 2.8424
Iteration: 2353; Percent complete: 58.8%; Average loss: 3.0593
Iteration: 2354; Percent complete: 58.9%; Average loss: 2.9864
Iteration: 2355; Percent complete: 58.9%; Average loss: 2.8760
Iteration: 2356; Percent complete: 58.9%; Average loss: 3.0984
Iteration: 2357; Percent complete: 58.9%; Average loss: 2.8478
Iteration: 2358; Percent complete: 59.0%; Average loss: 3.0162
Iteration: 2359; Percent complete: 59.0%; Average loss: 3.2318
Iteration: 2360; Percent complete: 59.0%; Average loss: 3.0287
Iteration: 2361; Percent complete: 59.0%; Average loss: 2.7538
Iteration: 2362; Percent complete: 59.1%; Average loss: 3.1500
Iteration: 2363; Percent complete: 59.1%; Average loss: 3.0721
Iteration: 2364; Percent complete: 59.1%; Average loss: 3.0951
Iteration: 2365; Percent complete: 59.1%; Average loss: 2.9500
Iteration: 2366; Percent complete: 59.2%; Average loss: 2.9660
Iteration: 2367; Percent complete: 59.2%; Average loss: 2.7414
Iteration: 2368; Percent complete: 59.2%; Average loss: 2.9864
Iteration: 2369; Percent complete: 59.2%; Average loss: 3.1212
Iteration: 2370; Percent complete: 59.2%; Average loss: 3.2346
Iteration: 2371; Percent complete: 59.3%; Average loss: 3.1441
Iteration: 2372; Percent complete: 59.3%; Average loss: 3.1006
Iteration: 2373; Percent complete: 59.3%; Average loss: 3.0701
Iteration: 2374; Percent complete: 59.4%; Average loss: 2.9075
Iteration: 2375; Percent complete: 59.4%; Average loss: 3.0309
Iteration: 2376; Percent complete: 59.4%; Average loss: 3.0294
Iteration: 2377; Percent complete: 59.4%; Average loss: 3.1920
Iteration: 2378; Percent complete: 59.5%; Average loss: 3.0414
Iteration: 2379; Percent complete: 59.5%; Average loss: 2.8533
Iteration: 2380; Percent complete: 59.5%; Average loss: 2.9480
Iteration: 2381; Percent complete: 59.5%; Average loss: 2.8021
Iteration: 2382; Percent complete: 59.6%; Average loss: 3.0445
Iteration: 2383; Percent complete: 59.6%; Average loss: 2.8975
Iteration: 2384; Percent complete: 59.6%; Average loss: 3.0275
Iteration: 2385; Percent complete: 59.6%; Average loss: 2.8667
Iteration: 2386; Percent complete: 59.7%; Average loss: 2.9247
Iteration: 2387; Percent complete: 59.7%; Average loss: 3.0654
Iteration: 2388; Percent complete: 59.7%; Average loss: 3.0075
Iteration: 2389; Percent complete: 59.7%; Average loss: 2.9086
Iteration: 2390; Percent complete: 59.8%; Average loss: 2.9439
Iteration: 2391; Percent complete: 59.8%; Average loss: 2.7919
Iteration: 2392; Percent complete: 59.8%; Average loss: 3.0385
Iteration: 2393; Percent complete: 59.8%; Average loss: 3.1199
Iteration: 2394; Percent complete: 59.9%; Average loss: 2.9937
Iteration: 2395; Percent complete: 59.9%; Average loss: 2.9436
Iteration: 2396; Percent complete: 59.9%; Average loss: 2.9662
Iteration: 2397; Percent complete: 59.9%; Average loss: 3.1765
Iteration: 2398; Percent complete: 60.0%; Average loss: 2.9851
Iteration: 2399; Percent complete: 60.0%; Average loss: 2.9261
Iteration: 2400; Percent complete: 60.0%; Average loss: 2.9861
Iteration: 2401; Percent complete: 60.0%; Average loss: 3.1232
Iteration: 2402; Percent complete: 60.1%; Average loss: 3.0609
Iteration: 2403; Percent complete: 60.1%; Average loss: 3.2369
Iteration: 2404; Percent complete: 60.1%; Average loss: 2.7616
Iteration: 2405; Percent complete: 60.1%; Average loss: 2.7583
Iteration: 2406; Percent complete: 60.2%; Average loss: 2.8754
Iteration: 2407; Percent complete: 60.2%; Average loss: 2.8935
Iteration: 2408; Percent complete: 60.2%; Average loss: 3.1375
Iteration: 2409; Percent complete: 60.2%; Average loss: 3.1906
Iteration: 2410; Percent complete: 60.2%; Average loss: 2.7267
Iteration: 2411; Percent complete: 60.3%; Average loss: 2.8295
Iteration: 2412; Percent complete: 60.3%; Average loss: 3.0083
Iteration: 2413; Percent complete: 60.3%; Average loss: 2.8893
Iteration: 2414; Percent complete: 60.4%; Average loss: 3.1129
Iteration: 2415; Percent complete: 60.4%; Average loss: 3.1260
Iteration: 2416; Percent complete: 60.4%; Average loss: 2.8005
Iteration: 2417; Percent complete: 60.4%; Average loss: 3.2467
Iteration: 2418; Percent complete: 60.5%; Average loss: 3.0640
Iteration: 2419; Percent complete: 60.5%; Average loss: 2.9233
Iteration: 2420; Percent complete: 60.5%; Average loss: 3.0951
Iteration: 2421; Percent complete: 60.5%; Average loss: 3.4027
Iteration: 2422; Percent complete: 60.6%; Average loss: 3.0305
Iteration: 2423; Percent complete: 60.6%; Average loss: 2.9048
Iteration: 2424; Percent complete: 60.6%; Average loss: 2.9874
Iteration: 2425; Percent complete: 60.6%; Average loss: 3.0807
Iteration: 2426; Percent complete: 60.7%; Average loss: 3.1068
Iteration: 2427; Percent complete: 60.7%; Average loss: 3.2999
Iteration: 2428; Percent complete: 60.7%; Average loss: 3.0327
Iteration: 2429; Percent complete: 60.7%; Average loss: 2.8207
Iteration: 2430; Percent complete: 60.8%; Average loss: 2.9036
Iteration: 2431; Percent complete: 60.8%; Average loss: 3.2238
Iteration: 2432; Percent complete: 60.8%; Average loss: 3.1610
Iteration: 2433; Percent complete: 60.8%; Average loss: 3.0359
Iteration: 2434; Percent complete: 60.9%; Average loss: 2.9152
Iteration: 2435; Percent complete: 60.9%; Average loss: 2.8745
Iteration: 2436; Percent complete: 60.9%; Average loss: 2.9779
Iteration: 2437; Percent complete: 60.9%; Average loss: 2.8621
Iteration: 2438; Percent complete: 61.0%; Average loss: 2.9801
Iteration: 2439; Percent complete: 61.0%; Average loss: 2.9215
Iteration: 2440; Percent complete: 61.0%; Average loss: 3.1568
Iteration: 2441; Percent complete: 61.0%; Average loss: 2.9317
Iteration: 2442; Percent complete: 61.1%; Average loss: 2.9461
Iteration: 2443; Percent complete: 61.1%; Average loss: 3.0861
Iteration: 2444; Percent complete: 61.1%; Average loss: 2.8709
Iteration: 2445; Percent complete: 61.1%; Average loss: 3.1337
Iteration: 2446; Percent complete: 61.2%; Average loss: 2.7972
Iteration: 2447; Percent complete: 61.2%; Average loss: 2.8512
Iteration: 2448; Percent complete: 61.2%; Average loss: 2.9440
Iteration: 2449; Percent complete: 61.2%; Average loss: 2.8319
Iteration: 2450; Percent complete: 61.3%; Average loss: 3.0836
Iteration: 2451; Percent complete: 61.3%; Average loss: 3.0179
Iteration: 2452; Percent complete: 61.3%; Average loss: 2.9611
Iteration: 2453; Percent complete: 61.3%; Average loss: 3.0233
Iteration: 2454; Percent complete: 61.4%; Average loss: 3.0490
Iteration: 2455; Percent complete: 61.4%; Average loss: 3.0973
Iteration: 2456; Percent complete: 61.4%; Average loss: 2.8025
Iteration: 2457; Percent complete: 61.4%; Average loss: 2.9784
Iteration: 2458; Percent complete: 61.5%; Average loss: 3.2447
Iteration: 2459; Percent complete: 61.5%; Average loss: 2.8971
Iteration: 2460; Percent complete: 61.5%; Average loss: 2.8591
Iteration: 2461; Percent complete: 61.5%; Average loss: 2.8168
Iteration: 2462; Percent complete: 61.6%; Average loss: 2.8948
Iteration: 2463; Percent complete: 61.6%; Average loss: 2.8582
Iteration: 2464; Percent complete: 61.6%; Average loss: 2.8939
Iteration: 2465; Percent complete: 61.6%; Average loss: 2.9863
Iteration: 2466; Percent complete: 61.7%; Average loss: 2.9158
Iteration: 2467; Percent complete: 61.7%; Average loss: 3.0416
Iteration: 2468; Percent complete: 61.7%; Average loss: 2.9339
Iteration: 2469; Percent complete: 61.7%; Average loss: 3.0377
Iteration: 2470; Percent complete: 61.8%; Average loss: 3.2639
Iteration: 2471; Percent complete: 61.8%; Average loss: 3.0384
Iteration: 2472; Percent complete: 61.8%; Average loss: 3.1077
Iteration: 2473; Percent complete: 61.8%; Average loss: 3.2697
Iteration: 2474; Percent complete: 61.9%; Average loss: 3.0561
Iteration: 2475; Percent complete: 61.9%; Average loss: 2.8379
Iteration: 2476; Percent complete: 61.9%; Average loss: 2.9430
Iteration: 2477; Percent complete: 61.9%; Average loss: 3.2399
Iteration: 2478; Percent complete: 62.0%; Average loss: 3.1931
Iteration: 2479; Percent complete: 62.0%; Average loss: 2.8510
Iteration: 2480; Percent complete: 62.0%; Average loss: 3.0177
Iteration: 2481; Percent complete: 62.0%; Average loss: 3.2742
Iteration: 2482; Percent complete: 62.1%; Average loss: 2.9571
Iteration: 2483; Percent complete: 62.1%; Average loss: 3.0343
Iteration: 2484; Percent complete: 62.1%; Average loss: 3.0912
Iteration: 2485; Percent complete: 62.1%; Average loss: 2.9765
Iteration: 2486; Percent complete: 62.2%; Average loss: 3.0256
Iteration: 2487; Percent complete: 62.2%; Average loss: 2.9732
Iteration: 2488; Percent complete: 62.2%; Average loss: 3.0239
Iteration: 2489; Percent complete: 62.2%; Average loss: 3.0886
Iteration: 2490; Percent complete: 62.3%; Average loss: 2.9888
Iteration: 2491; Percent complete: 62.3%; Average loss: 2.9141
Iteration: 2492; Percent complete: 62.3%; Average loss: 3.0464
Iteration: 2493; Percent complete: 62.3%; Average loss: 3.0122
Iteration: 2494; Percent complete: 62.4%; Average loss: 2.9708
Iteration: 2495; Percent complete: 62.4%; Average loss: 2.9323
Iteration: 2496; Percent complete: 62.4%; Average loss: 2.8488
Iteration: 2497; Percent complete: 62.4%; Average loss: 2.9253
Iteration: 2498; Percent complete: 62.5%; Average loss: 3.0681
Iteration: 2499; Percent complete: 62.5%; Average loss: 2.7832
Iteration: 2500; Percent complete: 62.5%; Average loss: 2.8715
Iteration: 2501; Percent complete: 62.5%; Average loss: 3.0450
Iteration: 2502; Percent complete: 62.5%; Average loss: 2.9108
Iteration: 2503; Percent complete: 62.6%; Average loss: 2.9643
Iteration: 2504; Percent complete: 62.6%; Average loss: 3.0899
Iteration: 2505; Percent complete: 62.6%; Average loss: 2.8849
Iteration: 2506; Percent complete: 62.6%; Average loss: 3.0371
Iteration: 2507; Percent complete: 62.7%; Average loss: 3.1647
Iteration: 2508; Percent complete: 62.7%; Average loss: 2.9842
Iteration: 2509; Percent complete: 62.7%; Average loss: 3.0317
Iteration: 2510; Percent complete: 62.7%; Average loss: 2.7992
Iteration: 2511; Percent complete: 62.8%; Average loss: 2.9652
Iteration: 2512; Percent complete: 62.8%; Average loss: 3.0385
Iteration: 2513; Percent complete: 62.8%; Average loss: 2.8004
Iteration: 2514; Percent complete: 62.8%; Average loss: 2.9280
Iteration: 2515; Percent complete: 62.9%; Average loss: 3.1458
Iteration: 2516; Percent complete: 62.9%; Average loss: 2.8654
Iteration: 2517; Percent complete: 62.9%; Average loss: 2.8814
Iteration: 2518; Percent complete: 62.9%; Average loss: 2.9488
Iteration: 2519; Percent complete: 63.0%; Average loss: 3.0369
Iteration: 2520; Percent complete: 63.0%; Average loss: 3.0725
Iteration: 2521; Percent complete: 63.0%; Average loss: 3.1321
Iteration: 2522; Percent complete: 63.0%; Average loss: 3.1842
Iteration: 2523; Percent complete: 63.1%; Average loss: 3.1337
Iteration: 2524; Percent complete: 63.1%; Average loss: 2.8462
Iteration: 2525; Percent complete: 63.1%; Average loss: 3.0014
Iteration: 2526; Percent complete: 63.1%; Average loss: 3.0639
Iteration: 2527; Percent complete: 63.2%; Average loss: 2.7729
Iteration: 2528; Percent complete: 63.2%; Average loss: 3.0549
Iteration: 2529; Percent complete: 63.2%; Average loss: 3.2726
Iteration: 2530; Percent complete: 63.2%; Average loss: 2.9164
Iteration: 2531; Percent complete: 63.3%; Average loss: 2.9463
Iteration: 2532; Percent complete: 63.3%; Average loss: 2.9438
Iteration: 2533; Percent complete: 63.3%; Average loss: 3.1394
Iteration: 2534; Percent complete: 63.3%; Average loss: 3.0675
Iteration: 2535; Percent complete: 63.4%; Average loss: 2.9061
Iteration: 2536; Percent complete: 63.4%; Average loss: 3.1657
Iteration: 2537; Percent complete: 63.4%; Average loss: 3.1781
Iteration: 2538; Percent complete: 63.4%; Average loss: 2.8368
Iteration: 2539; Percent complete: 63.5%; Average loss: 2.7469
Iteration: 2540; Percent complete: 63.5%; Average loss: 3.2070
Iteration: 2541; Percent complete: 63.5%; Average loss: 2.9454
Iteration: 2542; Percent complete: 63.5%; Average loss: 2.8763
Iteration: 2543; Percent complete: 63.6%; Average loss: 3.1761
Iteration: 2544; Percent complete: 63.6%; Average loss: 3.1015
Iteration: 2545; Percent complete: 63.6%; Average loss: 2.9154
Iteration: 2546; Percent complete: 63.6%; Average loss: 2.9002
Iteration: 2547; Percent complete: 63.7%; Average loss: 2.7805
Iteration: 2548; Percent complete: 63.7%; Average loss: 3.1240
Iteration: 2549; Percent complete: 63.7%; Average loss: 3.0030
Iteration: 2550; Percent complete: 63.7%; Average loss: 2.8313
Iteration: 2551; Percent complete: 63.8%; Average loss: 3.3141
Iteration: 2552; Percent complete: 63.8%; Average loss: 2.7902
Iteration: 2553; Percent complete: 63.8%; Average loss: 3.0395
Iteration: 2554; Percent complete: 63.8%; Average loss: 2.9705
Iteration: 2555; Percent complete: 63.9%; Average loss: 2.7356
Iteration: 2556; Percent complete: 63.9%; Average loss: 3.2632
Iteration: 2557; Percent complete: 63.9%; Average loss: 3.0531
Iteration: 2558; Percent complete: 63.9%; Average loss: 3.1071
Iteration: 2559; Percent complete: 64.0%; Average loss: 2.8984
Iteration: 2560; Percent complete: 64.0%; Average loss: 2.9913
Iteration: 2561; Percent complete: 64.0%; Average loss: 3.0100
Iteration: 2562; Percent complete: 64.0%; Average loss: 3.0175
Iteration: 2563; Percent complete: 64.1%; Average loss: 3.0984
Iteration: 2564; Percent complete: 64.1%; Average loss: 2.8290
Iteration: 2565; Percent complete: 64.1%; Average loss: 3.1592
Iteration: 2566; Percent complete: 64.1%; Average loss: 3.1238
Iteration: 2567; Percent complete: 64.2%; Average loss: 3.1201
Iteration: 2568; Percent complete: 64.2%; Average loss: 2.9585
Iteration: 2569; Percent complete: 64.2%; Average loss: 2.9344
Iteration: 2570; Percent complete: 64.2%; Average loss: 2.9741
Iteration: 2571; Percent complete: 64.3%; Average loss: 2.8713
Iteration: 2572; Percent complete: 64.3%; Average loss: 2.7485
Iteration: 2573; Percent complete: 64.3%; Average loss: 2.7546
Iteration: 2574; Percent complete: 64.3%; Average loss: 3.0241
Iteration: 2575; Percent complete: 64.4%; Average loss: 3.1305
Iteration: 2576; Percent complete: 64.4%; Average loss: 2.9570
Iteration: 2577; Percent complete: 64.4%; Average loss: 2.9941
Iteration: 2578; Percent complete: 64.5%; Average loss: 2.9178
Iteration: 2579; Percent complete: 64.5%; Average loss: 2.9063
Iteration: 2580; Percent complete: 64.5%; Average loss: 2.8284
Iteration: 2581; Percent complete: 64.5%; Average loss: 2.8888
Iteration: 2582; Percent complete: 64.5%; Average loss: 2.8609
Iteration: 2583; Percent complete: 64.6%; Average loss: 3.0145
Iteration: 2584; Percent complete: 64.6%; Average loss: 2.8814
Iteration: 2585; Percent complete: 64.6%; Average loss: 3.1180
Iteration: 2586; Percent complete: 64.6%; Average loss: 3.0549
Iteration: 2587; Percent complete: 64.7%; Average loss: 3.0193
Iteration: 2588; Percent complete: 64.7%; Average loss: 2.9203
Iteration: 2589; Percent complete: 64.7%; Average loss: 3.2526
Iteration: 2590; Percent complete: 64.8%; Average loss: 2.7973
Iteration: 2591; Percent complete: 64.8%; Average loss: 3.0983
Iteration: 2592; Percent complete: 64.8%; Average loss: 2.9755
Iteration: 2593; Percent complete: 64.8%; Average loss: 2.9263
Iteration: 2594; Percent complete: 64.8%; Average loss: 2.8518
Iteration: 2595; Percent complete: 64.9%; Average loss: 2.8980
Iteration: 2596; Percent complete: 64.9%; Average loss: 3.0326
Iteration: 2597; Percent complete: 64.9%; Average loss: 2.8617
Iteration: 2598; Percent complete: 65.0%; Average loss: 2.8229
Iteration: 2599; Percent complete: 65.0%; Average loss: 2.6414
Iteration: 2600; Percent complete: 65.0%; Average loss: 2.9680
Iteration: 2601; Percent complete: 65.0%; Average loss: 3.0690
Iteration: 2602; Percent complete: 65.0%; Average loss: 3.0013
Iteration: 2603; Percent complete: 65.1%; Average loss: 3.0177
Iteration: 2604; Percent complete: 65.1%; Average loss: 2.8392
Iteration: 2605; Percent complete: 65.1%; Average loss: 3.1620
Iteration: 2606; Percent complete: 65.1%; Average loss: 2.8734
Iteration: 2607; Percent complete: 65.2%; Average loss: 2.9489
Iteration: 2608; Percent complete: 65.2%; Average loss: 2.9738
Iteration: 2609; Percent complete: 65.2%; Average loss: 2.9622
Iteration: 2610; Percent complete: 65.2%; Average loss: 2.9740
Iteration: 2611; Percent complete: 65.3%; Average loss: 3.0380
Iteration: 2612; Percent complete: 65.3%; Average loss: 3.0515
Iteration: 2613; Percent complete: 65.3%; Average loss: 3.0205
Iteration: 2614; Percent complete: 65.3%; Average loss: 2.7348
Iteration: 2615; Percent complete: 65.4%; Average loss: 2.8274
Iteration: 2616; Percent complete: 65.4%; Average loss: 3.0904
Iteration: 2617; Percent complete: 65.4%; Average loss: 2.9929
Iteration: 2618; Percent complete: 65.5%; Average loss: 2.9875
Iteration: 2619; Percent complete: 65.5%; Average loss: 3.0905
Iteration: 2620; Percent complete: 65.5%; Average loss: 2.9603
Iteration: 2621; Percent complete: 65.5%; Average loss: 2.7603
Iteration: 2622; Percent complete: 65.5%; Average loss: 3.0318
Iteration: 2623; Percent complete: 65.6%; Average loss: 2.7008
Iteration: 2624; Percent complete: 65.6%; Average loss: 2.9250
Iteration: 2625; Percent complete: 65.6%; Average loss: 3.0490
Iteration: 2626; Percent complete: 65.6%; Average loss: 2.8646
Iteration: 2627; Percent complete: 65.7%; Average loss: 2.9224
Iteration: 2628; Percent complete: 65.7%; Average loss: 3.0065
Iteration: 2629; Percent complete: 65.7%; Average loss: 3.0513
Iteration: 2630; Percent complete: 65.8%; Average loss: 2.8554
Iteration: 2631; Percent complete: 65.8%; Average loss: 3.1178
Iteration: 2632; Percent complete: 65.8%; Average loss: 2.8158
Iteration: 2633; Percent complete: 65.8%; Average loss: 2.9303
Iteration: 2634; Percent complete: 65.8%; Average loss: 2.7909
Iteration: 2635; Percent complete: 65.9%; Average loss: 2.8660
Iteration: 2636; Percent complete: 65.9%; Average loss: 3.1726
Iteration: 2637; Percent complete: 65.9%; Average loss: 2.8448
Iteration: 2638; Percent complete: 66.0%; Average loss: 3.2791
Iteration: 2639; Percent complete: 66.0%; Average loss: 3.1924
Iteration: 2640; Percent complete: 66.0%; Average loss: 2.9812
Iteration: 2641; Percent complete: 66.0%; Average loss: 3.0375
Iteration: 2642; Percent complete: 66.0%; Average loss: 2.8506
Iteration: 2643; Percent complete: 66.1%; Average loss: 2.9880
Iteration: 2644; Percent complete: 66.1%; Average loss: 3.1087
Iteration: 2645; Percent complete: 66.1%; Average loss: 2.8995
Iteration: 2646; Percent complete: 66.1%; Average loss: 2.9919
Iteration: 2647; Percent complete: 66.2%; Average loss: 2.8236
Iteration: 2648; Percent complete: 66.2%; Average loss: 3.2478
Iteration: 2649; Percent complete: 66.2%; Average loss: 2.8549
Iteration: 2650; Percent complete: 66.2%; Average loss: 2.7916
Iteration: 2651; Percent complete: 66.3%; Average loss: 3.0931
Iteration: 2652; Percent complete: 66.3%; Average loss: 2.9544
Iteration: 2653; Percent complete: 66.3%; Average loss: 3.0016
Iteration: 2654; Percent complete: 66.3%; Average loss: 3.0195
Iteration: 2655; Percent complete: 66.4%; Average loss: 2.9979
Iteration: 2656; Percent complete: 66.4%; Average loss: 2.8326
Iteration: 2657; Percent complete: 66.4%; Average loss: 3.1443
Iteration: 2658; Percent complete: 66.5%; Average loss: 3.1718
Iteration: 2659; Percent complete: 66.5%; Average loss: 2.8870
Iteration: 2660; Percent complete: 66.5%; Average loss: 2.9253
Iteration: 2661; Percent complete: 66.5%; Average loss: 3.0169
Iteration: 2662; Percent complete: 66.5%; Average loss: 2.9196
Iteration: 2663; Percent complete: 66.6%; Average loss: 2.9299
Iteration: 2664; Percent complete: 66.6%; Average loss: 2.7494
Iteration: 2665; Percent complete: 66.6%; Average loss: 2.9419
Iteration: 2666; Percent complete: 66.6%; Average loss: 2.9841
Iteration: 2667; Percent complete: 66.7%; Average loss: 2.8377
Iteration: 2668; Percent complete: 66.7%; Average loss: 2.9130
Iteration: 2669; Percent complete: 66.7%; Average loss: 2.9682
Iteration: 2670; Percent complete: 66.8%; Average loss: 2.8729
Iteration: 2671; Percent complete: 66.8%; Average loss: 2.8496
Iteration: 2672; Percent complete: 66.8%; Average loss: 3.0554
Iteration: 2673; Percent complete: 66.8%; Average loss: 2.9380
Iteration: 2674; Percent complete: 66.8%; Average loss: 3.0321
Iteration: 2675; Percent complete: 66.9%; Average loss: 3.1066
Iteration: 2676; Percent complete: 66.9%; Average loss: 3.0756
Iteration: 2677; Percent complete: 66.9%; Average loss: 3.0320
Iteration: 2678; Percent complete: 67.0%; Average loss: 2.8361
Iteration: 2679; Percent complete: 67.0%; Average loss: 2.8654
Iteration: 2680; Percent complete: 67.0%; Average loss: 2.9264
Iteration: 2681; Percent complete: 67.0%; Average loss: 2.9929
Iteration: 2682; Percent complete: 67.0%; Average loss: 3.2228
Iteration: 2683; Percent complete: 67.1%; Average loss: 3.0344
Iteration: 2684; Percent complete: 67.1%; Average loss: 2.8277
Iteration: 2685; Percent complete: 67.1%; Average loss: 3.0306
Iteration: 2686; Percent complete: 67.2%; Average loss: 2.8452
Iteration: 2687; Percent complete: 67.2%; Average loss: 3.0918
Iteration: 2688; Percent complete: 67.2%; Average loss: 3.0151
Iteration: 2689; Percent complete: 67.2%; Average loss: 3.2232
Iteration: 2690; Percent complete: 67.2%; Average loss: 3.0424
Iteration: 2691; Percent complete: 67.3%; Average loss: 2.9597
Iteration: 2692; Percent complete: 67.3%; Average loss: 3.0330
Iteration: 2693; Percent complete: 67.3%; Average loss: 2.8379
Iteration: 2694; Percent complete: 67.3%; Average loss: 2.7171
Iteration: 2695; Percent complete: 67.4%; Average loss: 2.8086
Iteration: 2696; Percent complete: 67.4%; Average loss: 3.0114
Iteration: 2697; Percent complete: 67.4%; Average loss: 2.8618
Iteration: 2698; Percent complete: 67.5%; Average loss: 2.8445
Iteration: 2699; Percent complete: 67.5%; Average loss: 3.1497
Iteration: 2700; Percent complete: 67.5%; Average loss: 2.8909
Iteration: 2701; Percent complete: 67.5%; Average loss: 2.7332
Iteration: 2702; Percent complete: 67.5%; Average loss: 3.2743
Iteration: 2703; Percent complete: 67.6%; Average loss: 3.2282
Iteration: 2704; Percent complete: 67.6%; Average loss: 2.6389
Iteration: 2705; Percent complete: 67.6%; Average loss: 2.9407
Iteration: 2706; Percent complete: 67.7%; Average loss: 3.1029
Iteration: 2707; Percent complete: 67.7%; Average loss: 2.9915
Iteration: 2708; Percent complete: 67.7%; Average loss: 2.8258
Iteration: 2709; Percent complete: 67.7%; Average loss: 2.9405
Iteration: 2710; Percent complete: 67.8%; Average loss: 2.7940
Iteration: 2711; Percent complete: 67.8%; Average loss: 3.1707
Iteration: 2712; Percent complete: 67.8%; Average loss: 3.0576
Iteration: 2713; Percent complete: 67.8%; Average loss: 2.9125
Iteration: 2714; Percent complete: 67.8%; Average loss: 2.8023
Iteration: 2715; Percent complete: 67.9%; Average loss: 3.1103
Iteration: 2716; Percent complete: 67.9%; Average loss: 2.9682
Iteration: 2717; Percent complete: 67.9%; Average loss: 3.1150
Iteration: 2718; Percent complete: 68.0%; Average loss: 2.9299
Iteration: 2719; Percent complete: 68.0%; Average loss: 2.9364
Iteration: 2720; Percent complete: 68.0%; Average loss: 2.9841
Iteration: 2721; Percent complete: 68.0%; Average loss: 2.9456
Iteration: 2722; Percent complete: 68.0%; Average loss: 2.9518
Iteration: 2723; Percent complete: 68.1%; Average loss: 3.2147
Iteration: 2724; Percent complete: 68.1%; Average loss: 2.9776
Iteration: 2725; Percent complete: 68.1%; Average loss: 2.8675
Iteration: 2726; Percent complete: 68.2%; Average loss: 2.8288
Iteration: 2727; Percent complete: 68.2%; Average loss: 2.8866
Iteration: 2728; Percent complete: 68.2%; Average loss: 2.8408
Iteration: 2729; Percent complete: 68.2%; Average loss: 2.8168
Iteration: 2730; Percent complete: 68.2%; Average loss: 2.7443
Iteration: 2731; Percent complete: 68.3%; Average loss: 2.6526
Iteration: 2732; Percent complete: 68.3%; Average loss: 2.8933
Iteration: 2733; Percent complete: 68.3%; Average loss: 3.1361
Iteration: 2734; Percent complete: 68.3%; Average loss: 2.7559
Iteration: 2735; Percent complete: 68.4%; Average loss: 2.9760
Iteration: 2736; Percent complete: 68.4%; Average loss: 2.8123
Iteration: 2737; Percent complete: 68.4%; Average loss: 2.8568
Iteration: 2738; Percent complete: 68.5%; Average loss: 2.9215
Iteration: 2739; Percent complete: 68.5%; Average loss: 3.2648
Iteration: 2740; Percent complete: 68.5%; Average loss: 2.7746
Iteration: 2741; Percent complete: 68.5%; Average loss: 3.2718
Iteration: 2742; Percent complete: 68.5%; Average loss: 3.0082
Iteration: 2743; Percent complete: 68.6%; Average loss: 3.1594
Iteration: 2744; Percent complete: 68.6%; Average loss: 2.9615
Iteration: 2745; Percent complete: 68.6%; Average loss: 3.0496
Iteration: 2746; Percent complete: 68.7%; Average loss: 2.7515
Iteration: 2747; Percent complete: 68.7%; Average loss: 3.3479
Iteration: 2748; Percent complete: 68.7%; Average loss: 2.8162
Iteration: 2749; Percent complete: 68.7%; Average loss: 2.9385
Iteration: 2750; Percent complete: 68.8%; Average loss: 3.0187
Iteration: 2751; Percent complete: 68.8%; Average loss: 2.8833
Iteration: 2752; Percent complete: 68.8%; Average loss: 2.8035
Iteration: 2753; Percent complete: 68.8%; Average loss: 2.8249
Iteration: 2754; Percent complete: 68.8%; Average loss: 2.9506
Iteration: 2755; Percent complete: 68.9%; Average loss: 3.2661
Iteration: 2756; Percent complete: 68.9%; Average loss: 2.7653
Iteration: 2757; Percent complete: 68.9%; Average loss: 2.6405
Iteration: 2758; Percent complete: 69.0%; Average loss: 2.9182
Iteration: 2759; Percent complete: 69.0%; Average loss: 3.0208
Iteration: 2760; Percent complete: 69.0%; Average loss: 2.8340
Iteration: 2761; Percent complete: 69.0%; Average loss: 2.9512
Iteration: 2762; Percent complete: 69.0%; Average loss: 3.0217
Iteration: 2763; Percent complete: 69.1%; Average loss: 2.9737
Iteration: 2764; Percent complete: 69.1%; Average loss: 2.8724
Iteration: 2765; Percent complete: 69.1%; Average loss: 2.7283
Iteration: 2766; Percent complete: 69.2%; Average loss: 3.0662
Iteration: 2767; Percent complete: 69.2%; Average loss: 3.1547
Iteration: 2768; Percent complete: 69.2%; Average loss: 2.7494
Iteration: 2769; Percent complete: 69.2%; Average loss: 2.8986
Iteration: 2770; Percent complete: 69.2%; Average loss: 3.1523
Iteration: 2771; Percent complete: 69.3%; Average loss: 2.9750
Iteration: 2772; Percent complete: 69.3%; Average loss: 3.0364
Iteration: 2773; Percent complete: 69.3%; Average loss: 2.7380
Iteration: 2774; Percent complete: 69.3%; Average loss: 2.7735
Iteration: 2775; Percent complete: 69.4%; Average loss: 2.9634
Iteration: 2776; Percent complete: 69.4%; Average loss: 2.8822
Iteration: 2777; Percent complete: 69.4%; Average loss: 3.0105
Iteration: 2778; Percent complete: 69.5%; Average loss: 2.9972
Iteration: 2779; Percent complete: 69.5%; Average loss: 2.7438
Iteration: 2780; Percent complete: 69.5%; Average loss: 2.9456
Iteration: 2781; Percent complete: 69.5%; Average loss: 2.8388
Iteration: 2782; Percent complete: 69.5%; Average loss: 3.1030
Iteration: 2783; Percent complete: 69.6%; Average loss: 2.8733
Iteration: 2784; Percent complete: 69.6%; Average loss: 2.9627
Iteration: 2785; Percent complete: 69.6%; Average loss: 3.1214
Iteration: 2786; Percent complete: 69.7%; Average loss: 2.8982
Iteration: 2787; Percent complete: 69.7%; Average loss: 3.0439
Iteration: 2788; Percent complete: 69.7%; Average loss: 2.8603
Iteration: 2789; Percent complete: 69.7%; Average loss: 2.8937
Iteration: 2790; Percent complete: 69.8%; Average loss: 2.9386
Iteration: 2791; Percent complete: 69.8%; Average loss: 2.7611
Iteration: 2792; Percent complete: 69.8%; Average loss: 2.7815
Iteration: 2793; Percent complete: 69.8%; Average loss: 3.0887
Iteration: 2794; Percent complete: 69.8%; Average loss: 3.0014
Iteration: 2795; Percent complete: 69.9%; Average loss: 2.8276
Iteration: 2796; Percent complete: 69.9%; Average loss: 2.8008
Iteration: 2797; Percent complete: 69.9%; Average loss: 3.0055
Iteration: 2798; Percent complete: 70.0%; Average loss: 3.1506
Iteration: 2799; Percent complete: 70.0%; Average loss: 2.9961
Iteration: 2800; Percent complete: 70.0%; Average loss: 3.0065
Iteration: 2801; Percent complete: 70.0%; Average loss: 3.1213
Iteration: 2802; Percent complete: 70.0%; Average loss: 2.9813
Iteration: 2803; Percent complete: 70.1%; Average loss: 2.7898
Iteration: 2804; Percent complete: 70.1%; Average loss: 2.7242
Iteration: 2805; Percent complete: 70.1%; Average loss: 2.6516
Iteration: 2806; Percent complete: 70.2%; Average loss: 3.0728
Iteration: 2807; Percent complete: 70.2%; Average loss: 2.9519
Iteration: 2808; Percent complete: 70.2%; Average loss: 3.0163
Iteration: 2809; Percent complete: 70.2%; Average loss: 2.9589
Iteration: 2810; Percent complete: 70.2%; Average loss: 2.8901
Iteration: 2811; Percent complete: 70.3%; Average loss: 2.8755
Iteration: 2812; Percent complete: 70.3%; Average loss: 2.8147
Iteration: 2813; Percent complete: 70.3%; Average loss: 3.0504
Iteration: 2814; Percent complete: 70.3%; Average loss: 2.7967
Iteration: 2815; Percent complete: 70.4%; Average loss: 2.7567
Iteration: 2816; Percent complete: 70.4%; Average loss: 2.8252
Iteration: 2817; Percent complete: 70.4%; Average loss: 2.7938
Iteration: 2818; Percent complete: 70.5%; Average loss: 2.8131
Iteration: 2819; Percent complete: 70.5%; Average loss: 2.9629
Iteration: 2820; Percent complete: 70.5%; Average loss: 2.8660
Iteration: 2821; Percent complete: 70.5%; Average loss: 2.7960
Iteration: 2822; Percent complete: 70.5%; Average loss: 2.9319
Iteration: 2823; Percent complete: 70.6%; Average loss: 2.6700
Iteration: 2824; Percent complete: 70.6%; Average loss: 2.8764
Iteration: 2825; Percent complete: 70.6%; Average loss: 2.6894
Iteration: 2826; Percent complete: 70.7%; Average loss: 3.0043
Iteration: 2827; Percent complete: 70.7%; Average loss: 2.7058
Iteration: 2828; Percent complete: 70.7%; Average loss: 2.7777
Iteration: 2829; Percent complete: 70.7%; Average loss: 2.9759
Iteration: 2830; Percent complete: 70.8%; Average loss: 2.8413
Iteration: 2831; Percent complete: 70.8%; Average loss: 2.7853
Iteration: 2832; Percent complete: 70.8%; Average loss: 3.1736
Iteration: 2833; Percent complete: 70.8%; Average loss: 3.0816
Iteration: 2834; Percent complete: 70.9%; Average loss: 2.6507
Iteration: 2835; Percent complete: 70.9%; Average loss: 3.0491
Iteration: 2836; Percent complete: 70.9%; Average loss: 2.6582
Iteration: 2837; Percent complete: 70.9%; Average loss: 2.9121
Iteration: 2838; Percent complete: 71.0%; Average loss: 2.9151
Iteration: 2839; Percent complete: 71.0%; Average loss: 2.9356
Iteration: 2840; Percent complete: 71.0%; Average loss: 2.8286
Iteration: 2841; Percent complete: 71.0%; Average loss: 2.8881
Iteration: 2842; Percent complete: 71.0%; Average loss: 2.9509
Iteration: 2843; Percent complete: 71.1%; Average loss: 2.5810
Iteration: 2844; Percent complete: 71.1%; Average loss: 2.7565
Iteration: 2845; Percent complete: 71.1%; Average loss: 2.7776
Iteration: 2846; Percent complete: 71.2%; Average loss: 2.8107
Iteration: 2847; Percent complete: 71.2%; Average loss: 2.8167
Iteration: 2848; Percent complete: 71.2%; Average loss: 2.7951
Iteration: 2849; Percent complete: 71.2%; Average loss: 3.0590
Iteration: 2850; Percent complete: 71.2%; Average loss: 3.0458
Iteration: 2851; Percent complete: 71.3%; Average loss: 2.9645
Iteration: 2852; Percent complete: 71.3%; Average loss: 2.8046
Iteration: 2853; Percent complete: 71.3%; Average loss: 2.8812
Iteration: 2854; Percent complete: 71.4%; Average loss: 2.8768
Iteration: 2855; Percent complete: 71.4%; Average loss: 2.6910
Iteration: 2856; Percent complete: 71.4%; Average loss: 2.9706
Iteration: 2857; Percent complete: 71.4%; Average loss: 3.1564
Iteration: 2858; Percent complete: 71.5%; Average loss: 2.9692
Iteration: 2859; Percent complete: 71.5%; Average loss: 2.8607
Iteration: 2860; Percent complete: 71.5%; Average loss: 2.6504
Iteration: 2861; Percent complete: 71.5%; Average loss: 2.7879
Iteration: 2862; Percent complete: 71.5%; Average loss: 3.1938
Iteration: 2863; Percent complete: 71.6%; Average loss: 2.5741
Iteration: 2864; Percent complete: 71.6%; Average loss: 2.7467
Iteration: 2865; Percent complete: 71.6%; Average loss: 2.8547
Iteration: 2866; Percent complete: 71.7%; Average loss: 2.8256
Iteration: 2867; Percent complete: 71.7%; Average loss: 2.8162
Iteration: 2868; Percent complete: 71.7%; Average loss: 2.6445
Iteration: 2869; Percent complete: 71.7%; Average loss: 2.8304
Iteration: 2870; Percent complete: 71.8%; Average loss: 3.0075
Iteration: 2871; Percent complete: 71.8%; Average loss: 2.7961
Iteration: 2872; Percent complete: 71.8%; Average loss: 2.6466
Iteration: 2873; Percent complete: 71.8%; Average loss: 3.0782
Iteration: 2874; Percent complete: 71.9%; Average loss: 2.9572
Iteration: 2875; Percent complete: 71.9%; Average loss: 2.9112
Iteration: 2876; Percent complete: 71.9%; Average loss: 3.1231
Iteration: 2877; Percent complete: 71.9%; Average loss: 2.9373
Iteration: 2878; Percent complete: 72.0%; Average loss: 3.1215
Iteration: 2879; Percent complete: 72.0%; Average loss: 2.8912
Iteration: 2880; Percent complete: 72.0%; Average loss: 3.0560
Iteration: 2881; Percent complete: 72.0%; Average loss: 2.9098
Iteration: 2882; Percent complete: 72.0%; Average loss: 3.0696
Iteration: 2883; Percent complete: 72.1%; Average loss: 2.8604
Iteration: 2884; Percent complete: 72.1%; Average loss: 2.9647
Iteration: 2885; Percent complete: 72.1%; Average loss: 2.8408
Iteration: 2886; Percent complete: 72.2%; Average loss: 2.8023
Iteration: 2887; Percent complete: 72.2%; Average loss: 2.9699
Iteration: 2888; Percent complete: 72.2%; Average loss: 2.8899
Iteration: 2889; Percent complete: 72.2%; Average loss: 2.7509
Iteration: 2890; Percent complete: 72.2%; Average loss: 3.0695
Iteration: 2891; Percent complete: 72.3%; Average loss: 2.8292
Iteration: 2892; Percent complete: 72.3%; Average loss: 3.1199
Iteration: 2893; Percent complete: 72.3%; Average loss: 3.0404
Iteration: 2894; Percent complete: 72.4%; Average loss: 3.0264
Iteration: 2895; Percent complete: 72.4%; Average loss: 2.9741
Iteration: 2896; Percent complete: 72.4%; Average loss: 2.8900
Iteration: 2897; Percent complete: 72.4%; Average loss: 2.6038
Iteration: 2898; Percent complete: 72.5%; Average loss: 3.1104
Iteration: 2899; Percent complete: 72.5%; Average loss: 2.9016
Iteration: 2900; Percent complete: 72.5%; Average loss: 3.0194
Iteration: 2901; Percent complete: 72.5%; Average loss: 3.0457
Iteration: 2902; Percent complete: 72.5%; Average loss: 2.8499
Iteration: 2903; Percent complete: 72.6%; Average loss: 2.6927
Iteration: 2904; Percent complete: 72.6%; Average loss: 2.7626
Iteration: 2905; Percent complete: 72.6%; Average loss: 2.7896
Iteration: 2906; Percent complete: 72.7%; Average loss: 3.0149
Iteration: 2907; Percent complete: 72.7%; Average loss: 2.9876
Iteration: 2908; Percent complete: 72.7%; Average loss: 2.9340
Iteration: 2909; Percent complete: 72.7%; Average loss: 2.7335
Iteration: 2910; Percent complete: 72.8%; Average loss: 2.8175
Iteration: 2911; Percent complete: 72.8%; Average loss: 3.1439
Iteration: 2912; Percent complete: 72.8%; Average loss: 2.8987
Iteration: 2913; Percent complete: 72.8%; Average loss: 2.9784
Iteration: 2914; Percent complete: 72.9%; Average loss: 3.1104
Iteration: 2915; Percent complete: 72.9%; Average loss: 2.8192
Iteration: 2916; Percent complete: 72.9%; Average loss: 2.9292
Iteration: 2917; Percent complete: 72.9%; Average loss: 2.9930
Iteration: 2918; Percent complete: 73.0%; Average loss: 2.9864
Iteration: 2919; Percent complete: 73.0%; Average loss: 2.6454
Iteration: 2920; Percent complete: 73.0%; Average loss: 2.8230
Iteration: 2921; Percent complete: 73.0%; Average loss: 3.0225
Iteration: 2922; Percent complete: 73.0%; Average loss: 2.9086
Iteration: 2923; Percent complete: 73.1%; Average loss: 2.9529
Iteration: 2924; Percent complete: 73.1%; Average loss: 2.7299
Iteration: 2925; Percent complete: 73.1%; Average loss: 2.9488
Iteration: 2926; Percent complete: 73.2%; Average loss: 2.7491
Iteration: 2927; Percent complete: 73.2%; Average loss: 2.9054
Iteration: 2928; Percent complete: 73.2%; Average loss: 3.2599
Iteration: 2929; Percent complete: 73.2%; Average loss: 3.0000
Iteration: 2930; Percent complete: 73.2%; Average loss: 2.9290
Iteration: 2931; Percent complete: 73.3%; Average loss: 2.8529
Iteration: 2932; Percent complete: 73.3%; Average loss: 3.0988
Iteration: 2933; Percent complete: 73.3%; Average loss: 3.1299
Iteration: 2934; Percent complete: 73.4%; Average loss: 2.8664
Iteration: 2935; Percent complete: 73.4%; Average loss: 2.8559
Iteration: 2936; Percent complete: 73.4%; Average loss: 2.8654
Iteration: 2937; Percent complete: 73.4%; Average loss: 2.7880
Iteration: 2938; Percent complete: 73.5%; Average loss: 3.0441
Iteration: 2939; Percent complete: 73.5%; Average loss: 2.6604
Iteration: 2940; Percent complete: 73.5%; Average loss: 2.8349
Iteration: 2941; Percent complete: 73.5%; Average loss: 3.0409
Iteration: 2942; Percent complete: 73.6%; Average loss: 2.8176
Iteration: 2943; Percent complete: 73.6%; Average loss: 3.0730
Iteration: 2944; Percent complete: 73.6%; Average loss: 2.9639
Iteration: 2945; Percent complete: 73.6%; Average loss: 2.9665
Iteration: 2946; Percent complete: 73.7%; Average loss: 2.7805
Iteration: 2947; Percent complete: 73.7%; Average loss: 2.9751
Iteration: 2948; Percent complete: 73.7%; Average loss: 2.8000
Iteration: 2949; Percent complete: 73.7%; Average loss: 2.7891
Iteration: 2950; Percent complete: 73.8%; Average loss: 2.9583
Iteration: 2951; Percent complete: 73.8%; Average loss: 2.6572
Iteration: 2952; Percent complete: 73.8%; Average loss: 2.6221
Iteration: 2953; Percent complete: 73.8%; Average loss: 2.8492
Iteration: 2954; Percent complete: 73.9%; Average loss: 2.8088
Iteration: 2955; Percent complete: 73.9%; Average loss: 2.9195
Iteration: 2956; Percent complete: 73.9%; Average loss: 2.8466
Iteration: 2957; Percent complete: 73.9%; Average loss: 2.7799
Iteration: 2958; Percent complete: 74.0%; Average loss: 2.7785
Iteration: 2959; Percent complete: 74.0%; Average loss: 2.6991
Iteration: 2960; Percent complete: 74.0%; Average loss: 2.8399
Iteration: 2961; Percent complete: 74.0%; Average loss: 2.8793
Iteration: 2962; Percent complete: 74.1%; Average loss: 3.0733
Iteration: 2963; Percent complete: 74.1%; Average loss: 2.8135
Iteration: 2964; Percent complete: 74.1%; Average loss: 3.0782
Iteration: 2965; Percent complete: 74.1%; Average loss: 3.1157
Iteration: 2966; Percent complete: 74.2%; Average loss: 3.0654
Iteration: 2967; Percent complete: 74.2%; Average loss: 2.9643
Iteration: 2968; Percent complete: 74.2%; Average loss: 2.7630
Iteration: 2969; Percent complete: 74.2%; Average loss: 3.0331
Iteration: 2970; Percent complete: 74.2%; Average loss: 2.7721
Iteration: 2971; Percent complete: 74.3%; Average loss: 2.9151
Iteration: 2972; Percent complete: 74.3%; Average loss: 2.7340
Iteration: 2973; Percent complete: 74.3%; Average loss: 2.9196
Iteration: 2974; Percent complete: 74.4%; Average loss: 2.7246
Iteration: 2975; Percent complete: 74.4%; Average loss: 2.9367
Iteration: 2976; Percent complete: 74.4%; Average loss: 2.8102
Iteration: 2977; Percent complete: 74.4%; Average loss: 2.8156
Iteration: 2978; Percent complete: 74.5%; Average loss: 2.6986
Iteration: 2979; Percent complete: 74.5%; Average loss: 3.0047
Iteration: 2980; Percent complete: 74.5%; Average loss: 2.8991
Iteration: 2981; Percent complete: 74.5%; Average loss: 2.7441
Iteration: 2982; Percent complete: 74.6%; Average loss: 2.7808
Iteration: 2983; Percent complete: 74.6%; Average loss: 2.6880
Iteration: 2984; Percent complete: 74.6%; Average loss: 2.8648
Iteration: 2985; Percent complete: 74.6%; Average loss: 2.9676
Iteration: 2986; Percent complete: 74.7%; Average loss: 2.8475
Iteration: 2987; Percent complete: 74.7%; Average loss: 2.8878
Iteration: 2988; Percent complete: 74.7%; Average loss: 2.8059
Iteration: 2989; Percent complete: 74.7%; Average loss: 2.8022
Iteration: 2990; Percent complete: 74.8%; Average loss: 2.9271
Iteration: 2991; Percent complete: 74.8%; Average loss: 2.9626
Iteration: 2992; Percent complete: 74.8%; Average loss: 2.8918
Iteration: 2993; Percent complete: 74.8%; Average loss: 2.7667
Iteration: 2994; Percent complete: 74.9%; Average loss: 3.0281
Iteration: 2995; Percent complete: 74.9%; Average loss: 2.9550
Iteration: 2996; Percent complete: 74.9%; Average loss: 2.7574
Iteration: 2997; Percent complete: 74.9%; Average loss: 2.7256
Iteration: 2998; Percent complete: 75.0%; Average loss: 2.8683
Iteration: 2999; Percent complete: 75.0%; Average loss: 2.8786
Iteration: 3000; Percent complete: 75.0%; Average loss: 2.8660
Iteration: 3001; Percent complete: 75.0%; Average loss: 2.7284
Iteration: 3002; Percent complete: 75.0%; Average loss: 2.7878
Iteration: 3003; Percent complete: 75.1%; Average loss: 2.9030
Iteration: 3004; Percent complete: 75.1%; Average loss: 2.9096
Iteration: 3005; Percent complete: 75.1%; Average loss: 2.7714
Iteration: 3006; Percent complete: 75.1%; Average loss: 2.8542
Iteration: 3007; Percent complete: 75.2%; Average loss: 2.9423
Iteration: 3008; Percent complete: 75.2%; Average loss: 2.8661
Iteration: 3009; Percent complete: 75.2%; Average loss: 2.8734
Iteration: 3010; Percent complete: 75.2%; Average loss: 2.6916
Iteration: 3011; Percent complete: 75.3%; Average loss: 3.1328
Iteration: 3012; Percent complete: 75.3%; Average loss: 2.8939
Iteration: 3013; Percent complete: 75.3%; Average loss: 3.0649
Iteration: 3014; Percent complete: 75.3%; Average loss: 3.0677
Iteration: 3015; Percent complete: 75.4%; Average loss: 2.7846
Iteration: 3016; Percent complete: 75.4%; Average loss: 2.9783
Iteration: 3017; Percent complete: 75.4%; Average loss: 2.9391
Iteration: 3018; Percent complete: 75.4%; Average loss: 2.4462
Iteration: 3019; Percent complete: 75.5%; Average loss: 2.7641
Iteration: 3020; Percent complete: 75.5%; Average loss: 2.9491
Iteration: 3021; Percent complete: 75.5%; Average loss: 3.0156
Iteration: 3022; Percent complete: 75.5%; Average loss: 2.6814
Iteration: 3023; Percent complete: 75.6%; Average loss: 2.8038
Iteration: 3024; Percent complete: 75.6%; Average loss: 2.7470
Iteration: 3025; Percent complete: 75.6%; Average loss: 2.8319
Iteration: 3026; Percent complete: 75.6%; Average loss: 2.8545
Iteration: 3027; Percent complete: 75.7%; Average loss: 2.8310
Iteration: 3028; Percent complete: 75.7%; Average loss: 2.9625
Iteration: 3029; Percent complete: 75.7%; Average loss: 2.5480
Iteration: 3030; Percent complete: 75.8%; Average loss: 3.0777
Iteration: 3031; Percent complete: 75.8%; Average loss: 2.9779
Iteration: 3032; Percent complete: 75.8%; Average loss: 2.7116
Iteration: 3033; Percent complete: 75.8%; Average loss: 2.9719
Iteration: 3034; Percent complete: 75.8%; Average loss: 2.8230
Iteration: 3035; Percent complete: 75.9%; Average loss: 2.7200
Iteration: 3036; Percent complete: 75.9%; Average loss: 2.9900
Iteration: 3037; Percent complete: 75.9%; Average loss: 2.7683
Iteration: 3038; Percent complete: 75.9%; Average loss: 2.9495
Iteration: 3039; Percent complete: 76.0%; Average loss: 2.7560
Iteration: 3040; Percent complete: 76.0%; Average loss: 2.9140
Iteration: 3041; Percent complete: 76.0%; Average loss: 2.9743
Iteration: 3042; Percent complete: 76.0%; Average loss: 2.9181
Iteration: 3043; Percent complete: 76.1%; Average loss: 2.7850
Iteration: 3044; Percent complete: 76.1%; Average loss: 2.9136
Iteration: 3045; Percent complete: 76.1%; Average loss: 2.9894
Iteration: 3046; Percent complete: 76.1%; Average loss: 3.0439
Iteration: 3047; Percent complete: 76.2%; Average loss: 2.9099
Iteration: 3048; Percent complete: 76.2%; Average loss: 2.8954
Iteration: 3049; Percent complete: 76.2%; Average loss: 2.8565
Iteration: 3050; Percent complete: 76.2%; Average loss: 2.8707
Iteration: 3051; Percent complete: 76.3%; Average loss: 2.6346
Iteration: 3052; Percent complete: 76.3%; Average loss: 3.0119
Iteration: 3053; Percent complete: 76.3%; Average loss: 2.8128
Iteration: 3054; Percent complete: 76.3%; Average loss: 2.8710
Iteration: 3055; Percent complete: 76.4%; Average loss: 2.9352
Iteration: 3056; Percent complete: 76.4%; Average loss: 2.6098
Iteration: 3057; Percent complete: 76.4%; Average loss: 2.7163
Iteration: 3058; Percent complete: 76.4%; Average loss: 2.9396
Iteration: 3059; Percent complete: 76.5%; Average loss: 2.9285
Iteration: 3060; Percent complete: 76.5%; Average loss: 2.6758
Iteration: 3061; Percent complete: 76.5%; Average loss: 2.7809
Iteration: 3062; Percent complete: 76.5%; Average loss: 2.6872
Iteration: 3063; Percent complete: 76.6%; Average loss: 2.9680
Iteration: 3064; Percent complete: 76.6%; Average loss: 2.6993
Iteration: 3065; Percent complete: 76.6%; Average loss: 3.1684
Iteration: 3066; Percent complete: 76.6%; Average loss: 2.7985
Iteration: 3067; Percent complete: 76.7%; Average loss: 2.8227
Iteration: 3068; Percent complete: 76.7%; Average loss: 2.9221
Iteration: 3069; Percent complete: 76.7%; Average loss: 2.9861
Iteration: 3070; Percent complete: 76.8%; Average loss: 2.8992
Iteration: 3071; Percent complete: 76.8%; Average loss: 2.9716
Iteration: 3072; Percent complete: 76.8%; Average loss: 2.8103
Iteration: 3073; Percent complete: 76.8%; Average loss: 2.9729
Iteration: 3074; Percent complete: 76.8%; Average loss: 2.5817
Iteration: 3075; Percent complete: 76.9%; Average loss: 3.1386
Iteration: 3076; Percent complete: 76.9%; Average loss: 2.6488
Iteration: 3077; Percent complete: 76.9%; Average loss: 2.7730
Iteration: 3078; Percent complete: 77.0%; Average loss: 2.9037
Iteration: 3079; Percent complete: 77.0%; Average loss: 2.9510
Iteration: 3080; Percent complete: 77.0%; Average loss: 2.8846
Iteration: 3081; Percent complete: 77.0%; Average loss: 2.8219
Iteration: 3082; Percent complete: 77.0%; Average loss: 2.8143
Iteration: 3083; Percent complete: 77.1%; Average loss: 2.6795
Iteration: 3084; Percent complete: 77.1%; Average loss: 2.6040
Iteration: 3085; Percent complete: 77.1%; Average loss: 2.6131
Iteration: 3086; Percent complete: 77.1%; Average loss: 2.9391
Iteration: 3087; Percent complete: 77.2%; Average loss: 3.0224
Iteration: 3088; Percent complete: 77.2%; Average loss: 2.9159
Iteration: 3089; Percent complete: 77.2%; Average loss: 3.1732
Iteration: 3090; Percent complete: 77.2%; Average loss: 2.7634
Iteration: 3091; Percent complete: 77.3%; Average loss: 2.8343
Iteration: 3092; Percent complete: 77.3%; Average loss: 2.7162
Iteration: 3093; Percent complete: 77.3%; Average loss: 2.8631
Iteration: 3094; Percent complete: 77.3%; Average loss: 2.9865
Iteration: 3095; Percent complete: 77.4%; Average loss: 2.4542
Iteration: 3096; Percent complete: 77.4%; Average loss: 2.8780
Iteration: 3097; Percent complete: 77.4%; Average loss: 2.8969
Iteration: 3098; Percent complete: 77.5%; Average loss: 3.0746
Iteration: 3099; Percent complete: 77.5%; Average loss: 2.7979
Iteration: 3100; Percent complete: 77.5%; Average loss: 2.7735
Iteration: 3101; Percent complete: 77.5%; Average loss: 2.9612
Iteration: 3102; Percent complete: 77.5%; Average loss: 2.6964
Iteration: 3103; Percent complete: 77.6%; Average loss: 2.8639
Iteration: 3104; Percent complete: 77.6%; Average loss: 2.8788
Iteration: 3105; Percent complete: 77.6%; Average loss: 2.9585
Iteration: 3106; Percent complete: 77.6%; Average loss: 2.6613
Iteration: 3107; Percent complete: 77.7%; Average loss: 2.6155
Iteration: 3108; Percent complete: 77.7%; Average loss: 2.8788
Iteration: 3109; Percent complete: 77.7%; Average loss: 2.8288
Iteration: 3110; Percent complete: 77.8%; Average loss: 2.9695
Iteration: 3111; Percent complete: 77.8%; Average loss: 2.8063
Iteration: 3112; Percent complete: 77.8%; Average loss: 2.9602
Iteration: 3113; Percent complete: 77.8%; Average loss: 2.9596
Iteration: 3114; Percent complete: 77.8%; Average loss: 2.6918
Iteration: 3115; Percent complete: 77.9%; Average loss: 2.6063
Iteration: 3116; Percent complete: 77.9%; Average loss: 2.8094
Iteration: 3117; Percent complete: 77.9%; Average loss: 3.0742
Iteration: 3118; Percent complete: 78.0%; Average loss: 2.7664
Iteration: 3119; Percent complete: 78.0%; Average loss: 2.9759
Iteration: 3120; Percent complete: 78.0%; Average loss: 2.7745
Iteration: 3121; Percent complete: 78.0%; Average loss: 2.8831
Iteration: 3122; Percent complete: 78.0%; Average loss: 2.9238
Iteration: 3123; Percent complete: 78.1%; Average loss: 3.0202
Iteration: 3124; Percent complete: 78.1%; Average loss: 2.9664
Iteration: 3125; Percent complete: 78.1%; Average loss: 2.9893
Iteration: 3126; Percent complete: 78.1%; Average loss: 2.7833
Iteration: 3127; Percent complete: 78.2%; Average loss: 2.7254
Iteration: 3128; Percent complete: 78.2%; Average loss: 2.7047
Iteration: 3129; Percent complete: 78.2%; Average loss: 2.8485
Iteration: 3130; Percent complete: 78.2%; Average loss: 2.8778
Iteration: 3131; Percent complete: 78.3%; Average loss: 2.6820
Iteration: 3132; Percent complete: 78.3%; Average loss: 2.8719
Iteration: 3133; Percent complete: 78.3%; Average loss: 2.9471
Iteration: 3134; Percent complete: 78.3%; Average loss: 3.0049
Iteration: 3135; Percent complete: 78.4%; Average loss: 2.7481
Iteration: 3136; Percent complete: 78.4%; Average loss: 2.6723
Iteration: 3137; Percent complete: 78.4%; Average loss: 2.7954
Iteration: 3138; Percent complete: 78.5%; Average loss: 2.6773
Iteration: 3139; Percent complete: 78.5%; Average loss: 2.8710
Iteration: 3140; Percent complete: 78.5%; Average loss: 3.0994
Iteration: 3141; Percent complete: 78.5%; Average loss: 2.9561
Iteration: 3142; Percent complete: 78.5%; Average loss: 2.8988
Iteration: 3143; Percent complete: 78.6%; Average loss: 2.9474
Iteration: 3144; Percent complete: 78.6%; Average loss: 2.8829
Iteration: 3145; Percent complete: 78.6%; Average loss: 2.8460
Iteration: 3146; Percent complete: 78.6%; Average loss: 2.7989
Iteration: 3147; Percent complete: 78.7%; Average loss: 3.0243
Iteration: 3148; Percent complete: 78.7%; Average loss: 2.7625
Iteration: 3149; Percent complete: 78.7%; Average loss: 2.7797
Iteration: 3150; Percent complete: 78.8%; Average loss: 2.7505
Iteration: 3151; Percent complete: 78.8%; Average loss: 2.8664
Iteration: 3152; Percent complete: 78.8%; Average loss: 2.9901
Iteration: 3153; Percent complete: 78.8%; Average loss: 2.8944
Iteration: 3154; Percent complete: 78.8%; Average loss: 2.9198
Iteration: 3155; Percent complete: 78.9%; Average loss: 2.9553
Iteration: 3156; Percent complete: 78.9%; Average loss: 2.9632
Iteration: 3157; Percent complete: 78.9%; Average loss: 2.6734
Iteration: 3158; Percent complete: 79.0%; Average loss: 3.0616
Iteration: 3159; Percent complete: 79.0%; Average loss: 2.7618
Iteration: 3160; Percent complete: 79.0%; Average loss: 2.8259
Iteration: 3161; Percent complete: 79.0%; Average loss: 2.6977
Iteration: 3162; Percent complete: 79.0%; Average loss: 2.8469
Iteration: 3163; Percent complete: 79.1%; Average loss: 2.8519
Iteration: 3164; Percent complete: 79.1%; Average loss: 2.7574
Iteration: 3165; Percent complete: 79.1%; Average loss: 2.7939
Iteration: 3166; Percent complete: 79.1%; Average loss: 2.7051
Iteration: 3167; Percent complete: 79.2%; Average loss: 2.5123
Iteration: 3168; Percent complete: 79.2%; Average loss: 2.7841
Iteration: 3169; Percent complete: 79.2%; Average loss: 2.7374
Iteration: 3170; Percent complete: 79.2%; Average loss: 2.8603
Iteration: 3171; Percent complete: 79.3%; Average loss: 2.9687
Iteration: 3172; Percent complete: 79.3%; Average loss: 2.8836
Iteration: 3173; Percent complete: 79.3%; Average loss: 2.8438
Iteration: 3174; Percent complete: 79.3%; Average loss: 2.8248
Iteration: 3175; Percent complete: 79.4%; Average loss: 2.9028
Iteration: 3176; Percent complete: 79.4%; Average loss: 2.6294
Iteration: 3177; Percent complete: 79.4%; Average loss: 2.6121
Iteration: 3178; Percent complete: 79.5%; Average loss: 2.8041
Iteration: 3179; Percent complete: 79.5%; Average loss: 2.7732
Iteration: 3180; Percent complete: 79.5%; Average loss: 2.8355
Iteration: 3181; Percent complete: 79.5%; Average loss: 2.5950
Iteration: 3182; Percent complete: 79.5%; Average loss: 2.8513
Iteration: 3183; Percent complete: 79.6%; Average loss: 2.8277
Iteration: 3184; Percent complete: 79.6%; Average loss: 2.8895
Iteration: 3185; Percent complete: 79.6%; Average loss: 2.9913
Iteration: 3186; Percent complete: 79.7%; Average loss: 2.7981
Iteration: 3187; Percent complete: 79.7%; Average loss: 2.8441
Iteration: 3188; Percent complete: 79.7%; Average loss: 2.8503
Iteration: 3189; Percent complete: 79.7%; Average loss: 2.7479
Iteration: 3190; Percent complete: 79.8%; Average loss: 2.9700
Iteration: 3191; Percent complete: 79.8%; Average loss: 2.6950
Iteration: 3192; Percent complete: 79.8%; Average loss: 2.7678
Iteration: 3193; Percent complete: 79.8%; Average loss: 2.8524
Iteration: 3194; Percent complete: 79.8%; Average loss: 3.0515
Iteration: 3195; Percent complete: 79.9%; Average loss: 2.7007
Iteration: 3196; Percent complete: 79.9%; Average loss: 2.9352
Iteration: 3197; Percent complete: 79.9%; Average loss: 2.8417
Iteration: 3198; Percent complete: 80.0%; Average loss: 2.7072
Iteration: 3199; Percent complete: 80.0%; Average loss: 2.8420
Iteration: 3200; Percent complete: 80.0%; Average loss: 2.8674
Iteration: 3201; Percent complete: 80.0%; Average loss: 3.1892
Iteration: 3202; Percent complete: 80.0%; Average loss: 2.4289
Iteration: 3203; Percent complete: 80.1%; Average loss: 2.9772
Iteration: 3204; Percent complete: 80.1%; Average loss: 2.7571
Iteration: 3205; Percent complete: 80.1%; Average loss: 2.6203
Iteration: 3206; Percent complete: 80.2%; Average loss: 2.9890
Iteration: 3207; Percent complete: 80.2%; Average loss: 2.8670
Iteration: 3208; Percent complete: 80.2%; Average loss: 2.8168
Iteration: 3209; Percent complete: 80.2%; Average loss: 2.9105
Iteration: 3210; Percent complete: 80.2%; Average loss: 2.8197
Iteration: 3211; Percent complete: 80.3%; Average loss: 2.6331
Iteration: 3212; Percent complete: 80.3%; Average loss: 3.0290
Iteration: 3213; Percent complete: 80.3%; Average loss: 3.0913
Iteration: 3214; Percent complete: 80.3%; Average loss: 2.8203
Iteration: 3215; Percent complete: 80.4%; Average loss: 2.7016
Iteration: 3216; Percent complete: 80.4%; Average loss: 2.8070
Iteration: 3217; Percent complete: 80.4%; Average loss: 2.9093
Iteration: 3218; Percent complete: 80.5%; Average loss: 2.9426
Iteration: 3219; Percent complete: 80.5%; Average loss: 2.5730
Iteration: 3220; Percent complete: 80.5%; Average loss: 2.7584
Iteration: 3221; Percent complete: 80.5%; Average loss: 2.8234
Iteration: 3222; Percent complete: 80.5%; Average loss: 2.8280
Iteration: 3223; Percent complete: 80.6%; Average loss: 2.6794
Iteration: 3224; Percent complete: 80.6%; Average loss: 2.8971
Iteration: 3225; Percent complete: 80.6%; Average loss: 2.5182
Iteration: 3226; Percent complete: 80.7%; Average loss: 2.8160
Iteration: 3227; Percent complete: 80.7%; Average loss: 2.6571
Iteration: 3228; Percent complete: 80.7%; Average loss: 2.7915
Iteration: 3229; Percent complete: 80.7%; Average loss: 3.0486
Iteration: 3230; Percent complete: 80.8%; Average loss: 2.7986
Iteration: 3231; Percent complete: 80.8%; Average loss: 2.8247
Iteration: 3232; Percent complete: 80.8%; Average loss: 2.9859
Iteration: 3233; Percent complete: 80.8%; Average loss: 2.6955
Iteration: 3234; Percent complete: 80.8%; Average loss: 2.7260
Iteration: 3235; Percent complete: 80.9%; Average loss: 2.8349
Iteration: 3236; Percent complete: 80.9%; Average loss: 2.7593
Iteration: 3237; Percent complete: 80.9%; Average loss: 2.9380
Iteration: 3238; Percent complete: 81.0%; Average loss: 2.7825
Iteration: 3239; Percent complete: 81.0%; Average loss: 2.6686
Iteration: 3240; Percent complete: 81.0%; Average loss: 2.5404
Iteration: 3241; Percent complete: 81.0%; Average loss: 2.8765
Iteration: 3242; Percent complete: 81.0%; Average loss: 2.6742
Iteration: 3243; Percent complete: 81.1%; Average loss: 2.7210
Iteration: 3244; Percent complete: 81.1%; Average loss: 2.8333
Iteration: 3245; Percent complete: 81.1%; Average loss: 2.7739
Iteration: 3246; Percent complete: 81.2%; Average loss: 2.7323
Iteration: 3247; Percent complete: 81.2%; Average loss: 3.1470
Iteration: 3248; Percent complete: 81.2%; Average loss: 2.9450
Iteration: 3249; Percent complete: 81.2%; Average loss: 2.7036
Iteration: 3250; Percent complete: 81.2%; Average loss: 2.7075
Iteration: 3251; Percent complete: 81.3%; Average loss: 2.7517
Iteration: 3252; Percent complete: 81.3%; Average loss: 2.3869
Iteration: 3253; Percent complete: 81.3%; Average loss: 3.0827
Iteration: 3254; Percent complete: 81.3%; Average loss: 2.9837
Iteration: 3255; Percent complete: 81.4%; Average loss: 2.9710
Iteration: 3256; Percent complete: 81.4%; Average loss: 2.6789
Iteration: 3257; Percent complete: 81.4%; Average loss: 2.8450
Iteration: 3258; Percent complete: 81.5%; Average loss: 2.8875
Iteration: 3259; Percent complete: 81.5%; Average loss: 2.7825
Iteration: 3260; Percent complete: 81.5%; Average loss: 2.6028
Iteration: 3261; Percent complete: 81.5%; Average loss: 3.0198
Iteration: 3262; Percent complete: 81.5%; Average loss: 2.8607
Iteration: 3263; Percent complete: 81.6%; Average loss: 2.8137
Iteration: 3264; Percent complete: 81.6%; Average loss: 3.0101
Iteration: 3265; Percent complete: 81.6%; Average loss: 2.8307
Iteration: 3266; Percent complete: 81.7%; Average loss: 2.9138
Iteration: 3267; Percent complete: 81.7%; Average loss: 2.7983
Iteration: 3268; Percent complete: 81.7%; Average loss: 2.6847
Iteration: 3269; Percent complete: 81.7%; Average loss: 2.9225
Iteration: 3270; Percent complete: 81.8%; Average loss: 2.8390
Iteration: 3271; Percent complete: 81.8%; Average loss: 2.7793
Iteration: 3272; Percent complete: 81.8%; Average loss: 2.7659
Iteration: 3273; Percent complete: 81.8%; Average loss: 2.8979
Iteration: 3274; Percent complete: 81.8%; Average loss: 2.5170
Iteration: 3275; Percent complete: 81.9%; Average loss: 2.6929
Iteration: 3276; Percent complete: 81.9%; Average loss: 2.7807
Iteration: 3277; Percent complete: 81.9%; Average loss: 2.6119
Iteration: 3278; Percent complete: 82.0%; Average loss: 2.6734
Iteration: 3279; Percent complete: 82.0%; Average loss: 2.7308
Iteration: 3280; Percent complete: 82.0%; Average loss: 2.7371
Iteration: 3281; Percent complete: 82.0%; Average loss: 2.7730
Iteration: 3282; Percent complete: 82.0%; Average loss: 2.9093
Iteration: 3283; Percent complete: 82.1%; Average loss: 2.6939
Iteration: 3284; Percent complete: 82.1%; Average loss: 2.8230
Iteration: 3285; Percent complete: 82.1%; Average loss: 3.0503
Iteration: 3286; Percent complete: 82.2%; Average loss: 2.9990
Iteration: 3287; Percent complete: 82.2%; Average loss: 2.7982
Iteration: 3288; Percent complete: 82.2%; Average loss: 3.3228
Iteration: 3289; Percent complete: 82.2%; Average loss: 2.7910
Iteration: 3290; Percent complete: 82.2%; Average loss: 2.8468
Iteration: 3291; Percent complete: 82.3%; Average loss: 2.7388
Iteration: 3292; Percent complete: 82.3%; Average loss: 2.9395
Iteration: 3293; Percent complete: 82.3%; Average loss: 2.7116
Iteration: 3294; Percent complete: 82.3%; Average loss: 2.8661
Iteration: 3295; Percent complete: 82.4%; Average loss: 2.7137
Iteration: 3296; Percent complete: 82.4%; Average loss: 2.6625
Iteration: 3297; Percent complete: 82.4%; Average loss: 2.7418
Iteration: 3298; Percent complete: 82.5%; Average loss: 2.9062
Iteration: 3299; Percent complete: 82.5%; Average loss: 2.8732
Iteration: 3300; Percent complete: 82.5%; Average loss: 2.7206
Iteration: 3301; Percent complete: 82.5%; Average loss: 2.7957
Iteration: 3302; Percent complete: 82.5%; Average loss: 2.8014
Iteration: 3303; Percent complete: 82.6%; Average loss: 2.9042
Iteration: 3304; Percent complete: 82.6%; Average loss: 2.9719
Iteration: 3305; Percent complete: 82.6%; Average loss: 2.6783
Iteration: 3306; Percent complete: 82.7%; Average loss: 2.6946
Iteration: 3307; Percent complete: 82.7%; Average loss: 2.9033
Iteration: 3308; Percent complete: 82.7%; Average loss: 2.7838
Iteration: 3309; Percent complete: 82.7%; Average loss: 2.8935
Iteration: 3310; Percent complete: 82.8%; Average loss: 2.8305
Iteration: 3311; Percent complete: 82.8%; Average loss: 3.0090
Iteration: 3312; Percent complete: 82.8%; Average loss: 2.9229
Iteration: 3313; Percent complete: 82.8%; Average loss: 2.7763
Iteration: 3314; Percent complete: 82.8%; Average loss: 2.8008
Iteration: 3315; Percent complete: 82.9%; Average loss: 2.7398
Iteration: 3316; Percent complete: 82.9%; Average loss: 2.5765
Iteration: 3317; Percent complete: 82.9%; Average loss: 2.6972
Iteration: 3318; Percent complete: 83.0%; Average loss: 2.6617
Iteration: 3319; Percent complete: 83.0%; Average loss: 2.7152
Iteration: 3320; Percent complete: 83.0%; Average loss: 2.9970
Iteration: 3321; Percent complete: 83.0%; Average loss: 2.8408
Iteration: 3322; Percent complete: 83.0%; Average loss: 2.7249
Iteration: 3323; Percent complete: 83.1%; Average loss: 2.6764
Iteration: 3324; Percent complete: 83.1%; Average loss: 2.9331
Iteration: 3325; Percent complete: 83.1%; Average loss: 2.9259
Iteration: 3326; Percent complete: 83.2%; Average loss: 3.0200
Iteration: 3327; Percent complete: 83.2%; Average loss: 2.7532
Iteration: 3328; Percent complete: 83.2%; Average loss: 2.7900
Iteration: 3329; Percent complete: 83.2%; Average loss: 2.6527
Iteration: 3330; Percent complete: 83.2%; Average loss: 3.0303
Iteration: 3331; Percent complete: 83.3%; Average loss: 2.9479
Iteration: 3332; Percent complete: 83.3%; Average loss: 2.5832
Iteration: 3333; Percent complete: 83.3%; Average loss: 2.5444
Iteration: 3334; Percent complete: 83.4%; Average loss: 2.7994
Iteration: 3335; Percent complete: 83.4%; Average loss: 2.8109
Iteration: 3336; Percent complete: 83.4%; Average loss: 2.8856
Iteration: 3337; Percent complete: 83.4%; Average loss: 2.8218
Iteration: 3338; Percent complete: 83.5%; Average loss: 2.9308
Iteration: 3339; Percent complete: 83.5%; Average loss: 2.8147
Iteration: 3340; Percent complete: 83.5%; Average loss: 2.9483
Iteration: 3341; Percent complete: 83.5%; Average loss: 2.7606
Iteration: 3342; Percent complete: 83.5%; Average loss: 2.9137
Iteration: 3343; Percent complete: 83.6%; Average loss: 2.8295
Iteration: 3344; Percent complete: 83.6%; Average loss: 2.8034
Iteration: 3345; Percent complete: 83.6%; Average loss: 2.6374
Iteration: 3346; Percent complete: 83.7%; Average loss: 2.7545
Iteration: 3347; Percent complete: 83.7%; Average loss: 2.8430
Iteration: 3348; Percent complete: 83.7%; Average loss: 2.8868
Iteration: 3349; Percent complete: 83.7%; Average loss: 3.0428
Iteration: 3350; Percent complete: 83.8%; Average loss: 2.7735
Iteration: 3351; Percent complete: 83.8%; Average loss: 2.6057
Iteration: 3352; Percent complete: 83.8%; Average loss: 2.6737
Iteration: 3353; Percent complete: 83.8%; Average loss: 3.2117
Iteration: 3354; Percent complete: 83.9%; Average loss: 2.6917
Iteration: 3355; Percent complete: 83.9%; Average loss: 2.9053
Iteration: 3356; Percent complete: 83.9%; Average loss: 2.6933
Iteration: 3357; Percent complete: 83.9%; Average loss: 2.8306
Iteration: 3358; Percent complete: 84.0%; Average loss: 2.4906
Iteration: 3359; Percent complete: 84.0%; Average loss: 2.7201
Iteration: 3360; Percent complete: 84.0%; Average loss: 2.6749
Iteration: 3361; Percent complete: 84.0%; Average loss: 2.6971
Iteration: 3362; Percent complete: 84.0%; Average loss: 2.6295
Iteration: 3363; Percent complete: 84.1%; Average loss: 2.6723
Iteration: 3364; Percent complete: 84.1%; Average loss: 2.6856
Iteration: 3365; Percent complete: 84.1%; Average loss: 2.7942
Iteration: 3366; Percent complete: 84.2%; Average loss: 2.5913
Iteration: 3367; Percent complete: 84.2%; Average loss: 2.5310
Iteration: 3368; Percent complete: 84.2%; Average loss: 2.9799
Iteration: 3369; Percent complete: 84.2%; Average loss: 2.9018
Iteration: 3370; Percent complete: 84.2%; Average loss: 3.0234
Iteration: 3371; Percent complete: 84.3%; Average loss: 2.6471
Iteration: 3372; Percent complete: 84.3%; Average loss: 2.7674
Iteration: 3373; Percent complete: 84.3%; Average loss: 2.7833
Iteration: 3374; Percent complete: 84.4%; Average loss: 2.8454
Iteration: 3375; Percent complete: 84.4%; Average loss: 2.6523
Iteration: 3376; Percent complete: 84.4%; Average loss: 2.7187
Iteration: 3377; Percent complete: 84.4%; Average loss: 2.6630
Iteration: 3378; Percent complete: 84.5%; Average loss: 2.9629
Iteration: 3379; Percent complete: 84.5%; Average loss: 3.0223
Iteration: 3380; Percent complete: 84.5%; Average loss: 2.7837
Iteration: 3381; Percent complete: 84.5%; Average loss: 2.4609
Iteration: 3382; Percent complete: 84.5%; Average loss: 2.7489
Iteration: 3383; Percent complete: 84.6%; Average loss: 2.9737
Iteration: 3384; Percent complete: 84.6%; Average loss: 2.6263
Iteration: 3385; Percent complete: 84.6%; Average loss: 2.8694
Iteration: 3386; Percent complete: 84.7%; Average loss: 2.7276
Iteration: 3387; Percent complete: 84.7%; Average loss: 2.8025
Iteration: 3388; Percent complete: 84.7%; Average loss: 2.6566
Iteration: 3389; Percent complete: 84.7%; Average loss: 2.6718
Iteration: 3390; Percent complete: 84.8%; Average loss: 2.5843
Iteration: 3391; Percent complete: 84.8%; Average loss: 2.9255
Iteration: 3392; Percent complete: 84.8%; Average loss: 2.7295
Iteration: 3393; Percent complete: 84.8%; Average loss: 2.5791
Iteration: 3394; Percent complete: 84.9%; Average loss: 2.8686
Iteration: 3395; Percent complete: 84.9%; Average loss: 2.7228
Iteration: 3396; Percent complete: 84.9%; Average loss: 2.9507
Iteration: 3397; Percent complete: 84.9%; Average loss: 2.5846
Iteration: 3398; Percent complete: 85.0%; Average loss: 2.6277
Iteration: 3399; Percent complete: 85.0%; Average loss: 2.9141
Iteration: 3400; Percent complete: 85.0%; Average loss: 2.9965
Iteration: 3401; Percent complete: 85.0%; Average loss: 2.7601
Iteration: 3402; Percent complete: 85.0%; Average loss: 2.7519
Iteration: 3403; Percent complete: 85.1%; Average loss: 2.7807
Iteration: 3404; Percent complete: 85.1%; Average loss: 2.7004
Iteration: 3405; Percent complete: 85.1%; Average loss: 2.4194
Iteration: 3406; Percent complete: 85.2%; Average loss: 2.9029
Iteration: 3407; Percent complete: 85.2%; Average loss: 2.9457
Iteration: 3408; Percent complete: 85.2%; Average loss: 2.7234
Iteration: 3409; Percent complete: 85.2%; Average loss: 2.6489
Iteration: 3410; Percent complete: 85.2%; Average loss: 2.6174
Iteration: 3411; Percent complete: 85.3%; Average loss: 2.8160
Iteration: 3412; Percent complete: 85.3%; Average loss: 2.7753
Iteration: 3413; Percent complete: 85.3%; Average loss: 2.6950
Iteration: 3414; Percent complete: 85.4%; Average loss: 2.9450
Iteration: 3415; Percent complete: 85.4%; Average loss: 2.5648
Iteration: 3416; Percent complete: 85.4%; Average loss: 2.6003
Iteration: 3417; Percent complete: 85.4%; Average loss: 2.6189
Iteration: 3418; Percent complete: 85.5%; Average loss: 2.8207
Iteration: 3419; Percent complete: 85.5%; Average loss: 2.8575
Iteration: 3420; Percent complete: 85.5%; Average loss: 2.8143
Iteration: 3421; Percent complete: 85.5%; Average loss: 2.8400
Iteration: 3422; Percent complete: 85.5%; Average loss: 2.6878
Iteration: 3423; Percent complete: 85.6%; Average loss: 2.7033
Iteration: 3424; Percent complete: 85.6%; Average loss: 2.7199
Iteration: 3425; Percent complete: 85.6%; Average loss: 2.7694
Iteration: 3426; Percent complete: 85.7%; Average loss: 3.0422
Iteration: 3427; Percent complete: 85.7%; Average loss: 2.9299
Iteration: 3428; Percent complete: 85.7%; Average loss: 2.7796
Iteration: 3429; Percent complete: 85.7%; Average loss: 2.7234
Iteration: 3430; Percent complete: 85.8%; Average loss: 2.8077
Iteration: 3431; Percent complete: 85.8%; Average loss: 2.7511
Iteration: 3432; Percent complete: 85.8%; Average loss: 2.9422
Iteration: 3433; Percent complete: 85.8%; Average loss: 2.5570
Iteration: 3434; Percent complete: 85.9%; Average loss: 2.8807
Iteration: 3435; Percent complete: 85.9%; Average loss: 2.8321
Iteration: 3436; Percent complete: 85.9%; Average loss: 2.9768
Iteration: 3437; Percent complete: 85.9%; Average loss: 2.6186
Iteration: 3438; Percent complete: 86.0%; Average loss: 2.6096
Iteration: 3439; Percent complete: 86.0%; Average loss: 2.7342
Iteration: 3440; Percent complete: 86.0%; Average loss: 2.7546
Iteration: 3441; Percent complete: 86.0%; Average loss: 2.7113
Iteration: 3442; Percent complete: 86.1%; Average loss: 2.8070
Iteration: 3443; Percent complete: 86.1%; Average loss: 2.7718
Iteration: 3444; Percent complete: 86.1%; Average loss: 2.8847
Iteration: 3445; Percent complete: 86.1%; Average loss: 2.6368
Iteration: 3446; Percent complete: 86.2%; Average loss: 2.6735
Iteration: 3447; Percent complete: 86.2%; Average loss: 2.4824
Iteration: 3448; Percent complete: 86.2%; Average loss: 2.7033
Iteration: 3449; Percent complete: 86.2%; Average loss: 2.6412
Iteration: 3450; Percent complete: 86.2%; Average loss: 2.5635
Iteration: 3451; Percent complete: 86.3%; Average loss: 2.5975
Iteration: 3452; Percent complete: 86.3%; Average loss: 2.6986
Iteration: 3453; Percent complete: 86.3%; Average loss: 2.7983
Iteration: 3454; Percent complete: 86.4%; Average loss: 2.7767
Iteration: 3455; Percent complete: 86.4%; Average loss: 2.7236
Iteration: 3456; Percent complete: 86.4%; Average loss: 2.8067
Iteration: 3457; Percent complete: 86.4%; Average loss: 2.6293
Iteration: 3458; Percent complete: 86.5%; Average loss: 2.6763
Iteration: 3459; Percent complete: 86.5%; Average loss: 2.8430
Iteration: 3460; Percent complete: 86.5%; Average loss: 2.9697
Iteration: 3461; Percent complete: 86.5%; Average loss: 3.0289
Iteration: 3462; Percent complete: 86.6%; Average loss: 2.7118
Iteration: 3463; Percent complete: 86.6%; Average loss: 2.6859
Iteration: 3464; Percent complete: 86.6%; Average loss: 2.7126
Iteration: 3465; Percent complete: 86.6%; Average loss: 2.7423
Iteration: 3466; Percent complete: 86.7%; Average loss: 2.7845
Iteration: 3467; Percent complete: 86.7%; Average loss: 2.7970
Iteration: 3468; Percent complete: 86.7%; Average loss: 2.6866
Iteration: 3469; Percent complete: 86.7%; Average loss: 2.6727
Iteration: 3470; Percent complete: 86.8%; Average loss: 2.7852
Iteration: 3471; Percent complete: 86.8%; Average loss: 2.8646
Iteration: 3472; Percent complete: 86.8%; Average loss: 2.6594
Iteration: 3473; Percent complete: 86.8%; Average loss: 2.9783
Iteration: 3474; Percent complete: 86.9%; Average loss: 2.5720
Iteration: 3475; Percent complete: 86.9%; Average loss: 2.7058
Iteration: 3476; Percent complete: 86.9%; Average loss: 3.1377
Iteration: 3477; Percent complete: 86.9%; Average loss: 2.6406
Iteration: 3478; Percent complete: 87.0%; Average loss: 3.0365
Iteration: 3479; Percent complete: 87.0%; Average loss: 2.6235
Iteration: 3480; Percent complete: 87.0%; Average loss: 2.8066
Iteration: 3481; Percent complete: 87.0%; Average loss: 3.0855
Iteration: 3482; Percent complete: 87.1%; Average loss: 2.8604
Iteration: 3483; Percent complete: 87.1%; Average loss: 2.6627
Iteration: 3484; Percent complete: 87.1%; Average loss: 2.7266
Iteration: 3485; Percent complete: 87.1%; Average loss: 2.8166
Iteration: 3486; Percent complete: 87.2%; Average loss: 2.7188
Iteration: 3487; Percent complete: 87.2%; Average loss: 2.8411
Iteration: 3488; Percent complete: 87.2%; Average loss: 2.6620
Iteration: 3489; Percent complete: 87.2%; Average loss: 2.4129
Iteration: 3490; Percent complete: 87.2%; Average loss: 2.6078
Iteration: 3491; Percent complete: 87.3%; Average loss: 2.8307
Iteration: 3492; Percent complete: 87.3%; Average loss: 2.6687
Iteration: 3493; Percent complete: 87.3%; Average loss: 2.4638
Iteration: 3494; Percent complete: 87.4%; Average loss: 2.8494
Iteration: 3495; Percent complete: 87.4%; Average loss: 2.7564
Iteration: 3496; Percent complete: 87.4%; Average loss: 2.8182
Iteration: 3497; Percent complete: 87.4%; Average loss: 2.7345
Iteration: 3498; Percent complete: 87.5%; Average loss: 2.9887
Iteration: 3499; Percent complete: 87.5%; Average loss: 2.6523
Iteration: 3500; Percent complete: 87.5%; Average loss: 2.7320
Iteration: 3501; Percent complete: 87.5%; Average loss: 2.7633
Iteration: 3502; Percent complete: 87.5%; Average loss: 2.4386
Iteration: 3503; Percent complete: 87.6%; Average loss: 2.8388
Iteration: 3504; Percent complete: 87.6%; Average loss: 2.5370
Iteration: 3505; Percent complete: 87.6%; Average loss: 2.4669
Iteration: 3506; Percent complete: 87.6%; Average loss: 2.9926
Iteration: 3507; Percent complete: 87.7%; Average loss: 2.6725
Iteration: 3508; Percent complete: 87.7%; Average loss: 2.6873
Iteration: 3509; Percent complete: 87.7%; Average loss: 2.5766
Iteration: 3510; Percent complete: 87.8%; Average loss: 2.6081
Iteration: 3511; Percent complete: 87.8%; Average loss: 2.8676
Iteration: 3512; Percent complete: 87.8%; Average loss: 2.8776
Iteration: 3513; Percent complete: 87.8%; Average loss: 2.9378
Iteration: 3514; Percent complete: 87.8%; Average loss: 2.7550
Iteration: 3515; Percent complete: 87.9%; Average loss: 2.8618
Iteration: 3516; Percent complete: 87.9%; Average loss: 2.7719
Iteration: 3517; Percent complete: 87.9%; Average loss: 2.7347
Iteration: 3518; Percent complete: 87.9%; Average loss: 2.7007
Iteration: 3519; Percent complete: 88.0%; Average loss: 2.9882
Iteration: 3520; Percent complete: 88.0%; Average loss: 2.8019
Iteration: 3521; Percent complete: 88.0%; Average loss: 2.5686
Iteration: 3522; Percent complete: 88.0%; Average loss: 2.5869
Iteration: 3523; Percent complete: 88.1%; Average loss: 2.8603
Iteration: 3524; Percent complete: 88.1%; Average loss: 2.7648
Iteration: 3525; Percent complete: 88.1%; Average loss: 2.7472
Iteration: 3526; Percent complete: 88.1%; Average loss: 2.6052
Iteration: 3527; Percent complete: 88.2%; Average loss: 2.8161
Iteration: 3528; Percent complete: 88.2%; Average loss: 2.7018
Iteration: 3529; Percent complete: 88.2%; Average loss: 2.5086
Iteration: 3530; Percent complete: 88.2%; Average loss: 2.6513
Iteration: 3531; Percent complete: 88.3%; Average loss: 2.7080
Iteration: 3532; Percent complete: 88.3%; Average loss: 2.7803
Iteration: 3533; Percent complete: 88.3%; Average loss: 2.6616
Iteration: 3534; Percent complete: 88.3%; Average loss: 2.5179
Iteration: 3535; Percent complete: 88.4%; Average loss: 2.6981
Iteration: 3536; Percent complete: 88.4%; Average loss: 2.6719
Iteration: 3537; Percent complete: 88.4%; Average loss: 2.8602
Iteration: 3538; Percent complete: 88.4%; Average loss: 2.5302
Iteration: 3539; Percent complete: 88.5%; Average loss: 2.6080
Iteration: 3540; Percent complete: 88.5%; Average loss: 2.8503
Iteration: 3541; Percent complete: 88.5%; Average loss: 2.6825
Iteration: 3542; Percent complete: 88.5%; Average loss: 2.7332
Iteration: 3543; Percent complete: 88.6%; Average loss: 2.9248
Iteration: 3544; Percent complete: 88.6%; Average loss: 2.7558
Iteration: 3545; Percent complete: 88.6%; Average loss: 2.5693
Iteration: 3546; Percent complete: 88.6%; Average loss: 2.5825
Iteration: 3547; Percent complete: 88.7%; Average loss: 2.7040
Iteration: 3548; Percent complete: 88.7%; Average loss: 2.7138
Iteration: 3549; Percent complete: 88.7%; Average loss: 2.5938
Iteration: 3550; Percent complete: 88.8%; Average loss: 2.9927
Iteration: 3551; Percent complete: 88.8%; Average loss: 2.7267
Iteration: 3552; Percent complete: 88.8%; Average loss: 2.7862
Iteration: 3553; Percent complete: 88.8%; Average loss: 2.6187
Iteration: 3554; Percent complete: 88.8%; Average loss: 2.7605
Iteration: 3555; Percent complete: 88.9%; Average loss: 2.6760
Iteration: 3556; Percent complete: 88.9%; Average loss: 2.8145
Iteration: 3557; Percent complete: 88.9%; Average loss: 2.6237
Iteration: 3558; Percent complete: 88.9%; Average loss: 2.7495
Iteration: 3559; Percent complete: 89.0%; Average loss: 2.8565
Iteration: 3560; Percent complete: 89.0%; Average loss: 2.7866
Iteration: 3561; Percent complete: 89.0%; Average loss: 2.7635
Iteration: 3562; Percent complete: 89.0%; Average loss: 2.8033
Iteration: 3563; Percent complete: 89.1%; Average loss: 2.8570
Iteration: 3564; Percent complete: 89.1%; Average loss: 2.6161
Iteration: 3565; Percent complete: 89.1%; Average loss: 3.0662
Iteration: 3566; Percent complete: 89.1%; Average loss: 2.6610
Iteration: 3567; Percent complete: 89.2%; Average loss: 2.5021
Iteration: 3568; Percent complete: 89.2%; Average loss: 2.7584
Iteration: 3569; Percent complete: 89.2%; Average loss: 2.8292
Iteration: 3570; Percent complete: 89.2%; Average loss: 3.0123
Iteration: 3571; Percent complete: 89.3%; Average loss: 2.5561
Iteration: 3572; Percent complete: 89.3%; Average loss: 2.5552
Iteration: 3573; Percent complete: 89.3%; Average loss: 2.6323
Iteration: 3574; Percent complete: 89.3%; Average loss: 2.7862
Iteration: 3575; Percent complete: 89.4%; Average loss: 2.8052
Iteration: 3576; Percent complete: 89.4%; Average loss: 2.7338
Iteration: 3577; Percent complete: 89.4%; Average loss: 2.9041
Iteration: 3578; Percent complete: 89.5%; Average loss: 2.7679
Iteration: 3579; Percent complete: 89.5%; Average loss: 2.7296
Iteration: 3580; Percent complete: 89.5%; Average loss: 2.7006
Iteration: 3581; Percent complete: 89.5%; Average loss: 2.7829
Iteration: 3582; Percent complete: 89.5%; Average loss: 2.7210
Iteration: 3583; Percent complete: 89.6%; Average loss: 2.5025
Iteration: 3584; Percent complete: 89.6%; Average loss: 2.5286
Iteration: 3585; Percent complete: 89.6%; Average loss: 2.6560
Iteration: 3586; Percent complete: 89.6%; Average loss: 2.6939
Iteration: 3587; Percent complete: 89.7%; Average loss: 2.5932
Iteration: 3588; Percent complete: 89.7%; Average loss: 2.7565
Iteration: 3589; Percent complete: 89.7%; Average loss: 2.6245
Iteration: 3590; Percent complete: 89.8%; Average loss: 2.8300
Iteration: 3591; Percent complete: 89.8%; Average loss: 2.4968
Iteration: 3592; Percent complete: 89.8%; Average loss: 2.8761
Iteration: 3593; Percent complete: 89.8%; Average loss: 2.7206
Iteration: 3594; Percent complete: 89.8%; Average loss: 2.5199
Iteration: 3595; Percent complete: 89.9%; Average loss: 2.6695
Iteration: 3596; Percent complete: 89.9%; Average loss: 2.7426
Iteration: 3597; Percent complete: 89.9%; Average loss: 2.7982
Iteration: 3598; Percent complete: 90.0%; Average loss: 2.9291
Iteration: 3599; Percent complete: 90.0%; Average loss: 2.3853
Iteration: 3600; Percent complete: 90.0%; Average loss: 2.5922
Iteration: 3601; Percent complete: 90.0%; Average loss: 2.6512
Iteration: 3602; Percent complete: 90.0%; Average loss: 2.8278
Iteration: 3603; Percent complete: 90.1%; Average loss: 2.7017
Iteration: 3604; Percent complete: 90.1%; Average loss: 2.8667
Iteration: 3605; Percent complete: 90.1%; Average loss: 2.7480
Iteration: 3606; Percent complete: 90.1%; Average loss: 2.8490
Iteration: 3607; Percent complete: 90.2%; Average loss: 2.8669
Iteration: 3608; Percent complete: 90.2%; Average loss: 2.8575
Iteration: 3609; Percent complete: 90.2%; Average loss: 2.9408
Iteration: 3610; Percent complete: 90.2%; Average loss: 2.6934
Iteration: 3611; Percent complete: 90.3%; Average loss: 2.7542
Iteration: 3612; Percent complete: 90.3%; Average loss: 2.5443
Iteration: 3613; Percent complete: 90.3%; Average loss: 2.5178
Iteration: 3614; Percent complete: 90.3%; Average loss: 2.9873
Iteration: 3615; Percent complete: 90.4%; Average loss: 2.7863
Iteration: 3616; Percent complete: 90.4%; Average loss: 2.5634
Iteration: 3617; Percent complete: 90.4%; Average loss: 2.6462
Iteration: 3618; Percent complete: 90.5%; Average loss: 2.5402
Iteration: 3619; Percent complete: 90.5%; Average loss: 2.6818
Iteration: 3620; Percent complete: 90.5%; Average loss: 2.6770
Iteration: 3621; Percent complete: 90.5%; Average loss: 2.5107
Iteration: 3622; Percent complete: 90.5%; Average loss: 2.5150
Iteration: 3623; Percent complete: 90.6%; Average loss: 2.6404
Iteration: 3624; Percent complete: 90.6%; Average loss: 2.5390
Iteration: 3625; Percent complete: 90.6%; Average loss: 2.6470
Iteration: 3626; Percent complete: 90.6%; Average loss: 2.5209
Iteration: 3627; Percent complete: 90.7%; Average loss: 2.7787
Iteration: 3628; Percent complete: 90.7%; Average loss: 2.5480
Iteration: 3629; Percent complete: 90.7%; Average loss: 2.6002
Iteration: 3630; Percent complete: 90.8%; Average loss: 2.6463
Iteration: 3631; Percent complete: 90.8%; Average loss: 2.6405
Iteration: 3632; Percent complete: 90.8%; Average loss: 2.6098
Iteration: 3633; Percent complete: 90.8%; Average loss: 2.7260
Iteration: 3634; Percent complete: 90.8%; Average loss: 2.8445
Iteration: 3635; Percent complete: 90.9%; Average loss: 2.7967
Iteration: 3636; Percent complete: 90.9%; Average loss: 2.4763
Iteration: 3637; Percent complete: 90.9%; Average loss: 2.5463
Iteration: 3638; Percent complete: 91.0%; Average loss: 2.4524
Iteration: 3639; Percent complete: 91.0%; Average loss: 2.8433
Iteration: 3640; Percent complete: 91.0%; Average loss: 2.7126
Iteration: 3641; Percent complete: 91.0%; Average loss: 2.7256
Iteration: 3642; Percent complete: 91.0%; Average loss: 2.5756
Iteration: 3643; Percent complete: 91.1%; Average loss: 2.5495
Iteration: 3644; Percent complete: 91.1%; Average loss: 2.8295
Iteration: 3645; Percent complete: 91.1%; Average loss: 2.7971
Iteration: 3646; Percent complete: 91.1%; Average loss: 2.6023
Iteration: 3647; Percent complete: 91.2%; Average loss: 2.6371
Iteration: 3648; Percent complete: 91.2%; Average loss: 2.6218
Iteration: 3649; Percent complete: 91.2%; Average loss: 2.7540
Iteration: 3650; Percent complete: 91.2%; Average loss: 2.6551
Iteration: 3651; Percent complete: 91.3%; Average loss: 2.6504
Iteration: 3652; Percent complete: 91.3%; Average loss: 2.9196
Iteration: 3653; Percent complete: 91.3%; Average loss: 2.5312
Iteration: 3654; Percent complete: 91.3%; Average loss: 2.6925
Iteration: 3655; Percent complete: 91.4%; Average loss: 2.7639
Iteration: 3656; Percent complete: 91.4%; Average loss: 2.5865
Iteration: 3657; Percent complete: 91.4%; Average loss: 2.7975
Iteration: 3658; Percent complete: 91.5%; Average loss: 2.4645
Iteration: 3659; Percent complete: 91.5%; Average loss: 2.6574
Iteration: 3660; Percent complete: 91.5%; Average loss: 2.7030
Iteration: 3661; Percent complete: 91.5%; Average loss: 2.7001
Iteration: 3662; Percent complete: 91.5%; Average loss: 2.7018
Iteration: 3663; Percent complete: 91.6%; Average loss: 2.7640
Iteration: 3664; Percent complete: 91.6%; Average loss: 2.5977
Iteration: 3665; Percent complete: 91.6%; Average loss: 2.7763
Iteration: 3666; Percent complete: 91.6%; Average loss: 2.6627
Iteration: 3667; Percent complete: 91.7%; Average loss: 2.6436
Iteration: 3668; Percent complete: 91.7%; Average loss: 3.0191
Iteration: 3669; Percent complete: 91.7%; Average loss: 2.7853
Iteration: 3670; Percent complete: 91.8%; Average loss: 2.7102
Iteration: 3671; Percent complete: 91.8%; Average loss: 2.6898
Iteration: 3672; Percent complete: 91.8%; Average loss: 2.8479
Iteration: 3673; Percent complete: 91.8%; Average loss: 2.6149
Iteration: 3674; Percent complete: 91.8%; Average loss: 2.7377
Iteration: 3675; Percent complete: 91.9%; Average loss: 2.5989
Iteration: 3676; Percent complete: 91.9%; Average loss: 2.6045
Iteration: 3677; Percent complete: 91.9%; Average loss: 2.5743
Iteration: 3678; Percent complete: 92.0%; Average loss: 2.7003
Iteration: 3679; Percent complete: 92.0%; Average loss: 2.9480
Iteration: 3680; Percent complete: 92.0%; Average loss: 2.8007
Iteration: 3681; Percent complete: 92.0%; Average loss: 2.7673
Iteration: 3682; Percent complete: 92.0%; Average loss: 2.6789
Iteration: 3683; Percent complete: 92.1%; Average loss: 2.6302
Iteration: 3684; Percent complete: 92.1%; Average loss: 2.4982
Iteration: 3685; Percent complete: 92.1%; Average loss: 2.6455
Iteration: 3686; Percent complete: 92.2%; Average loss: 2.6710
Iteration: 3687; Percent complete: 92.2%; Average loss: 2.6479
Iteration: 3688; Percent complete: 92.2%; Average loss: 2.6643
Iteration: 3689; Percent complete: 92.2%; Average loss: 2.7574
Iteration: 3690; Percent complete: 92.2%; Average loss: 2.7849
Iteration: 3691; Percent complete: 92.3%; Average loss: 2.8541
Iteration: 3692; Percent complete: 92.3%; Average loss: 2.9683
Iteration: 3693; Percent complete: 92.3%; Average loss: 2.7890
Iteration: 3694; Percent complete: 92.3%; Average loss: 2.6577
Iteration: 3695; Percent complete: 92.4%; Average loss: 2.8924
Iteration: 3696; Percent complete: 92.4%; Average loss: 2.6164
Iteration: 3697; Percent complete: 92.4%; Average loss: 2.7565
Iteration: 3698; Percent complete: 92.5%; Average loss: 2.6701
Iteration: 3699; Percent complete: 92.5%; Average loss: 2.8743
Iteration: 3700; Percent complete: 92.5%; Average loss: 2.7919
Iteration: 3701; Percent complete: 92.5%; Average loss: 2.4863
Iteration: 3702; Percent complete: 92.5%; Average loss: 2.5959
Iteration: 3703; Percent complete: 92.6%; Average loss: 2.6119
Iteration: 3704; Percent complete: 92.6%; Average loss: 2.6401
Iteration: 3705; Percent complete: 92.6%; Average loss: 2.6751
Iteration: 3706; Percent complete: 92.7%; Average loss: 2.5803
Iteration: 3707; Percent complete: 92.7%; Average loss: 2.6006
Iteration: 3708; Percent complete: 92.7%; Average loss: 2.6261
Iteration: 3709; Percent complete: 92.7%; Average loss: 2.5499
Iteration: 3710; Percent complete: 92.8%; Average loss: 2.6565
Iteration: 3711; Percent complete: 92.8%; Average loss: 2.6676
Iteration: 3712; Percent complete: 92.8%; Average loss: 2.6319
Iteration: 3713; Percent complete: 92.8%; Average loss: 2.8743
Iteration: 3714; Percent complete: 92.8%; Average loss: 2.7307
Iteration: 3715; Percent complete: 92.9%; Average loss: 2.6274
Iteration: 3716; Percent complete: 92.9%; Average loss: 2.7856
Iteration: 3717; Percent complete: 92.9%; Average loss: 2.5660
Iteration: 3718; Percent complete: 93.0%; Average loss: 2.7019
Iteration: 3719; Percent complete: 93.0%; Average loss: 2.5477
Iteration: 3720; Percent complete: 93.0%; Average loss: 2.6592
Iteration: 3721; Percent complete: 93.0%; Average loss: 2.7091
Iteration: 3722; Percent complete: 93.0%; Average loss: 2.5278
Iteration: 3723; Percent complete: 93.1%; Average loss: 2.7083
Iteration: 3724; Percent complete: 93.1%; Average loss: 2.8468
Iteration: 3725; Percent complete: 93.1%; Average loss: 2.6704
Iteration: 3726; Percent complete: 93.2%; Average loss: 2.6770
Iteration: 3727; Percent complete: 93.2%; Average loss: 2.4896
Iteration: 3728; Percent complete: 93.2%; Average loss: 2.4423
Iteration: 3729; Percent complete: 93.2%; Average loss: 2.8452
Iteration: 3730; Percent complete: 93.2%; Average loss: 2.5801
Iteration: 3731; Percent complete: 93.3%; Average loss: 2.6024
Iteration: 3732; Percent complete: 93.3%; Average loss: 2.6449
Iteration: 3733; Percent complete: 93.3%; Average loss: 2.7873
Iteration: 3734; Percent complete: 93.3%; Average loss: 2.5487
Iteration: 3735; Percent complete: 93.4%; Average loss: 2.8463
Iteration: 3736; Percent complete: 93.4%; Average loss: 2.7594
Iteration: 3737; Percent complete: 93.4%; Average loss: 2.4859
Iteration: 3738; Percent complete: 93.5%; Average loss: 2.6629
Iteration: 3739; Percent complete: 93.5%; Average loss: 2.6899
Iteration: 3740; Percent complete: 93.5%; Average loss: 2.5750
Iteration: 3741; Percent complete: 93.5%; Average loss: 2.8426
Iteration: 3742; Percent complete: 93.5%; Average loss: 2.8465
Iteration: 3743; Percent complete: 93.6%; Average loss: 2.8562
Iteration: 3744; Percent complete: 93.6%; Average loss: 2.6475
Iteration: 3745; Percent complete: 93.6%; Average loss: 2.5558
Iteration: 3746; Percent complete: 93.7%; Average loss: 2.6536
Iteration: 3747; Percent complete: 93.7%; Average loss: 2.7793
Iteration: 3748; Percent complete: 93.7%; Average loss: 2.8998
Iteration: 3749; Percent complete: 93.7%; Average loss: 2.7141
Iteration: 3750; Percent complete: 93.8%; Average loss: 2.7583
Iteration: 3751; Percent complete: 93.8%; Average loss: 2.8058
Iteration: 3752; Percent complete: 93.8%; Average loss: 2.6941
Iteration: 3753; Percent complete: 93.8%; Average loss: 2.4967
Iteration: 3754; Percent complete: 93.8%; Average loss: 2.8174
Iteration: 3755; Percent complete: 93.9%; Average loss: 2.6734
Iteration: 3756; Percent complete: 93.9%; Average loss: 2.5873
Iteration: 3757; Percent complete: 93.9%; Average loss: 2.5952
Iteration: 3758; Percent complete: 94.0%; Average loss: 2.5812
Iteration: 3759; Percent complete: 94.0%; Average loss: 2.4992
Iteration: 3760; Percent complete: 94.0%; Average loss: 2.7437
Iteration: 3761; Percent complete: 94.0%; Average loss: 2.5995
Iteration: 3762; Percent complete: 94.0%; Average loss: 2.6941
Iteration: 3763; Percent complete: 94.1%; Average loss: 2.6670
Iteration: 3764; Percent complete: 94.1%; Average loss: 2.6052
Iteration: 3765; Percent complete: 94.1%; Average loss: 2.6517
Iteration: 3766; Percent complete: 94.2%; Average loss: 2.7186
Iteration: 3767; Percent complete: 94.2%; Average loss: 2.4369
Iteration: 3768; Percent complete: 94.2%; Average loss: 2.5326
Iteration: 3769; Percent complete: 94.2%; Average loss: 2.5461
Iteration: 3770; Percent complete: 94.2%; Average loss: 2.8263
Iteration: 3771; Percent complete: 94.3%; Average loss: 2.5634
Iteration: 3772; Percent complete: 94.3%; Average loss: 2.6783
Iteration: 3773; Percent complete: 94.3%; Average loss: 2.6562
Iteration: 3774; Percent complete: 94.3%; Average loss: 2.4988
Iteration: 3775; Percent complete: 94.4%; Average loss: 2.6183
Iteration: 3776; Percent complete: 94.4%; Average loss: 2.6875
Iteration: 3777; Percent complete: 94.4%; Average loss: 2.7508
Iteration: 3778; Percent complete: 94.5%; Average loss: 2.8036
Iteration: 3779; Percent complete: 94.5%; Average loss: 2.6496
Iteration: 3780; Percent complete: 94.5%; Average loss: 2.6780
Iteration: 3781; Percent complete: 94.5%; Average loss: 2.7644
Iteration: 3782; Percent complete: 94.5%; Average loss: 2.5741
Iteration: 3783; Percent complete: 94.6%; Average loss: 2.5947
Iteration: 3784; Percent complete: 94.6%; Average loss: 2.5652
Iteration: 3785; Percent complete: 94.6%; Average loss: 2.8126
Iteration: 3786; Percent complete: 94.7%; Average loss: 2.6269
Iteration: 3787; Percent complete: 94.7%; Average loss: 3.0408
Iteration: 3788; Percent complete: 94.7%; Average loss: 2.5992
Iteration: 3789; Percent complete: 94.7%; Average loss: 2.6722
Iteration: 3790; Percent complete: 94.8%; Average loss: 2.6924
Iteration: 3791; Percent complete: 94.8%; Average loss: 2.8040
Iteration: 3792; Percent complete: 94.8%; Average loss: 2.8246
Iteration: 3793; Percent complete: 94.8%; Average loss: 2.6672
Iteration: 3794; Percent complete: 94.8%; Average loss: 2.5778
Iteration: 3795; Percent complete: 94.9%; Average loss: 2.7271
Iteration: 3796; Percent complete: 94.9%; Average loss: 2.8960
Iteration: 3797; Percent complete: 94.9%; Average loss: 2.7663
Iteration: 3798; Percent complete: 95.0%; Average loss: 2.4768
Iteration: 3799; Percent complete: 95.0%; Average loss: 2.7284
Iteration: 3800; Percent complete: 95.0%; Average loss: 2.7215
Iteration: 3801; Percent complete: 95.0%; Average loss: 2.8193
Iteration: 3802; Percent complete: 95.0%; Average loss: 2.7403
Iteration: 3803; Percent complete: 95.1%; Average loss: 2.5791
Iteration: 3804; Percent complete: 95.1%; Average loss: 2.6712
Iteration: 3805; Percent complete: 95.1%; Average loss: 2.5432
Iteration: 3806; Percent complete: 95.2%; Average loss: 2.8291
Iteration: 3807; Percent complete: 95.2%; Average loss: 2.5778
Iteration: 3808; Percent complete: 95.2%; Average loss: 2.6153
Iteration: 3809; Percent complete: 95.2%; Average loss: 2.8352
Iteration: 3810; Percent complete: 95.2%; Average loss: 2.5486
Iteration: 3811; Percent complete: 95.3%; Average loss: 2.6662
Iteration: 3812; Percent complete: 95.3%; Average loss: 2.6209
Iteration: 3813; Percent complete: 95.3%; Average loss: 2.5016
Iteration: 3814; Percent complete: 95.3%; Average loss: 2.7097
Iteration: 3815; Percent complete: 95.4%; Average loss: 2.5524
Iteration: 3816; Percent complete: 95.4%; Average loss: 2.6863
Iteration: 3817; Percent complete: 95.4%; Average loss: 2.7439
Iteration: 3818; Percent complete: 95.5%; Average loss: 2.3187
Iteration: 3819; Percent complete: 95.5%; Average loss: 2.5212
Iteration: 3820; Percent complete: 95.5%; Average loss: 2.5820
Iteration: 3821; Percent complete: 95.5%; Average loss: 2.7956
Iteration: 3822; Percent complete: 95.5%; Average loss: 2.6784
Iteration: 3823; Percent complete: 95.6%; Average loss: 2.5494
Iteration: 3824; Percent complete: 95.6%; Average loss: 2.4850
Iteration: 3825; Percent complete: 95.6%; Average loss: 2.7105
Iteration: 3826; Percent complete: 95.7%; Average loss: 2.3590
Iteration: 3827; Percent complete: 95.7%; Average loss: 2.6944
Iteration: 3828; Percent complete: 95.7%; Average loss: 2.7869
Iteration: 3829; Percent complete: 95.7%; Average loss: 2.8915
Iteration: 3830; Percent complete: 95.8%; Average loss: 2.6966
Iteration: 3831; Percent complete: 95.8%; Average loss: 2.5605
Iteration: 3832; Percent complete: 95.8%; Average loss: 2.7068
Iteration: 3833; Percent complete: 95.8%; Average loss: 2.7641
Iteration: 3834; Percent complete: 95.9%; Average loss: 2.3943
Iteration: 3835; Percent complete: 95.9%; Average loss: 2.5629
Iteration: 3836; Percent complete: 95.9%; Average loss: 2.8073
Iteration: 3837; Percent complete: 95.9%; Average loss: 2.5561
Iteration: 3838; Percent complete: 96.0%; Average loss: 2.5662
Iteration: 3839; Percent complete: 96.0%; Average loss: 2.3929
Iteration: 3840; Percent complete: 96.0%; Average loss: 2.6164
Iteration: 3841; Percent complete: 96.0%; Average loss: 2.7972
Iteration: 3842; Percent complete: 96.0%; Average loss: 2.8086
Iteration: 3843; Percent complete: 96.1%; Average loss: 2.4698
Iteration: 3844; Percent complete: 96.1%; Average loss: 2.4471
Iteration: 3845; Percent complete: 96.1%; Average loss: 2.5949
Iteration: 3846; Percent complete: 96.2%; Average loss: 2.8657
Iteration: 3847; Percent complete: 96.2%; Average loss: 2.6372
Iteration: 3848; Percent complete: 96.2%; Average loss: 2.5893
Iteration: 3849; Percent complete: 96.2%; Average loss: 2.7329
Iteration: 3850; Percent complete: 96.2%; Average loss: 2.5392
Iteration: 3851; Percent complete: 96.3%; Average loss: 2.8005
Iteration: 3852; Percent complete: 96.3%; Average loss: 2.7891
Iteration: 3853; Percent complete: 96.3%; Average loss: 2.5004
Iteration: 3854; Percent complete: 96.4%; Average loss: 2.7216
Iteration: 3855; Percent complete: 96.4%; Average loss: 2.6836
Iteration: 3856; Percent complete: 96.4%; Average loss: 2.4229
Iteration: 3857; Percent complete: 96.4%; Average loss: 2.5904
Iteration: 3858; Percent complete: 96.5%; Average loss: 2.6803
Iteration: 3859; Percent complete: 96.5%; Average loss: 2.5149
Iteration: 3860; Percent complete: 96.5%; Average loss: 2.6981
Iteration: 3861; Percent complete: 96.5%; Average loss: 2.7019
Iteration: 3862; Percent complete: 96.5%; Average loss: 2.6566
Iteration: 3863; Percent complete: 96.6%; Average loss: 2.5671
Iteration: 3864; Percent complete: 96.6%; Average loss: 2.4374
Iteration: 3865; Percent complete: 96.6%; Average loss: 2.3093
Iteration: 3866; Percent complete: 96.7%; Average loss: 2.7161
Iteration: 3867; Percent complete: 96.7%; Average loss: 2.3997
Iteration: 3868; Percent complete: 96.7%; Average loss: 2.8414
Iteration: 3869; Percent complete: 96.7%; Average loss: 2.4619
Iteration: 3870; Percent complete: 96.8%; Average loss: 2.5981
Iteration: 3871; Percent complete: 96.8%; Average loss: 2.5813
Iteration: 3872; Percent complete: 96.8%; Average loss: 2.5201
Iteration: 3873; Percent complete: 96.8%; Average loss: 2.7890
Iteration: 3874; Percent complete: 96.9%; Average loss: 2.7256
Iteration: 3875; Percent complete: 96.9%; Average loss: 2.7206
Iteration: 3876; Percent complete: 96.9%; Average loss: 2.5568
Iteration: 3877; Percent complete: 96.9%; Average loss: 2.7123
Iteration: 3878; Percent complete: 97.0%; Average loss: 2.6450
Iteration: 3879; Percent complete: 97.0%; Average loss: 2.8107
Iteration: 3880; Percent complete: 97.0%; Average loss: 2.6259
Iteration: 3881; Percent complete: 97.0%; Average loss: 2.6394
Iteration: 3882; Percent complete: 97.0%; Average loss: 2.5637
Iteration: 3883; Percent complete: 97.1%; Average loss: 2.6182
Iteration: 3884; Percent complete: 97.1%; Average loss: 2.6959
Iteration: 3885; Percent complete: 97.1%; Average loss: 2.5846
Iteration: 3886; Percent complete: 97.2%; Average loss: 2.6166
Iteration: 3887; Percent complete: 97.2%; Average loss: 2.5981
Iteration: 3888; Percent complete: 97.2%; Average loss: 2.7536
Iteration: 3889; Percent complete: 97.2%; Average loss: 2.8334
Iteration: 3890; Percent complete: 97.2%; Average loss: 2.4413
Iteration: 3891; Percent complete: 97.3%; Average loss: 2.8050
Iteration: 3892; Percent complete: 97.3%; Average loss: 2.5959
Iteration: 3893; Percent complete: 97.3%; Average loss: 2.5715
Iteration: 3894; Percent complete: 97.4%; Average loss: 2.4807
Iteration: 3895; Percent complete: 97.4%; Average loss: 2.5177
Iteration: 3896; Percent complete: 97.4%; Average loss: 2.6718
Iteration: 3897; Percent complete: 97.4%; Average loss: 2.6649
Iteration: 3898; Percent complete: 97.5%; Average loss: 2.4426
Iteration: 3899; Percent complete: 97.5%; Average loss: 2.5055
Iteration: 3900; Percent complete: 97.5%; Average loss: 2.6718
Iteration: 3901; Percent complete: 97.5%; Average loss: 2.2094
Iteration: 3902; Percent complete: 97.5%; Average loss: 2.4592
Iteration: 3903; Percent complete: 97.6%; Average loss: 2.4397
Iteration: 3904; Percent complete: 97.6%; Average loss: 2.5079
Iteration: 3905; Percent complete: 97.6%; Average loss: 2.5195
Iteration: 3906; Percent complete: 97.7%; Average loss: 2.7979
Iteration: 3907; Percent complete: 97.7%; Average loss: 2.6095
Iteration: 3908; Percent complete: 97.7%; Average loss: 2.5901
Iteration: 3909; Percent complete: 97.7%; Average loss: 2.3588
Iteration: 3910; Percent complete: 97.8%; Average loss: 2.5203
Iteration: 3911; Percent complete: 97.8%; Average loss: 2.4988
Iteration: 3912; Percent complete: 97.8%; Average loss: 2.6507
Iteration: 3913; Percent complete: 97.8%; Average loss: 2.7173
Iteration: 3914; Percent complete: 97.9%; Average loss: 2.7059
Iteration: 3915; Percent complete: 97.9%; Average loss: 2.4849
Iteration: 3916; Percent complete: 97.9%; Average loss: 2.5447
Iteration: 3917; Percent complete: 97.9%; Average loss: 2.7097
Iteration: 3918; Percent complete: 98.0%; Average loss: 2.5743
Iteration: 3919; Percent complete: 98.0%; Average loss: 2.5924
Iteration: 3920; Percent complete: 98.0%; Average loss: 2.5019
Iteration: 3921; Percent complete: 98.0%; Average loss: 2.4624
Iteration: 3922; Percent complete: 98.0%; Average loss: 2.6257
Iteration: 3923; Percent complete: 98.1%; Average loss: 2.6022
Iteration: 3924; Percent complete: 98.1%; Average loss: 2.5482
Iteration: 3925; Percent complete: 98.1%; Average loss: 2.5896
Iteration: 3926; Percent complete: 98.2%; Average loss: 2.4777
Iteration: 3927; Percent complete: 98.2%; Average loss: 2.5192
Iteration: 3928; Percent complete: 98.2%; Average loss: 2.5195
Iteration: 3929; Percent complete: 98.2%; Average loss: 2.5938
Iteration: 3930; Percent complete: 98.2%; Average loss: 2.6053
Iteration: 3931; Percent complete: 98.3%; Average loss: 2.5692
Iteration: 3932; Percent complete: 98.3%; Average loss: 2.5751
Iteration: 3933; Percent complete: 98.3%; Average loss: 2.7327
Iteration: 3934; Percent complete: 98.4%; Average loss: 2.9702
Iteration: 3935; Percent complete: 98.4%; Average loss: 2.6305
Iteration: 3936; Percent complete: 98.4%; Average loss: 2.6409
Iteration: 3937; Percent complete: 98.4%; Average loss: 2.5191
Iteration: 3938; Percent complete: 98.5%; Average loss: 2.6808
Iteration: 3939; Percent complete: 98.5%; Average loss: 2.4338
Iteration: 3940; Percent complete: 98.5%; Average loss: 2.5983
Iteration: 3941; Percent complete: 98.5%; Average loss: 2.5861
Iteration: 3942; Percent complete: 98.6%; Average loss: 2.5468
Iteration: 3943; Percent complete: 98.6%; Average loss: 2.6634
Iteration: 3944; Percent complete: 98.6%; Average loss: 2.3722
Iteration: 3945; Percent complete: 98.6%; Average loss: 2.5887
Iteration: 3946; Percent complete: 98.7%; Average loss: 2.6352
Iteration: 3947; Percent complete: 98.7%; Average loss: 2.4547
Iteration: 3948; Percent complete: 98.7%; Average loss: 2.5711
Iteration: 3949; Percent complete: 98.7%; Average loss: 2.7015
Iteration: 3950; Percent complete: 98.8%; Average loss: 2.4189
Iteration: 3951; Percent complete: 98.8%; Average loss: 2.7698
Iteration: 3952; Percent complete: 98.8%; Average loss: 2.5308
Iteration: 3953; Percent complete: 98.8%; Average loss: 2.4634
Iteration: 3954; Percent complete: 98.9%; Average loss: 2.6429
Iteration: 3955; Percent complete: 98.9%; Average loss: 2.5571
Iteration: 3956; Percent complete: 98.9%; Average loss: 2.7450
Iteration: 3957; Percent complete: 98.9%; Average loss: 2.5575
Iteration: 3958; Percent complete: 99.0%; Average loss: 2.5224
Iteration: 3959; Percent complete: 99.0%; Average loss: 2.5633
Iteration: 3960; Percent complete: 99.0%; Average loss: 2.5138
Iteration: 3961; Percent complete: 99.0%; Average loss: 2.5317
Iteration: 3962; Percent complete: 99.1%; Average loss: 2.5643
Iteration: 3963; Percent complete: 99.1%; Average loss: 2.7406
Iteration: 3964; Percent complete: 99.1%; Average loss: 2.6734
Iteration: 3965; Percent complete: 99.1%; Average loss: 2.5716
Iteration: 3966; Percent complete: 99.2%; Average loss: 2.7365
Iteration: 3967; Percent complete: 99.2%; Average loss: 2.7365
Iteration: 3968; Percent complete: 99.2%; Average loss: 2.6220
Iteration: 3969; Percent complete: 99.2%; Average loss: 2.7400
Iteration: 3970; Percent complete: 99.2%; Average loss: 2.3759
Iteration: 3971; Percent complete: 99.3%; Average loss: 2.5277
Iteration: 3972; Percent complete: 99.3%; Average loss: 2.6963
Iteration: 3973; Percent complete: 99.3%; Average loss: 2.4363
Iteration: 3974; Percent complete: 99.4%; Average loss: 2.6505
Iteration: 3975; Percent complete: 99.4%; Average loss: 2.7278
Iteration: 3976; Percent complete: 99.4%; Average loss: 2.8151
Iteration: 3977; Percent complete: 99.4%; Average loss: 2.6118
Iteration: 3978; Percent complete: 99.5%; Average loss: 2.5100
Iteration: 3979; Percent complete: 99.5%; Average loss: 2.5858
Iteration: 3980; Percent complete: 99.5%; Average loss: 2.5389
Iteration: 3981; Percent complete: 99.5%; Average loss: 2.4725
Iteration: 3982; Percent complete: 99.6%; Average loss: 2.6167
Iteration: 3983; Percent complete: 99.6%; Average loss: 2.5766
Iteration: 3984; Percent complete: 99.6%; Average loss: 2.6161
Iteration: 3985; Percent complete: 99.6%; Average loss: 2.6456
Iteration: 3986; Percent complete: 99.7%; Average loss: 2.5953
Iteration: 3987; Percent complete: 99.7%; Average loss: 2.5636
Iteration: 3988; Percent complete: 99.7%; Average loss: 2.4881
Iteration: 3989; Percent complete: 99.7%; Average loss: 2.5400
Iteration: 3990; Percent complete: 99.8%; Average loss: 2.6992
Iteration: 3991; Percent complete: 99.8%; Average loss: 2.6701
Iteration: 3992; Percent complete: 99.8%; Average loss: 2.6955
Iteration: 3993; Percent complete: 99.8%; Average loss: 2.7961
Iteration: 3994; Percent complete: 99.9%; Average loss: 2.7424
Iteration: 3995; Percent complete: 99.9%; Average loss: 2.8025
Iteration: 3996; Percent complete: 99.9%; Average loss: 2.6529
Iteration: 3997; Percent complete: 99.9%; Average loss: 2.5416
Iteration: 3998; Percent complete: 100.0%; Average loss: 2.5808
Iteration: 3999; Percent complete: 100.0%; Average loss: 2.5044
Iteration: 4000; Percent complete: 100.0%; Average loss: 2.4801

평가 수행하기

여러분의 모델과 채팅을 해보고 싶다면 다음 블록을 수행하면 됩니다.

# Dropout 레이어를 평가 모드로 설정합니다
encoder.eval()
decoder.eval()

# 탐색 모듈을 초기화합니다
searcher = GreedySearchDecoder(encoder, decoder)

# 채팅을 시작합니다 (다음 줄의 주석을 제거하면 시작해볼 수 있습니다)
# evaluateInput(encoder, decoder, searcher, voc)

맺음말

이번 튜토리얼을 이것으로 마무리하겠습니다. 축하합니다! 여러분은 이제 생성 챗봇 모델을 만들기 위한 기초 지식을 습득했습니다. 만약 좀 더 관심이 있다면 모델이나 학습 패러미터를 수정해 보면서, 혹은 모델을 학습할 데이터를 바꿔 보면서 챗봇의 행동을 수정해볼 수 있을 것입니다.

그 외에도 딥러닝의 멋진 활용 예에 대한 PyTorch 튜토리얼이 있으니 한 번 확인해 보기 바랍니다!

Total running time of the script: ( 4 minutes 33.278 seconds)

Gallery generated by Sphinx-Gallery


이 문서가 도움이 되었나요?
피드백을 주셔서 감사합니다.

© Copyright 2019, PyTorch.

Built with Sphinx using a theme provided by Read the Docs.

Docs

Access comprehensive developer documentation for PyTorch

View Docs

Tutorials

Get in-depth tutorials for beginners and advanced developers

View Tutorials

Resources

Find development resources and get your questions answered

View Resources