wyx100
wyx100

官网实例详解4.17（imdb_lstm.py）-keras学习笔记四

IMDB数据集情感分类任务训练LSTM（长短时记忆网络）模型

Keras实例目录

代码注释

'''Trains an LSTM model on the IMDB sentiment classification task.
IMDB数据集情感分类任务训练LSTM（长短时记忆网络）模型
The dataset is actually too small for LSTM to be of any advantage
compared to simpler, much faster methods such as TF-IDF + LogReg.
数据集实际上太小，与简单、更快的方法（如TF IDF+LoReGG）相比，LSTM不具有任何优势，

TF-IDF（term frequency–inverse document frequency）是一种用于信息检索与数据挖掘的常用加权技术。
TF意思是词频(Term Frequency)，IDF意思是逆文本频率指数(Inverse Document Frequency)。

# Notes
注意
- RNNs are tricky. Choice of batch size is important,
RNNS很附件，批次大小的选择很重要
choice of loss and optimizer is critical, etc.
损失（函数）和优化器的选择至关重要，等等。
Some configurations won't converge.
有些配置不会收敛。

- LSTM loss decrease patterns during training can be quite different
from what you see with CNNs/MLPs/etc.
LSTM在训练过程中的损失减少模式和CNNs/MLPs/etc完全不同
'''
from __future__ import print_function

from keras.preprocessing import sequence
from keras.models import Sequential
from keras.layers import Dense, Embedding
from keras.layers import LSTM
from keras.datasets import imdb

max_features = 20000
maxlen = 80  # cut texts after this number of words (among top max_features most common words)
# 剪切超出maxlen = 80 长度文本（在Max最常用的单词中）
batch_size = 32

print('Loading data...')
(x_train, y_train), (x_test, y_test) = imdb.load_data(num_words=max_features)
print(len(x_train), 'train sequences')
print(len(x_test), 'test sequences')

print('Pad sequences (samples x time)')
x_train = sequence.pad_sequences(x_train, maxlen=maxlen) # 每个句子maxlen = 80长度
x_test = sequence.pad_sequences(x_test, maxlen=maxlen)
print('x_train shape:', x_train.shape)
print('x_test shape:', x_test.shape)


print('Build model...')
model = Sequential()
model.add(Embedding(max_features, 128)) #  max_features = 20000 词汇量（20000个不同单词）， 每个词用128向量表示，句子长度80个单词（pad_sequences已处理）
model.add(LSTM(128, dropout=0.2, recurrent_dropout=0.2)) 
model.add(Dense(1, activation='sigmoid'))

# try using different optimizers and different optimizer configs
# 尝试不同的优化器和优化器配置
model.compile(loss='binary_crossentropy',
              optimizer='adam',
              metrics=['accuracy'])

print('Train...')
model.fit(x_train, y_train,
          batch_size=batch_size,
          epochs=15,
          validation_data=(x_test, y_test))
score, acc = model.evaluate(x_test, y_test,
                            batch_size=batch_size)
print('Test score:', score)
print('Test accuracy:', acc)

代码执行

C:\ProgramData\Anaconda3\python.exe E:/keras-master/examples/imdb_lstm.py
Using TensorFlow backend.
Loading data...
25000 train sequences
25000 test sequences
Pad sequences (samples x time)
x_train shape: (25000, 80)
x_test shape: (25000, 80)
Build model...
Train...
Train on 25000 samples, validate on 25000 samples
Epoch 1/15

   32/25000 [..............................] - ETA: 59:26 - loss: 0.6931 - acc: 0.5312
   64/25000 [..............................] - ETA: 30:58 - loss: 0.6943 - acc: 0.5000
   96/25000 [..............................] - ETA: 21:58 - loss: 0.6931 - acc: 0.5625
  128/25000 [..............................] - ETA: 17:30 - loss: 0.6930 - acc: 0.5859
  160/25000 [..............................] - ETA: 14:42 - loss: 0.6937 - acc: 0.5312
  192/25000 [..............................] - ETA: 12:43 - loss: 0.6935 - acc: 0.5260
  224/25000 [..............................] - ETA: 11:20 - loss: 0.6937 - acc: 0.5089
  256/25000 [..............................] - ETA: 10:22 - loss: 0.6933 - acc: 0.5156
  288/25000 [..............................] - ETA: 9:36 - loss: 0.6935 - acc: 0.4896 
  320/25000 [..............................] - ETA: 9:17 - loss: 0.6934 - acc: 0.4938
  352/25000 [..............................] - ETA: 8:54 - loss: 0.6930 - acc: 0.5028
  384/25000 [..............................] - ETA: 8:25 - loss: 0.6927 - acc: 0.5182
  416/25000 [..............................] - ETA: 8:01 - loss: 0.6921 - acc: 0.5385
  448/25000 [..............................] - ETA: 7:42 - loss: 0.6918 - acc: 0.5469
  480/25000 [..............................] - ETA: 7:24 - loss: 0.6918 - acc: 0.5521
  512/25000 [..............................] - ETA: 7:09 - loss: 0.6919 - acc: 0.5469
  544/25000 [..............................] - ETA: 6:58 - loss: 0.6917 - acc: 0.5478
  576/25000 [..............................] - ETA: 6:47 - loss: 0.6918 - acc: 0.5434
  608/25000 [..............................] - ETA: 6:36 - loss: 0.6917 - acc: 0.5428
  640/25000 [..............................] - ETA: 6:24 - loss: 0.6914 - acc: 0.5453
  672/25000 [..............................] - ETA: 6:12 - loss: 0.6911 - acc: 0.5506
  704/25000 [..............................] - ETA: 6:04 - loss: 0.6908 - acc: 0.5554
  736/25000 [..............................] - ETA: 5:58 - loss: 0.6902 - acc: 0.5611
  768/25000 [..............................] - ETA: 5:50 - loss: 0.6899 - acc: 0.5625
  800/25000 [..............................] - ETA: 5:44 - loss: 0.6900 - acc: 0.5587
  832/25000 [..............................] - ETA: 5:40 - loss: 0.6900 - acc: 0.5565
  864/25000 [>.............................] - ETA: 5:35 - loss: 0.6894 - acc: 0.5556
  896/25000 [>.............................] - ETA: 5:29 - loss: 0.6886 - acc: 0.5569
  928/25000 [>.............................] - ETA: 5:23 - loss: 0.6891 - acc: 0.5528
  960/25000 [>.............................] - ETA: 5:18 - loss: 0.6892 - acc: 0.5521
  992/25000 [>.............................] - ETA: 5:13 - loss: 0.6887 - acc: 0.5524
 1024/25000 [>.............................] - ETA: 5:07 - loss: 0.6867 - acc: 0.5576
 1056/25000 [>.............................] - ETA: 5:03 - loss: 0.6865 - acc: 0.5587
 1088/25000 [>.............................] - ETA: 5:01 - loss: 0.6872 - acc: 0.5551
 1120/25000 [>.............................] - ETA: 4:57 - loss: 0.6882 - acc: 0.5509
 1152/25000 [>.............................] - ETA: 4:52 - loss: 0.6875 - acc: 0.5512
 1184/25000 [>.............................] - ETA: 4:48 - loss: 0.6873 - acc: 0.5507
 1216/25000 [>.............................] - ETA: 4:44 - loss: 0.6868 - acc: 0.5526
 1248/25000 [>.............................] - ETA: 4:40 - loss: 0.6861 - acc: 0.5569
 1280/25000 [>.............................] - ETA: 4:38 - loss: 0.6859 - acc: 0.5586
 1312/25000 [>.............................] - ETA: 4:37 - loss: 0.6855 - acc: 0.5595
 1344/25000 [>.............................] - ETA: 4:37 - loss: 0.6851 - acc: 0.5603
 1376/25000 [>.............................] - ETA: 4:34 - loss: 0.6849 - acc: 0.5603
 1408/25000 [>.............................] - ETA: 4:32 - loss: 0.6848 - acc: 0.5611
 1440/25000 [>.............................] - ETA: 4:30 - loss: 0.6842 - acc: 0.5646
 1472/25000 [>.............................] - ETA: 4:29 - loss: 0.6842 - acc: 0.5639
 1504/25000 [>.............................] - ETA: 4:26 - loss: 0.6832 - acc: 0.5638
 1536/25000 [>.............................] - ETA: 4:24 - loss: 0.6832 - acc: 0.5632
 1568/25000 [>.............................] - ETA: 4:22 - loss: 0.6833 - acc: 0.5644
 1600/25000 [>.............................] - ETA: 4:19 - loss: 0.6829 - acc: 0.5650
 1632/25000 [>.............................] - ETA: 4:17 - loss: 0.6829 - acc: 0.5643
 1664/25000 [>.............................] - ETA: 4:15 - loss: 0.6822 - acc: 0.5667
 1696/25000 [=>............................] - ETA: 4:14 - loss: 0.6810 - acc: 0.5708
 1728/25000 [=>............................] - ETA: 4:12 - loss: 0.6804 - acc: 0.5712
 1760/25000 [=>............................] - ETA: 4:10 - loss: 0.6797 - acc: 0.5739
 1792/25000 [=>............................] - ETA: 4:10 - loss: 0.6777 - acc: 0.5781
 1824/25000 [=>............................] - ETA: 4:09 - loss: 0.6756 - acc: 0.5811
 1856/25000 [=>............................] - ETA: 4:08 - loss: 0.6723 - acc: 0.5851
 1888/25000 [=>............................] - ETA: 4:06 - loss: 0.6767 - acc: 0.5858
 1920/25000 [=>............................] - ETA: 4:05 - loss: 0.6742 - acc: 0.5885
 1952/25000 [=>............................] - ETA: 4:03 - loss: 0.6706 - acc: 0.5917
 1984/25000 [=>............................] - ETA: 4:02 - loss: 0.6694 - acc: 0.5932
 2016/25000 [=>............................] - ETA: 4:01 - loss: 0.6684 - acc: 0.5947
 2048/25000 [=>............................] - ETA: 4:01 - loss: 0.6665 - acc: 0.5967
 2080/25000 [=>............................] - ETA: 3:59 - loss: 0.6646 - acc: 0.6005
 2112/25000 [=>............................] - ETA: 3:58 - loss: 0.6631 - acc: 0.6013
 2144/25000 [=>............................] - ETA: 3:57 - loss: 0.6619 - acc: 0.6031
 2176/25000 [=>............................] - ETA: 3:55 - loss: 0.6591 - acc: 0.6057
 2208/25000 [=>............................] - ETA: 3:54 - loss: 0.6575 - acc: 0.6069
 2240/25000 [=>............................] - ETA: 3:52 - loss: 0.6560 - acc: 0.6085
 2272/25000 [=>............................] - ETA: 3:51 - loss: 0.6565 - acc: 0.6087
 2304/25000 [=>............................] - ETA: 3:50 - loss: 0.6552 - acc: 0.6102
 2336/25000 [=>............................] - ETA: 3:49 - loss: 0.6517 - acc: 0.6126
 2368/25000 [=>............................] - ETA: 3:48 - loss: 0.6489 - acc: 0.6161
 2400/25000 [=>............................] - ETA: 3:47 - loss: 0.6484 - acc: 0.6179
 2432/25000 [=>............................] - ETA: 3:45 - loss: 0.6478 - acc: 0.6184
 2464/25000 [=>............................] - ETA: 3:44 - loss: 0.6476 - acc: 0.6189
 2496/25000 [=>............................] - ETA: 3:43 - loss: 0.6455 - acc: 0.6214
 2528/25000 [==>...........................] - ETA: 3:41 - loss: 0.6432 - acc: 0.6234
 2560/25000 [==>...........................] - ETA: 3:40 - loss: 0.6418 - acc: 0.6238
 2592/25000 [==>...........................] - ETA: 3:39 - loss: 0.6417 - acc: 0.6246
 2624/25000 [==>...........................] - ETA: 3:37 - loss: 0.6433 - acc: 0.6239
 2656/25000 [==>...........................] - ETA: 3:37 - loss: 0.6425 - acc: 0.6242
 2688/25000 [==>...........................] - ETA: 3:35 - loss: 0.6423 - acc: 0.6239
 2720/25000 [==>...........................] - ETA: 3:34 - loss: 0.6422 - acc: 0.6243
 2752/25000 [==>...........................] - ETA: 3:33 - loss: 0.6418 - acc: 0.6250
 2784/25000 [==>...........................] - ETA: 3:32 - loss: 0.6408 - acc: 0.6275
 2816/25000 [==>...........................] - ETA: 3:31 - loss: 0.6399 - acc: 0.6282
 2848/25000 [==>...........................] - ETA: 3:31 - loss: 0.6391 - acc: 0.6289
 2880/25000 [==>...........................] - ETA: 3:30 - loss: 0.6388 - acc: 0.6288
 2912/25000 [==>...........................] - ETA: 3:29 - loss: 0.6387 - acc: 0.6284
 2944/25000 [==>...........................] - ETA: 3:28 - loss: 0.6379 - acc: 0.6301
 2976/25000 [==>...........................] - ETA: 3:27 - loss: 0.6378 - acc: 0.6300
 3008/25000 [==>...........................] - ETA: 3:27 - loss: 0.6368 - acc: 0.6310
 3040/25000 [==>...........................] - ETA: 3:26 - loss: 0.6349 - acc: 0.6322
 3072/25000 [==>...........................] - ETA: 3:25 - loss: 0.6337 - acc: 0.6331
 3104/25000 [==>...........................] - ETA: 3:24 - loss: 0.6327 - acc: 0.6340
 3136/25000 [==>...........................] - ETA: 3:23 - loss: 0.6319 - acc: 0.6342
 3168/25000 [==>...........................] - ETA: 3:22 - loss: 0.6317 - acc: 0.6342
 3200/25000 [==>...........................] - ETA: 3:21 - loss: 0.6301 - acc: 0.6362
 3232/25000 [==>...........................] - ETA: 3:20 - loss: 0.6275 - acc: 0.6383
 3264/25000 [==>...........................] - ETA: 3:19 - loss: 0.6264 - acc: 0.6400
 3296/25000 [==>...........................] - ETA: 3:18 - loss: 0.6251 - acc: 0.6408
 3328/25000 [==>...........................] - ETA: 3:17 - loss: 0.6241 - acc: 0.6415
 3360/25000 [===>..........................] - ETA: 3:16 - loss: 0.6236 - acc: 0.6423
 3392/25000 [===>..........................] - ETA: 3:16 - loss: 0.6225 - acc: 0.6433
 3424/25000 [===>..........................] - ETA: 3:15 - loss: 0.6203 - acc: 0.6454
 3456/25000 [===>..........................] - ETA: 3:14 - loss: 0.6184 - acc: 0.6461
 3488/25000 [===>..........................] - ETA: 3:13 - loss: 0.6185 - acc: 0.6454
 3520/25000 [===>..........................] - ETA: 3:12 - loss: 0.6168 - acc: 0.6463
 3552/25000 [===>..........................] - ETA: 3:11 - loss: 0.6157 - acc: 0.6478
 3584/25000 [===>..........................] - ETA: 3:11 - loss: 0.6148 - acc: 0.6487
 3616/25000 [===>..........................] - ETA: 3:10 - loss: 0.6134 - acc: 0.6493
 3648/25000 [===>..........................] - ETA: 3:09 - loss: 0.6134 - acc: 0.6497
 3680/25000 [===>..........................] - ETA: 3:08 - loss: 0.6122 - acc: 0.6511
 3712/25000 [===>..........................] - ETA: 3:08 - loss: 0.6130 - acc: 0.6509
 3744/25000 [===>..........................] - ETA: 3:07 - loss: 0.6119 - acc: 0.6522
 3776/25000 [===>..........................] - ETA: 3:07 - loss: 0.6114 - acc: 0.6523
 3808/25000 [===>..........................] - ETA: 3:06 - loss: 0.6104 - acc: 0.6534
 3840/25000 [===>..........................] - ETA: 3:05 - loss: 0.6089 - acc: 0.6547
 3872/25000 [===>..........................] - ETA: 3:05 - loss: 0.6084 - acc: 0.6555
 3904/25000 [===>..........................] - ETA: 3:04 - loss: 0.6078 - acc: 0.6565
 3936/25000 [===>..........................] - ETA: 3:03 - loss: 0.6060 - acc: 0.6580
 3968/25000 [===>..........................] - ETA: 3:03 - loss: 0.6047 - acc: 0.6593
 4000/25000 [===>..........................] - ETA: 3:02 - loss: 0.6034 - acc: 0.6603
 4032/25000 [===>..........................] - ETA: 3:02 - loss: 0.6032 - acc: 0.6617
 4064/25000 [===>..........................] - ETA: 3:01 - loss: 0.6028 - acc: 0.6619
 4096/25000 [===>..........................] - ETA: 3:00 - loss: 0.6019 - acc: 0.6626
 4128/25000 [===>..........................] - ETA: 3:00 - loss: 0.6024 - acc: 0.6623
 4160/25000 [===>..........................] - ETA: 2:59 - loss: 0.6008 - acc: 0.6642
 4192/25000 [====>.........................] - ETA: 2:59 - loss: 0.6003 - acc: 0.6648
 4224/25000 [====>.........................] - ETA: 2:59 - loss: 0.5992 - acc: 0.6660
 4256/25000 [====>.........................] - ETA: 2:58 - loss: 0.5979 - acc: 0.6673
 4288/25000 [====>.........................] - ETA: 2:57 - loss: 0.5967 - acc: 0.6684
 4320/25000 [====>.........................] - ETA: 2:57 - loss: 0.5954 - acc: 0.6697
 4352/25000 [====>.........................] - ETA: 2:57 - loss: 0.5953 - acc: 0.6698
 4384/25000 [====>.........................] - ETA: 2:57 - loss: 0.5941 - acc: 0.6708
 4416/25000 [====>.........................] - ETA: 2:56 - loss: 0.5931 - acc: 0.6721
 4448/25000 [====>.........................] - ETA: 2:55 - loss: 0.5922 - acc: 0.6731
 4480/25000 [====>.........................] - ETA: 2:55 - loss: 0.5915 - acc: 0.6739
 4512/25000 [====>.........................] - ETA: 2:55 - loss: 0.5906 - acc: 0.6746
 4544/25000 [====>.........................] - ETA: 2:54 - loss: 0.5906 - acc: 0.6745
 4576/25000 [====>.........................] - ETA: 2:54 - loss: 0.5893 - acc: 0.6757
 4608/25000 [====>.........................] - ETA: 2:54 - loss: 0.5888 - acc: 0.6764
 4640/25000 [====>.........................] - ETA: 2:54 - loss: 0.5876 - acc: 0.6772
 4672/25000 [====>.........................] - ETA: 2:53 - loss: 0.5857 - acc: 0.6787
 4704/25000 [====>.........................] - ETA: 2:53 - loss: 0.5838 - acc: 0.6798
 4736/25000 [====>.........................] - ETA: 2:53 - loss: 0.5835 - acc: 0.6803
 4768/25000 [====>.........................] - ETA: 2:52 - loss: 0.5825 - acc: 0.6810
 4800/25000 [====>.........................] - ETA: 2:52 - loss: 0.5818 - acc: 0.6817
 4832/25000 [====>.........................] - ETA: 2:52 - loss: 0.5806 - acc: 0.6829
 4864/25000 [====>.........................] - ETA: 2:52 - loss: 0.5814 - acc: 0.6824
 4896/25000 [====>.........................] - ETA: 2:51 - loss: 0.5801 - acc: 0.6838
 4928/25000 [====>.........................] - ETA: 2:51 - loss: 0.5785 - acc: 0.6849
 4960/25000 [====>.........................] - ETA: 2:50 - loss: 0.5781 - acc: 0.6849
 4992/25000 [====>.........................] - ETA: 2:49 - loss: 0.5775 - acc: 0.6853
 5024/25000 [=====>........................] - ETA: 2:49 - loss: 0.5769 - acc: 0.6861
 5056/25000 [=====>........................] - ETA: 2:48 - loss: 0.5758 - acc: 0.6873
 5088/25000 [=====>........................] - ETA: 2:48 - loss: 0.5750 - acc: 0.6883
 5120/25000 [=====>........................] - ETA: 2:47 - loss: 0.5755 - acc: 0.6873
 5152/25000 [=====>........................] - ETA: 2:47 - loss: 0.5753 - acc: 0.6873
 5184/25000 [=====>........................] - ETA: 2:46 - loss: 0.5746 - acc: 0.6879
 5216/25000 [=====>........................] - ETA: 2:46 - loss: 0.5738 - acc: 0.6885
 5248/25000 [=====>........................] - ETA: 2:45 - loss: 0.5732 - acc: 0.6883
 5280/25000 [=====>........................] - ETA: 2:45 - loss: 0.5725 - acc: 0.6890
 5312/25000 [=====>........................] - ETA: 2:44 - loss: 0.5716 - acc: 0.6899
 5344/25000 [=====>........................] - ETA: 2:44 - loss: 0.5706 - acc: 0.6905
 5376/25000 [=====>........................] - ETA: 2:43 - loss: 0.5699 - acc: 0.6912
 5408/25000 [=====>........................] - ETA: 2:43 - loss: 0.5695 - acc: 0.6919
 5440/25000 [=====>........................] - ETA: 2:42 - loss: 0.5685 - acc: 0.6923
 5472/25000 [=====>........................] - ETA: 2:42 - loss: 0.5696 - acc: 0.6923
 5504/25000 [=====>........................] - ETA: 2:41 - loss: 0.5686 - acc: 0.6933
 5536/25000 [=====>........................] - ETA: 2:41 - loss: 0.5677 - acc: 0.6938
 5568/25000 [=====>........................] - ETA: 2:40 - loss: 0.5674 - acc: 0.6936
 5600/25000 [=====>........................] - ETA: 2:40 - loss: 0.5677 - acc: 0.6934
 5632/25000 [=====>........................] - ETA: 2:39 - loss: 0.5669 - acc: 0.6939
 5664/25000 [=====>........................] - ETA: 2:39 - loss: 0.5665 - acc: 0.6940
 5696/25000 [=====>........................] - ETA: 2:38 - loss: 0.5664 - acc: 0.6945
 5728/25000 [=====>........................] - ETA: 2:38 - loss: 0.5661 - acc: 0.6947
 5760/25000 [=====>........................] - ETA: 2:38 - loss: 0.5656 - acc: 0.6951
 5792/25000 [=====>........................] - ETA: 2:37 - loss: 0.5652 - acc: 0.6961
 5824/25000 [=====>........................] - ETA: 2:37 - loss: 0.5644 - acc: 0.6966
 5856/25000 [======>.......................] - ETA: 2:36 - loss: 0.5639 - acc: 0.6972
 5888/25000 [======>.......................] - ETA: 2:36 - loss: 0.5637 - acc: 0.6972
 5920/25000 [======>.......................] - ETA: 2:35 - loss: 0.5629 - acc: 0.6983
 5952/25000 [======>.......................] - ETA: 2:35 - loss: 0.5627 - acc: 0.6986
 5984/25000 [======>.......................] - ETA: 2:34 - loss: 0.5626 - acc: 0.6987
 6016/25000 [======>.......................] - ETA: 2:34 - loss: 0.5617 - acc: 0.6996
 6048/25000 [======>.......................] - ETA: 2:33 - loss: 0.5610 - acc: 0.7002
 6080/25000 [======>.......................] - ETA: 2:33 - loss: 0.5608 - acc: 0.7003
 6112/25000 [======>.......................] - ETA: 2:33 - loss: 0.5603 - acc: 0.7006
 6144/25000 [======>.......................] - ETA: 2:32 - loss: 0.5610 - acc: 0.7004
 6176/25000 [======>.......................] - ETA: 2:32 - loss: 0.5605 - acc: 0.7006
 6208/25000 [======>.......................] - ETA: 2:31 - loss: 0.5604 - acc: 0.7007
 6240/25000 [======>.......................] - ETA: 2:31 - loss: 0.5592 - acc: 0.7014
 6272/25000 [======>.......................] - ETA: 2:31 - loss: 0.5581 - acc: 0.7022
 6304/25000 [======>.......................] - ETA: 2:30 - loss: 0.5580 - acc: 0.7024
 6336/25000 [======>.......................] - ETA: 2:30 - loss: 0.5571 - acc: 0.7030
 6368/25000 [======>.......................] - ETA: 2:30 - loss: 0.5565 - acc: 0.7038
 6400/25000 [======>.......................] - ETA: 2:29 - loss: 0.5566 - acc: 0.7041
 6432/25000 [======>.......................] - ETA: 2:29 - loss: 0.5555 - acc: 0.7051
 6464/25000 [======>.......................] - ETA: 2:29 - loss: 0.5545 - acc: 0.7061
 6496/25000 [======>.......................] - ETA: 2:28 - loss: 0.5537 - acc: 0.7071
 6528/25000 [======>.......................] - ETA: 2:28 - loss: 0.5536 - acc: 0.7070
 6560/25000 [======>.......................] - ETA: 2:28 - loss: 0.5532 - acc: 0.7073
 6592/25000 [======>.......................] - ETA: 2:28 - loss: 0.5530 - acc: 0.7074
 6624/25000 [======>.......................] - ETA: 2:27 - loss: 0.5522 - acc: 0.7082
 6656/25000 [======>.......................] - ETA: 2:27 - loss: 0.5514 - acc: 0.7088
 6688/25000 [=======>......................] - ETA: 2:27 - loss: 0.5508 - acc: 0.7095
 6720/25000 [=======>......................] - ETA: 2:27 - loss: 0.5497 - acc: 0.7103
 6752/25000 [=======>......................] - ETA: 2:26 - loss: 0.5493 - acc: 0.7108
 6784/25000 [=======>......................] - ETA: 2:26 - loss: 0.5499 - acc: 0.7108
 6816/25000 [=======>......................] - ETA: 2:25 - loss: 0.5496 - acc: 0.7108
 6848/25000 [=======>......................] - ETA: 2:25 - loss: 0.5492 - acc: 0.7112
 6880/25000 [=======>......................] - ETA: 2:25 - loss: 0.5481 - acc: 0.7118
 6912/25000 [=======>......................] - ETA: 2:24 - loss: 0.5476 - acc: 0.7121
 6944/25000 [=======>......................] - ETA: 2:24 - loss: 0.5466 - acc: 0.7127
 6976/25000 [=======>......................] - ETA: 2:24 - loss: 0.5453 - acc: 0.7134
 7008/25000 [=======>......................] - ETA: 2:23 - loss: 0.5438 - acc: 0.7145
 7040/25000 [=======>......................] - ETA: 2:23 - loss: 0.5446 - acc: 0.7141
 7072/25000 [=======>......................] - ETA: 2:23 - loss: 0.5446 - acc: 0.7145
 7104/25000 [=======>......................] - ETA: 2:23 - loss: 0.5441 - acc: 0.7152
 7136/25000 [=======>......................] - ETA: 2:22 - loss: 0.5436 - acc: 0.7159
 7168/25000 [=======>......................] - ETA: 2:22 - loss: 0.5435 - acc: 0.7158
 7200/25000 [=======>......................] - ETA: 2:22 - loss: 0.5433 - acc: 0.7161
 7232/25000 [=======>......................] - ETA: 2:21 - loss: 0.5429 - acc: 0.7165
 7264/25000 [=======>......................] - ETA: 2:21 - loss: 0.5428 - acc: 0.7168
 7296/25000 [=======>......................] - ETA: 2:21 - loss: 0.5425 - acc: 0.7171
 7328/25000 [=======>......................] - ETA: 2:21 - loss: 0.5421 - acc: 0.7177
 7360/25000 [=======>......................] - ETA: 2:20 - loss: 0.5412 - acc: 0.7185
 7392/25000 [=======>......................] - ETA: 2:20 - loss: 0.5409 - acc: 0.7189
 7424/25000 [=======>......................] - ETA: 2:20 - loss: 0.5405 - acc: 0.7193
 7456/25000 [=======>......................] - ETA: 2:19 - loss: 0.5399 - acc: 0.7198
 7488/25000 [=======>......................] - ETA: 2:19 - loss: 0.5401 - acc: 0.7197
 7520/25000 [========>.....................] - ETA: 2:19 - loss: 0.5398 - acc: 0.7202
 7552/25000 [========>.....................] - ETA: 2:19 - loss: 0.5395 - acc: 0.7205
 7584/25000 [========>.....................] - ETA: 2:18 - loss: 0.5397 - acc: 0.7202
 7616/25000 [========>.....................] - ETA: 2:18 - loss: 0.5390 - acc: 0.7209
 7648/25000 [========>.....................] - ETA: 2:18 - loss: 0.5389 - acc: 0.7211
 7680/25000 [========>.....................] - ETA: 2:18 - loss: 0.5386 - acc: 0.7211
 7712/25000 [========>.....................] - ETA: 2:17 - loss: 0.5382 - acc: 0.7216
 7744/25000 [========>.....................] - ETA: 2:17 - loss: 0.5377 - acc: 0.7218
 7776/25000 [========>.....................] - ETA: 2:17 - loss: 0.5375 - acc: 0.7217
 7808/25000 [========>.....................] - ETA: 2:16 - loss: 0.5377 - acc: 0.7216
 7840/25000 [========>.....................] - ETA: 2:16 - loss: 0.5369 - acc: 0.7221
 7872/25000 [========>.....................] - ETA: 2:16 - loss: 0.5367 - acc: 0.7223
 7904/25000 [========>.....................] - ETA: 2:16 - loss: 0.5362 - acc: 0.7228
 7936/25000 [========>.....................] - ETA: 2:15 - loss: 0.5352 - acc: 0.7235
 7968/25000 [========>.....................] - ETA: 2:15 - loss: 0.5345 - acc: 0.7240
 8000/25000 [========>.....................] - ETA: 2:14 - loss: 0.5342 - acc: 0.7242
 8032/25000 [========>.....................] - ETA: 2:14 - loss: 0.5339 - acc: 0.7246
 8064/25000 [========>.....................] - ETA: 2:14 - loss: 0.5339 - acc: 0.7247
 8096/25000 [========>.....................] - ETA: 2:13 - loss: 0.5335 - acc: 0.7249
 8128/25000 [========>.....................] - ETA: 2:13 - loss: 0.5330 - acc: 0.7249
 8160/25000 [========>.....................] - ETA: 2:13 - loss: 0.5327 - acc: 0.7252
 8192/25000 [========>.....................] - ETA: 2:12 - loss: 0.5325 - acc: 0.7256
 8224/25000 [========>.....................] - ETA: 2:12 - loss: 0.5324 - acc: 0.7254
 8256/25000 [========>.....................] - ETA: 2:12 - loss: 0.5322 - acc: 0.7255
 8288/25000 [========>.....................] - ETA: 2:12 - loss: 0.5320 - acc: 0.7260
 8320/25000 [========>.....................] - ETA: 2:11 - loss: 0.5315 - acc: 0.7266
 8352/25000 [=========>....................] - ETA: 2:11 - loss: 0.5308 - acc: 0.7273
 8384/25000 [=========>....................] - ETA: 2:10 - loss: 0.5297 - acc: 0.7279
 8416/25000 [=========>....................] - ETA: 2:10 - loss: 0.5291 - acc: 0.7284
 8448/25000 [=========>....................] - ETA: 2:10 - loss: 0.5289 - acc: 0.7288
 8480/25000 [=========>....................] - ETA: 2:09 - loss: 0.5292 - acc: 0.7285
 8512/25000 [=========>....................] - ETA: 2:09 - loss: 0.5287 - acc: 0.7289
 8544/25000 [=========>....................] - ETA: 2:09 - loss: 0.5281 - acc: 0.7294
 8576/25000 [=========>....................] - ETA: 2:08 - loss: 0.5278 - acc: 0.7295
 8608/25000 [=========>....................] - ETA: 2:08 - loss: 0.5267 - acc: 0.7301
 8640/25000 [=========>....................] - ETA: 2:08 - loss: 0.5259 - acc: 0.7307
 8672/25000 [=========>....................] - ETA: 2:07 - loss: 0.5250 - acc: 0.7311
 8704/25000 [=========>....................] - ETA: 2:07 - loss: 0.5251 - acc: 0.7310
 8736/25000 [=========>....................] - ETA: 2:07 - loss: 0.5250 - acc: 0.7313
 8768/25000 [=========>....................] - ETA: 2:06 - loss: 0.5246 - acc: 0.7316
 8800/25000 [=========>....................] - ETA: 2:06 - loss: 0.5247 - acc: 0.7317
 8832/25000 [=========>....................] - ETA: 2:06 - loss: 0.5248 - acc: 0.7318
 8864/25000 [=========>....................] - ETA: 2:05 - loss: 0.5245 - acc: 0.7322
 8896/25000 [=========>....................] - ETA: 2:05 - loss: 0.5245 - acc: 0.7322
 8928/25000 [=========>....................] - ETA: 2:05 - loss: 0.5245 - acc: 0.7323
 8960/25000 [=========>....................] - ETA: 2:04 - loss: 0.5244 - acc: 0.7321
 8992/25000 [=========>....................] - ETA: 2:04 - loss: 0.5244 - acc: 0.7322
 9024/25000 [=========>....................] - ETA: 2:04 - loss: 0.5236 - acc: 0.7328
 9056/25000 [=========>....................] - ETA: 2:04 - loss: 0.5232 - acc: 0.7334
 9088/25000 [=========>....................] - ETA: 2:03 - loss: 0.5229 - acc: 0.7337
 9120/25000 [=========>....................] - ETA: 2:03 - loss: 0.5229 - acc: 0.7341
 9152/25000 [=========>....................] - ETA: 2:03 - loss: 0.5227 - acc: 0.7345
 9184/25000 [==========>...................] - ETA: 2:02 - loss: 0.5222 - acc: 0.7350
 9216/25000 [==========>...................] - ETA: 2:02 - loss: 0.5223 - acc: 0.7348
 9248/25000 [==========>...................] - ETA: 2:02 - loss: 0.5224 - acc: 0.7348
 9280/25000 [==========>...................] - ETA: 2:01 - loss: 0.5217 - acc: 0.7352
 9312/25000 [==========>...................] - ETA: 2:01 - loss: 0.5214 - acc: 0.7355
 9344/25000 [==========>...................] - ETA: 2:01 - loss: 0.5222 - acc: 0.7351
 9376/25000 [==========>...................] - ETA: 2:00 - loss: 0.5218 - acc: 0.7356
 9408/25000 [==========>...................] - ETA: 2:00 - loss: 0.5211 - acc: 0.7361
 9440/25000 [==========>...................] - ETA: 2:00 - loss: 0.5203 - acc: 0.7367
 9472/25000 [==========>...................] - ETA: 1:59 - loss: 0.5192 - acc: 0.7374
 9504/25000 [==========>...................] - ETA: 1:59 - loss: 0.5194 - acc: 0.7375
 9536/25000 [==========>...................] - ETA: 1:59 - loss: 0.5200 - acc: 0.7373
 9568/25000 [==========>...................] - ETA: 1:59 - loss: 0.5203 - acc: 0.7370
 9600/25000 [==========>...................] - ETA: 1:58 - loss: 0.5207 - acc: 0.7370
 9632/25000 [==========>...................] - ETA: 1:58 - loss: 0.5206 - acc: 0.7370
 9664/25000 [==========>...................] - ETA: 1:58 - loss: 0.5201 - acc: 0.7373
 9696/25000 [==========>...................] - ETA: 1:57 - loss: 0.5198 - acc: 0.7376
 9728/25000 [==========>...................] - ETA: 1:57 - loss: 0.5199 - acc: 0.7378
 9760/25000 [==========>...................] - ETA: 1:57 - loss: 0.5197 - acc: 0.7379
 9792/25000 [==========>...................] - ETA: 1:56 - loss: 0.5191 - acc: 0.7384
 9824/25000 [==========>...................] - ETA: 1:56 - loss: 0.5198 - acc: 0.7378
 9856/25000 [==========>...................] - ETA: 1:56 - loss: 0.5197 - acc: 0.7378
 9888/25000 [==========>...................] - ETA: 1:55 - loss: 0.5194 - acc: 0.7381
 9920/25000 [==========>...................] - ETA: 1:55 - loss: 0.5194 - acc: 0.7380
 9952/25000 [==========>...................] - ETA: 1:55 - loss: 0.5188 - acc: 0.7386
 9984/25000 [==========>...................] - ETA: 1:55 - loss: 0.5183 - acc: 0.7388
10016/25000 [===========>..................] - ETA: 1:54 - loss: 0.5182 - acc: 0.7387
10048/25000 [===========>..................] - ETA: 1:54 - loss: 0.5181 - acc: 0.7392
10080/25000 [===========>..................] - ETA: 1:54 - loss: 0.5175 - acc: 0.7395
10112/25000 [===========>..................] - ETA: 1:53 - loss: 0.5168 - acc: 0.7401
10144/25000 [===========>..................] - ETA: 1:53 - loss: 0.5164 - acc: 0.7404
10176/25000 [===========>..................] - ETA: 1:53 - loss: 0.5161 - acc: 0.7408
10208/25000 [===========>..................] - ETA: 1:52 - loss: 0.5162 - acc: 0.7408
10240/25000 [===========>..................] - ETA: 1:52 - loss: 0.5167 - acc: 0.7409
10272/25000 [===========>..................] - ETA: 1:52 - loss: 0.5175 - acc: 0.7408
10304/25000 [===========>..................] - ETA: 1:51 - loss: 0.5178 - acc: 0.7408
10336/25000 [===========>..................] - ETA: 1:51 - loss: 0.5183 - acc: 0.7406
10368/25000 [===========>..................] - ETA: 1:51 - loss: 0.5180 - acc: 0.7407
10400/25000 [===========>..................] - ETA: 1:51 - loss: 0.5175 - acc: 0.7411
10432/25000 [===========>..................] - ETA: 1:50 - loss: 0.5172 - acc: 0.7413
10464/25000 [===========>..................] - ETA: 1:50 - loss: 0.5169 - acc: 0.7415
10496/25000 [===========>..................] - ETA: 1:50 - loss: 0.5164 - acc: 0.7418
10528/25000 [===========>..................] - ETA: 1:49 - loss: 0.5161 - acc: 0.7420
10560/25000 [===========>..................] - ETA: 1:49 - loss: 0.5159 - acc: 0.7420
10592/25000 [===========>..................] - ETA: 1:49 - loss: 0.5155 - acc: 0.7424
10624/25000 [===========>..................] - ETA: 1:49 - loss: 0.5151 - acc: 0.7426
10656/25000 [===========>..................] - ETA: 1:48 - loss: 0.5145 - acc: 0.7430
10688/25000 [===========>..................] - ETA: 1:48 - loss: 0.5145 - acc: 0.7430
10720/25000 [===========>..................] - ETA: 1:48 - loss: 0.5149 - acc: 0.7430
10752/25000 [===========>..................] - ETA: 1:47 - loss: 0.5147 - acc: 0.7430
10784/25000 [===========>..................] - ETA: 1:47 - loss: 0.5144 - acc: 0.7433
10816/25000 [===========>..................] - ETA: 1:47 - loss: 0.5143 - acc: 0.7434
10848/25000 [============>.................] - ETA: 1:47 - loss: 0.5144 - acc: 0.7432
10880/25000 [============>.................] - ETA: 1:46 - loss: 0.5139 - acc: 0.7437
10912/25000 [============>.................] - ETA: 1:46 - loss: 0.5141 - acc: 0.7434
10944/25000 [============>.................] - ETA: 1:46 - loss: 0.5139 - acc: 0.7436
10976/25000 [============>.................] - ETA: 1:45 - loss: 0.5139 - acc: 0.7435
11008/25000 [============>.................] - ETA: 1:45 - loss: 0.5136 - acc: 0.7440
11040/25000 [============>.................] - ETA: 1:45 - loss: 0.5134 - acc: 0.7442
11072/25000 [============>.................] - ETA: 1:45 - loss: 0.5133 - acc: 0.7443
11104/25000 [============>.................] - ETA: 1:44 - loss: 0.5130 - acc: 0.7444
11136/25000 [============>.................] - ETA: 1:44 - loss: 0.5131 - acc: 0.7444
11168/25000 [============>.................] - ETA: 1:44 - loss: 0.5128 - acc: 0.7446
11200/25000 [============>.................] - ETA: 1:43 - loss: 0.5128 - acc: 0.7444
11232/25000 [============>.................] - ETA: 1:43 - loss: 0.5129 - acc: 0.7443
11264/25000 [============>.................] - ETA: 1:43 - loss: 0.5127 - acc: 0.7444
11296/25000 [============>.................] - ETA: 1:43 - loss: 0.5123 - acc: 0.7447
11328/25000 [============>.................] - ETA: 1:42 - loss: 0.5123 - acc: 0.7448
11360/25000 [============>.................] - ETA: 1:42 - loss: 0.5117 - acc: 0.7453
11392/25000 [============>.................] - ETA: 1:42 - loss: 0.5115 - acc: 0.7453
11424/25000 [============>.................] - ETA: 1:41 - loss: 0.5118 - acc: 0.7454
11456/25000 [============>.................] - ETA: 1:41 - loss: 0.5111 - acc: 0.7457
11488/25000 [============>.................] - ETA: 1:41 - loss: 0.5109 - acc: 0.7461
11520/25000 [============>.................] - ETA: 1:41 - loss: 0.5107 - acc: 0.7461
11552/25000 [============>.................] - ETA: 1:40 - loss: 0.5103 - acc: 0.7464
11584/25000 [============>.................] - ETA: 1:40 - loss: 0.5102 - acc: 0.7465
11616/25000 [============>.................] - ETA: 1:40 - loss: 0.5103 - acc: 0.7467
11648/25000 [============>.................] - ETA: 1:39 - loss: 0.5105 - acc: 0.7467
11680/25000 [=============>................] - ETA: 1:39 - loss: 0.5107 - acc: 0.7467
11712/25000 [=============>................] - ETA: 1:39 - loss: 0.5108 - acc: 0.7466
11744/25000 [=============>................] - ETA: 1:39 - loss: 0.5107 - acc: 0.7465
11776/25000 [=============>................] - ETA: 1:38 - loss: 0.5105 - acc: 0.7466
11808/25000 [=============>................] - ETA: 1:38 - loss: 0.5101 - acc: 0.7470
11840/25000 [=============>................] - ETA: 1:38 - loss: 0.5098 - acc: 0.7470
11872/25000 [=============>................] - ETA: 1:38 - loss: 0.5092 - acc: 0.7473
11904/25000 [=============>................] - ETA: 1:37 - loss: 0.5090 - acc: 0.7474
11936/25000 [=============>................] - ETA: 1:37 - loss: 0.5090 - acc: 0.7472
11968/25000 [=============>................] - ETA: 1:37 - loss: 0.5092 - acc: 0.7467
12000/25000 [=============>................] - ETA: 1:36 - loss: 0.5088 - acc: 0.7471
12032/25000 [=============>................] - ETA: 1:36 - loss: 0.5084 - acc: 0.7474
12064/25000 [=============>................] - ETA: 1:36 - loss: 0.5084 - acc: 0.7473
12096/25000 [=============>................] - ETA: 1:36 - loss: 0.5081 - acc: 0.7475
12128/25000 [=============>................] - ETA: 1:35 - loss: 0.5081 - acc: 0.7475
12160/25000 [=============>................] - ETA: 1:35 - loss: 0.5079 - acc: 0.7476
12192/25000 [=============>................] - ETA: 1:35 - loss: 0.5078 - acc: 0.7475
12224/25000 [=============>................] - ETA: 1:35 - loss: 0.5075 - acc: 0.7477
12256/25000 [=============>................] - ETA: 1:34 - loss: 0.5070 - acc: 0.7481
12288/25000 [=============>................] - ETA: 1:34 - loss: 0.5067 - acc: 0.7482
12320/25000 [=============>................] - ETA: 1:34 - loss: 0.5067 - acc: 0.7485
12352/25000 [=============>................] - ETA: 1:33 - loss: 0.5062 - acc: 0.7488
12384/25000 [=============>................] - ETA: 1:33 - loss: 0.5065 - acc: 0.7487
12416/25000 [=============>................] - ETA: 1:33 - loss: 0.5066 - acc: 0.7486
12448/25000 [=============>................] - ETA: 1:33 - loss: 0.5063 - acc: 0.7487
12480/25000 [=============>................] - ETA: 1:32 - loss: 0.5058 - acc: 0.7491
12512/25000 [==============>...............] - ETA: 1:32 - loss: 0.5057 - acc: 0.7493
12544/25000 [==============>...............] - ETA: 1:32 - loss: 0.5058 - acc: 0.7494
12576/25000 [==============>...............] - ETA: 1:32 - loss: 0.5056 - acc: 0.7494
12608/25000 [==============>...............] - ETA: 1:31 - loss: 0.5058 - acc: 0.7494
12640/25000 [==============>...............] - ETA: 1:31 - loss: 0.5055 - acc: 0.7496
12672/25000 [==============>...............] - ETA: 1:31 - loss: 0.5052 - acc: 0.7498
12704/25000 [==============>...............] - ETA: 1:30 - loss: 0.5048 - acc: 0.7501
12736/25000 [==============>...............] - ETA: 1:30 - loss: 0.5048 - acc: 0.7502
12768/25000 [==============>...............] - ETA: 1:30 - loss: 0.5043 - acc: 0.7503
12800/25000 [==============>...............] - ETA: 1:30 - loss: 0.5034 - acc: 0.7509
12832/25000 [==============>...............] - ETA: 1:29 - loss: 0.5029 - acc: 0.7512
12864/25000 [==============>...............] - ETA: 1:29 - loss: 0.5026 - acc: 0.7514
12896/25000 [==============>...............] - ETA: 1:29 - loss: 0.5022 - acc: 0.7517
12928/25000 [==============>...............] - ETA: 1:29 - loss: 0.5018 - acc: 0.7519
12960/25000 [==============>...............] - ETA: 1:28 - loss: 0.5019 - acc: 0.7519
12992/25000 [==============>...............] - ETA: 1:28 - loss: 0.5015 - acc: 0.7521
13024/25000 [==============>...............] - ETA: 1:28 - loss: 0.5009 - acc: 0.7525
13056/25000 [==============>...............] - ETA: 1:28 - loss: 0.5009 - acc: 0.7529
13088/25000 [==============>...............] - ETA: 1:27 - loss: 0.5008 - acc: 0.7530
13120/25000 [==============>...............] - ETA: 1:27 - loss: 0.5008 - acc: 0.7529
13152/25000 [==============>...............] - ETA: 1:27 - loss: 0.5004 - acc: 0.7533
13184/25000 [==============>...............] - ETA: 1:26 - loss: 0.5004 - acc: 0.7533
13216/25000 [==============>...............] - ETA: 1:26 - loss: 0.5004 - acc: 0.7534
13248/25000 [==============>...............] - ETA: 1:26 - loss: 0.4998 - acc: 0.7538
13280/25000 [==============>...............] - ETA: 1:26 - loss: 0.4994 - acc: 0.7541
13312/25000 [==============>...............] - ETA: 1:25 - loss: 0.5000 - acc: 0.7540
13344/25000 [===============>..............] - ETA: 1:25 - loss: 0.5001 - acc: 0.7540
13376/25000 [===============>..............] - ETA: 1:25 - loss: 0.5001 - acc: 0.7541
13408/25000 [===============>..............] - ETA: 1:25 - loss: 0.5000 - acc: 0.7541
13440/25000 [===============>..............] - ETA: 1:24 - loss: 0.5003 - acc: 0.7540
13472/25000 [===============>..............] - ETA: 1:24 - loss: 0.5000 - acc: 0.7542
13504/25000 [===============>..............] - ETA: 1:24 - loss: 0.4999 - acc: 0.7542
13536/25000 [===============>..............] - ETA: 1:24 - loss: 0.4998 - acc: 0.7543
13568/25000 [===============>..............] - ETA: 1:23 - loss: 0.4998 - acc: 0.7541
13600/25000 [===============>..............] - ETA: 1:23 - loss: 0.4998 - acc: 0.7543
13632/25000 [===============>..............] - ETA: 1:23 - loss: 0.4995 - acc: 0.7545
13664/25000 [===============>..............] - ETA: 1:23 - loss: 0.4994 - acc: 0.7544
13696/25000 [===============>..............] - ETA: 1:22 - loss: 0.4992 - acc: 0.7546
13728/25000 [===============>..............] - ETA: 1:22 - loss: 0.4992 - acc: 0.7547
13760/25000 [===============>..............] - ETA: 1:22 - loss: 0.4992 - acc: 0.7546
13792/25000 [===============>..............] - ETA: 1:21 - loss: 0.4992 - acc: 0.7547
13824/25000 [===============>..............] - ETA: 1:21 - loss: 0.4989 - acc: 0.7549
13856/25000 [===============>..............] - ETA: 1:21 - loss: 0.4991 - acc: 0.7548
13888/25000 [===============>..............] - ETA: 1:21 - loss: 0.4991 - acc: 0.7548
13920/25000 [===============>..............] - ETA: 1:20 - loss: 0.4994 - acc: 0.7547
13952/25000 [===============>..............] - ETA: 1:20 - loss: 0.4988 - acc: 0.7552
13984/25000 [===============>..............] - ETA: 1:20 - loss: 0.4984 - acc: 0.7555
14016/25000 [===============>..............] - ETA: 1:20 - loss: 0.4987 - acc: 0.7556
14048/25000 [===============>..............] - ETA: 1:19 - loss: 0.4984 - acc: 0.7556
14080/25000 [===============>..............] - ETA: 1:19 - loss: 0.4982 - acc: 0.7558
14112/25000 [===============>..............] - ETA: 1:19 - loss: 0.4978 - acc: 0.7561
14144/25000 [===============>..............] - ETA: 1:19 - loss: 0.4975 - acc: 0.7563
14176/25000 [================>.............] - ETA: 1:18 - loss: 0.4974 - acc: 0.7563
14208/25000 [================>.............] - ETA: 1:18 - loss: 0.4973 - acc: 0.7564
14240/25000 [================>.............] - ETA: 1:18 - loss: 0.4970 - acc: 0.7565
14272/25000 [================>.............] - ETA: 1:18 - loss: 0.4966 - acc: 0.7567
14304/25000 [================>.............] - ETA: 1:17 - loss: 0.4964 - acc: 0.7567
14336/25000 [================>.............] - ETA: 1:17 - loss: 0.4958 - acc: 0.7570
14368/25000 [================>.............] - ETA: 1:17 - loss: 0.4955 - acc: 0.7572
14400/25000 [================>.............] - ETA: 1:17 - loss: 0.4953 - acc: 0.7573
14432/25000 [================>.............] - ETA: 1:16 - loss: 0.4952 - acc: 0.7573
14464/25000 [================>.............] - ETA: 1:16 - loss: 0.4952 - acc: 0.7573
14496/25000 [================>.............] - ETA: 1:16 - loss: 0.4948 - acc: 0.7575
14528/25000 [================>.............] - ETA: 1:16 - loss: 0.4949 - acc: 0.7574
14560/25000 [================>.............] - ETA: 1:15 - loss: 0.4944 - acc: 0.7578
14592/25000 [================>.............] - ETA: 1:15 - loss: 0.4943 - acc: 0.7577
14624/25000 [================>.............] - ETA: 1:15 - loss: 0.4945 - acc: 0.7577
14656/25000 [================>.............] - ETA: 1:15 - loss: 0.4944 - acc: 0.7578
14688/25000 [================>.............] - ETA: 1:14 - loss: 0.4944 - acc: 0.7577
14720/25000 [================>.............] - ETA: 1:14 - loss: 0.4942 - acc: 0.7577
14752/25000 [================>.............] - ETA: 1:14 - loss: 0.4942 - acc: 0.7579
14784/25000 [================>.............] - ETA: 1:14 - loss: 0.4938 - acc: 0.7582
14816/25000 [================>.............] - ETA: 1:13 - loss: 0.4934 - acc: 0.7584
14848/25000 [================>.............] - ETA: 1:13 - loss: 0.4930 - acc: 0.7586
14880/25000 [================>.............] - ETA: 1:13 - loss: 0.4932 - acc: 0.7585
14912/25000 [================>.............] - ETA: 1:13 - loss: 0.4930 - acc: 0.7587
14944/25000 [================>.............] - ETA: 1:12 - loss: 0.4930 - acc: 0.7588
14976/25000 [================>.............] - ETA: 1:12 - loss: 0.4928 - acc: 0.7589
15008/25000 [=================>............] - ETA: 1:12 - loss: 0.4923 - acc: 0.7592
15040/25000 [=================>............] - ETA: 1:12 - loss: 0.4923 - acc: 0.7593
15072/25000 [=================>............] - ETA: 1:11 - loss: 0.4923 - acc: 0.7594
15104/25000 [=================>............] - ETA: 1:11 - loss: 0.4918 - acc: 0.7597
15136/25000 [=================>............] - ETA: 1:11 - loss: 0.4917 - acc: 0.7596
15168/25000 [=================>............] - ETA: 1:11 - loss: 0.4916 - acc: 0.7596
15200/25000 [=================>............] - ETA: 1:10 - loss: 0.4919 - acc: 0.7593
15232/25000 [=================>............] - ETA: 1:10 - loss: 0.4915 - acc: 0.7597
15264/25000 [=================>............] - ETA: 1:10 - loss: 0.4915 - acc: 0.7596
15296/25000 [=================>............] - ETA: 1:10 - loss: 0.4911 - acc: 0.7598
15328/25000 [=================>............] - ETA: 1:09 - loss: 0.4909 - acc: 0.7599
15360/25000 [=================>............] - ETA: 1:09 - loss: 0.4907 - acc: 0.7601
15392/25000 [=================>............] - ETA: 1:09 - loss: 0.4907 - acc: 0.7601
15424/25000 [=================>............] - ETA: 1:09 - loss: 0.4910 - acc: 0.7599
15456/25000 [=================>............] - ETA: 1:09 - loss: 0.4911 - acc: 0.7598
15488/25000 [=================>............] - ETA: 1:08 - loss: 0.4906 - acc: 0.7600
15520/25000 [=================>............] - ETA: 1:08 - loss: 0.4906 - acc: 0.7600
15552/25000 [=================>............] - ETA: 1:08 - loss: 0.4906 - acc: 0.7600
15584/25000 [=================>............] - ETA: 1:08 - loss: 0.4905 - acc: 0.7600
15616/25000 [=================>............] - ETA: 1:07 - loss: 0.4905 - acc: 0.7601
15648/25000 [=================>............] - ETA: 1:07 - loss: 0.4901 - acc: 0.7602
15680/25000 [=================>............] - ETA: 1:07 - loss: 0.4899 - acc: 0.7604
15712/25000 [=================>............] - ETA: 1:07 - loss: 0.4895 - acc: 0.7608
15744/25000 [=================>............] - ETA: 1:06 - loss: 0.4893 - acc: 0.7608
15776/25000 [=================>............] - ETA: 1:06 - loss: 0.4894 - acc: 0.7606
15808/25000 [=================>............] - ETA: 1:06 - loss: 0.4891 - acc: 0.7607
15840/25000 [==================>...........] - ETA: 1:06 - loss: 0.4890 - acc: 0.7609
15872/25000 [==================>...........] - ETA: 1:05 - loss: 0.4886 - acc: 0.7611
15904/25000 [==================>...........] - ETA: 1:05 - loss: 0.4884 - acc: 0.7614
15936/25000 [==================>...........] - ETA: 1:05 - loss: 0.4887 - acc: 0.7613
15968/25000 [==================>...........] - ETA: 1:05 - loss: 0.4884 - acc: 0.7615
16000/25000 [==================>...........] - ETA: 1:04 - loss: 0.4883 - acc: 0.7614
16032/25000 [==================>...........] - ETA: 1:04 - loss: 0.4883 - acc: 0.7614
16064/25000 [==================>...........] - ETA: 1:04 - loss: 0.4881 - acc: 0.7615
16096/25000 [==================>...........] - ETA: 1:04 - loss: 0.4884 - acc: 0.7614
16128/25000 [==================>...........] - ETA: 1:03 - loss: 0.4881 - acc: 0.7617
16160/25000 [==================>...........] - ETA: 1:03 - loss: 0.4879 - acc: 0.7617
16192/25000 [==================>...........] - ETA: 1:03 - loss: 0.4878 - acc: 0.7618
16224/25000 [==================>...........] - ETA: 1:03 - loss: 0.4875 - acc: 0.7621
16256/25000 [==================>...........] - ETA: 1:02 - loss: 0.4873 - acc: 0.7622
16288/25000 [==================>...........] - ETA: 1:02 - loss: 0.4874 - acc: 0.7622
16320/25000 [==================>...........] - ETA: 1:02 - loss: 0.4874 - acc: 0.7621
16352/25000 [==================>...........] - ETA: 1:02 - loss: 0.4870 - acc: 0.7624
16384/25000 [==================>...........] - ETA: 1:01 - loss: 0.4867 - acc: 0.7626
16416/25000 [==================>...........] - ETA: 1:01 - loss: 0.4865 - acc: 0.7626
16448/25000 [==================>...........] - ETA: 1:01 - loss: 0.4867 - acc: 0.7624
16480/25000 [==================>...........] - ETA: 1:01 - loss: 0.4866 - acc: 0.7624
16512/25000 [==================>...........] - ETA: 1:00 - loss: 0.4866 - acc: 0.7624
16544/25000 [==================>...........] - ETA: 1:00 - loss: 0.4865 - acc: 0.7625
16576/25000 [==================>...........] - ETA: 1:00 - loss: 0.4864 - acc: 0.7625
16608/25000 [==================>...........] - ETA: 1:00 - loss: 0.4864 - acc: 0.7625
16640/25000 [==================>...........] - ETA: 59s - loss: 0.4862 - acc: 0.7627 
16672/25000 [===================>..........] - ETA: 59s - loss: 0.4859 - acc: 0.7629
16704/25000 [===================>..........] - ETA: 59s - loss: 0.4855 - acc: 0.7632
16736/25000 [===================>..........] - ETA: 59s - loss: 0.4853 - acc: 0.7634
16768/25000 [===================>..........] - ETA: 58s - loss: 0.4852 - acc: 0.7635
16800/25000 [===================>..........] - ETA: 58s - loss: 0.4850 - acc: 0.7638
16832/25000 [===================>..........] - ETA: 58s - loss: 0.4849 - acc: 0.7638
16864/25000 [===================>..........] - ETA: 58s - loss: 0.4856 - acc: 0.7635
16896/25000 [===================>..........] - ETA: 57s - loss: 0.4854 - acc: 0.7636
16928/25000 [===================>..........] - ETA: 57s - loss: 0.4853 - acc: 0.7637
16960/25000 [===================>..........] - ETA: 57s - loss: 0.4849 - acc: 0.7640
16992/25000 [===================>..........] - ETA: 57s - loss: 0.4846 - acc: 0.7642
17024/25000 [===================>..........] - ETA: 57s - loss: 0.4843 - acc: 0.7643
17056/25000 [===================>..........] - ETA: 56s - loss: 0.4842 - acc: 0.7644
17088/25000 [===================>..........] - ETA: 56s - loss: 0.4841 - acc: 0.7645
17120/25000 [===================>..........] - ETA: 56s - loss: 0.4839 - acc: 0.7647
17152/25000 [===================>..........] - ETA: 56s - loss: 0.4837 - acc: 0.7647
17184/25000 [===================>..........] - ETA: 55s - loss: 0.4834 - acc: 0.7650
17216/25000 [===================>..........] - ETA: 55s - loss: 0.4830 - acc: 0.7651
17248/25000 [===================>..........] - ETA: 55s - loss: 0.4827 - acc: 0.7652
17280/25000 [===================>..........] - ETA: 55s - loss: 0.4827 - acc: 0.7652
17312/25000 [===================>..........] - ETA: 54s - loss: 0.4827 - acc: 0.7652
17344/25000 [===================>..........] - ETA: 54s - loss: 0.4826 - acc: 0.7652
17376/25000 [===================>..........] - ETA: 54s - loss: 0.4828 - acc: 0.7652
17408/25000 [===================>..........] - ETA: 54s - loss: 0.4830 - acc: 0.7651
17440/25000 [===================>..........] - ETA: 53s - loss: 0.4829 - acc: 0.7651
17472/25000 [===================>..........] - ETA: 53s - loss: 0.4829 - acc: 0.7651
17504/25000 [====================>.........] - ETA: 53s - loss: 0.4827 - acc: 0.7651
17536/25000 [====================>.........] - ETA: 53s - loss: 0.4828 - acc: 0.7651
17568/25000 [====================>.........] - ETA: 52s - loss: 0.4827 - acc: 0.7651
17600/25000 [====================>.........] - ETA: 52s - loss: 0.4823 - acc: 0.7653
17632/25000 [====================>.........] - ETA: 52s - loss: 0.4825 - acc: 0.7653
17664/25000 [====================>.........] - ETA: 52s - loss: 0.4824 - acc: 0.7653
17696/25000 [====================>.........] - ETA: 52s - loss: 0.4824 - acc: 0.7654
17728/25000 [====================>.........] - ETA: 51s - loss: 0.4821 - acc: 0.7657
17760/25000 [====================>.........] - ETA: 51s - loss: 0.4818 - acc: 0.7659
17792/25000 [====================>.........] - ETA: 51s - loss: 0.4818 - acc: 0.7660
17824/25000 [====================>.........] - ETA: 51s - loss: 0.4817 - acc: 0.7660
17856/25000 [====================>.........] - ETA: 50s - loss: 0.4819 - acc: 0.7659
17888/25000 [====================>.........] - ETA: 50s - loss: 0.4817 - acc: 0.7659
17920/25000 [====================>.........] - ETA: 50s - loss: 0.4816 - acc: 0.7660
17952/25000 [====================>.........] - ETA: 50s - loss: 0.4812 - acc: 0.7663
17984/25000 [====================>.........] - ETA: 49s - loss: 0.4809 - acc: 0.7665
18016/25000 [====================>.........] - ETA: 49s - loss: 0.4808 - acc: 0.7665
18048/25000 [====================>.........] - ETA: 49s - loss: 0.4807 - acc: 0.7666
18080/25000 [====================>.........] - ETA: 49s - loss: 0.4805 - acc: 0.7668
18112/25000 [====================>.........] - ETA: 48s - loss: 0.4802 - acc: 0.7669
18144/25000 [====================>.........] - ETA: 48s - loss: 0.4800 - acc: 0.7670
18176/25000 [====================>.........] - ETA: 48s - loss: 0.4797 - acc: 0.7672
18208/25000 [====================>.........] - ETA: 48s - loss: 0.4796 - acc: 0.7674
18240/25000 [====================>.........] - ETA: 48s - loss: 0.4796 - acc: 0.7674
18272/25000 [====================>.........] - ETA: 47s - loss: 0.4796 - acc: 0.7674
18304/25000 [====================>.........] - ETA: 47s - loss: 0.4795 - acc: 0.7674
18336/25000 [=====================>........] - ETA: 47s - loss: 0.4795 - acc: 0.7674
18368/25000 [=====================>........] - ETA: 47s - loss: 0.4791 - acc: 0.7676
18400/25000 [=====================>........] - ETA: 46s - loss: 0.4789 - acc: 0.7678
18432/25000 [=====================>........] - ETA: 46s - loss: 0.4788 - acc: 0.7680
18464/25000 [=====================>........] - ETA: 46s - loss: 0.4786 - acc: 0.7680
18496/25000 [=====================>........] - ETA: 46s - loss: 0.4783 - acc: 0.7682
18528/25000 [=====================>........] - ETA: 45s - loss: 0.4783 - acc: 0.7682
18560/25000 [=====================>........] - ETA: 45s - loss: 0.4783 - acc: 0.7683
18592/25000 [=====================>........] - ETA: 45s - loss: 0.4784 - acc: 0.7682
18624/25000 [=====================>........] - ETA: 45s - loss: 0.4782 - acc: 0.7684
18656/25000 [=====================>........] - ETA: 44s - loss: 0.4778 - acc: 0.7686
18688/25000 [=====================>........] - ETA: 44s - loss: 0.4777 - acc: 0.7687
18720/25000 [=====================>........] - ETA: 44s - loss: 0.4777 - acc: 0.7688
18752/25000 [=====================>........] - ETA: 44s - loss: 0.4775 - acc: 0.7689
18784/25000 [=====================>........] - ETA: 44s - loss: 0.4775 - acc: 0.7690
18816/25000 [=====================>........] - ETA: 43s - loss: 0.4775 - acc: 0.7689
18848/25000 [=====================>........] - ETA: 43s - loss: 0.4776 - acc: 0.7689
18880/25000 [=====================>........] - ETA: 43s - loss: 0.4775 - acc: 0.7690
18912/25000 [=====================>........] - ETA: 43s - loss: 0.4772 - acc: 0.7694
18944/25000 [=====================>........] - ETA: 42s - loss: 0.4776 - acc: 0.7691
18976/25000 [=====================>........] - ETA: 42s - loss: 0.4772 - acc: 0.7694
19008/25000 [=====================>........] - ETA: 42s - loss: 0.4772 - acc: 0.7694
19040/25000 [=====================>........] - ETA: 42s - loss: 0.4772 - acc: 0.7694
19072/25000 [=====================>........] - ETA: 41s - loss: 0.4769 - acc: 0.7695
19104/25000 [=====================>........] - ETA: 41s - loss: 0.4768 - acc: 0.7696
19136/25000 [=====================>........] - ETA: 41s - loss: 0.4771 - acc: 0.7695
19168/25000 [======================>.......] - ETA: 41s - loss: 0.4769 - acc: 0.7697
19200/25000 [======================>.......] - ETA: 41s - loss: 0.4764 - acc: 0.7699
19232/25000 [======================>.......] - ETA: 40s - loss: 0.4763 - acc: 0.7700
19264/25000 [======================>.......] - ETA: 40s - loss: 0.4762 - acc: 0.7700
19296/25000 [======================>.......] - ETA: 40s - loss: 0.4761 - acc: 0.7702
19328/25000 [======================>.......] - ETA: 40s - loss: 0.4759 - acc: 0.7703
19360/25000 [======================>.......] - ETA: 39s - loss: 0.4757 - acc: 0.7704
19392/25000 [======================>.......] - ETA: 39s - loss: 0.4757 - acc: 0.7705
19424/25000 [======================>.......] - ETA: 39s - loss: 0.4759 - acc: 0.7703
19456/25000 [======================>.......] - ETA: 39s - loss: 0.4758 - acc: 0.7704
19488/25000 [======================>.......] - ETA: 38s - loss: 0.4755 - acc: 0.7706
19520/25000 [======================>.......] - ETA: 38s - loss: 0.4753 - acc: 0.7707
19552/25000 [======================>.......] - ETA: 38s - loss: 0.4752 - acc: 0.7709
19584/25000 [======================>.......] - ETA: 38s - loss: 0.4756 - acc: 0.7707
19616/25000 [======================>.......] - ETA: 38s - loss: 0.4752 - acc: 0.7709
19648/25000 [======================>.......] - ETA: 37s - loss: 0.4750 - acc: 0.7710
19680/25000 [======================>.......] - ETA: 37s - loss: 0.4747 - acc: 0.7712
19712/25000 [======================>.......] - ETA: 37s - loss: 0.4747 - acc: 0.7712
19744/25000 [======================>.......] - ETA: 37s - loss: 0.4748 - acc: 0.7711
19776/25000 [======================>.......] - ETA: 36s - loss: 0.4745 - acc: 0.7713
19808/25000 [======================>.......] - ETA: 36s - loss: 0.4746 - acc: 0.7713
19840/25000 [======================>.......] - ETA: 36s - loss: 0.4743 - acc: 0.7714
19872/25000 [======================>.......] - ETA: 36s - loss: 0.4741 - acc: 0.7716
19904/25000 [======================>.......] - ETA: 35s - loss: 0.4737 - acc: 0.7719
19936/25000 [======================>.......] - ETA: 35s - loss: 0.4734 - acc: 0.7721
19968/25000 [======================>.......] - ETA: 35s - loss: 0.4731 - acc: 0.7723
20000/25000 [=======================>......] - ETA: 35s - loss: 0.4729 - acc: 0.7724
20032/25000 [=======================>......] - ETA: 35s - loss: 0.4730 - acc: 0.7725
20064/25000 [=======================>......] - ETA: 34s - loss: 0.4728 - acc: 0.7726
20096/25000 [=======================>......] - ETA: 34s - loss: 0.4727 - acc: 0.7726
20128/25000 [=======================>......] - ETA: 34s - loss: 0.4724 - acc: 0.7728
20160/25000 [=======================>......] - ETA: 34s - loss: 0.4723 - acc: 0.7729
20192/25000 [=======================>......] - ETA: 33s - loss: 0.4722 - acc: 0.7729
20224/25000 [=======================>......] - ETA: 33s - loss: 0.4722 - acc: 0.7729
20256/25000 [=======================>......] - ETA: 33s - loss: 0.4720 - acc: 0.7731
20288/25000 [=======================>......] - ETA: 33s - loss: 0.4717 - acc: 0.7732
20320/25000 [=======================>......] - ETA: 32s - loss: 0.4720 - acc: 0.7731
20352/25000 [=======================>......] - ETA: 32s - loss: 0.4718 - acc: 0.7733
20384/25000 [=======================>......] - ETA: 32s - loss: 0.4714 - acc: 0.7735
20416/25000 [=======================>......] - ETA: 32s - loss: 0.4716 - acc: 0.7734
20448/25000 [=======================>......] - ETA: 32s - loss: 0.4714 - acc: 0.7735
20480/25000 [=======================>......] - ETA: 31s - loss: 0.4712 - acc: 0.7735
20512/25000 [=======================>......] - ETA: 31s - loss: 0.4712 - acc: 0.7735
20544/25000 [=======================>......] - ETA: 31s - loss: 0.4708 - acc: 0.7738
20576/25000 [=======================>......] - ETA: 31s - loss: 0.4704 - acc: 0.7740
20608/25000 [=======================>......] - ETA: 30s - loss: 0.4702 - acc: 0.7742
20640/25000 [=======================>......] - ETA: 30s - loss: 0.4699 - acc: 0.7744
20672/25000 [=======================>......] - ETA: 30s - loss: 0.4696 - acc: 0.7746
20704/25000 [=======================>......] - ETA: 30s - loss: 0.4693 - acc: 0.7747
20736/25000 [=======================>......] - ETA: 29s - loss: 0.4691 - acc: 0.7748
20768/25000 [=======================>......] - ETA: 29s - loss: 0.4689 - acc: 0.7749
20800/25000 [=======================>......] - ETA: 29s - loss: 0.4688 - acc: 0.7750
20832/25000 [=======================>......] - ETA: 29s - loss: 0.4688 - acc: 0.7752
20864/25000 [========================>.....] - ETA: 29s - loss: 0.4688 - acc: 0.7752
20896/25000 [========================>.....] - ETA: 28s - loss: 0.4687 - acc: 0.7753
20928/25000 [========================>.....] - ETA: 28s - loss: 0.4685 - acc: 0.7753
20960/25000 [========================>.....] - ETA: 28s - loss: 0.4687 - acc: 0.7753
20992/25000 [========================>.....] - ETA: 28s - loss: 0.4687 - acc: 0.7754
21024/25000 [========================>.....] - ETA: 27s - loss: 0.4688 - acc: 0.7755
21056/25000 [========================>.....] - ETA: 27s - loss: 0.4689 - acc: 0.7755
21088/25000 [========================>.....] - ETA: 27s - loss: 0.4688 - acc: 0.7756
21120/25000 [========================>.....] - ETA: 27s - loss: 0.4686 - acc: 0.7758
21152/25000 [========================>.....] - ETA: 26s - loss: 0.4684 - acc: 0.7758
21184/25000 [========================>.....] - ETA: 26s - loss: 0.4684 - acc: 0.7758
21216/25000 [========================>.....] - ETA: 26s - loss: 0.4683 - acc: 0.7759
21248/25000 [========================>.....] - ETA: 26s - loss: 0.4685 - acc: 0.7757
21280/25000 [========================>.....] - ETA: 26s - loss: 0.4685 - acc: 0.7758
21312/25000 [========================>.....] - ETA: 25s - loss: 0.4684 - acc: 0.7758
21344/25000 [========================>.....] - ETA: 25s - loss: 0.4681 - acc: 0.7759
21376/25000 [========================>.....] - ETA: 25s - loss: 0.4679 - acc: 0.7761
21408/25000 [========================>.....] - ETA: 25s - loss: 0.4680 - acc: 0.7760
21440/25000 [========================>.....] - ETA: 24s - loss: 0.4683 - acc: 0.7758
21472/25000 [========================>.....] - ETA: 24s - loss: 0.4682 - acc: 0.7758
21504/25000 [========================>.....] - ETA: 24s - loss: 0.4681 - acc: 0.7759
21536/25000 [========================>.....] - ETA: 24s - loss: 0.4679 - acc: 0.7760
21568/25000 [========================>.....] - ETA: 24s - loss: 0.4678 - acc: 0.7761
21600/25000 [========================>.....] - ETA: 23s - loss: 0.4677 - acc: 0.7762
21632/25000 [========================>.....] - ETA: 23s - loss: 0.4676 - acc: 0.7763
21664/25000 [========================>.....] - ETA: 23s - loss: 0.4673 - acc: 0.7764
21696/25000 [=========================>....] - ETA: 23s - loss: 0.4672 - acc: 0.7765
21728/25000 [=========================>....] - ETA: 22s - loss: 0.4672 - acc: 0.7764
21760/25000 [=========================>....] - ETA: 22s - loss: 0.4670 - acc: 0.7765
21792/25000 [=========================>....] - ETA: 22s - loss: 0.4669 - acc: 0.7767
21824/25000 [=========================>....] - ETA: 22s - loss: 0.4674 - acc: 0.7766
21856/25000 [=========================>....] - ETA: 21s - loss: 0.4674 - acc: 0.7765
21888/25000 [=========================>....] - ETA: 21s - loss: 0.4671 - acc: 0.7768
21920/25000 [=========================>....] - ETA: 21s - loss: 0.4668 - acc: 0.7770
21952/25000 [=========================>....] - ETA: 21s - loss: 0.4667 - acc: 0.7769
21984/25000 [=========================>....] - ETA: 21s - loss: 0.4668 - acc: 0.7769
22016/25000 [=========================>....] - ETA: 20s - loss: 0.4668 - acc: 0.7770
22048/25000 [=========================>....] - ETA: 20s - loss: 0.4665 - acc: 0.7772
22080/25000 [=========================>....] - ETA: 20s - loss: 0.4667 - acc: 0.7773
22112/25000 [=========================>....] - ETA: 20s - loss: 0.4665 - acc: 0.7773
22144/25000 [=========================>....] - ETA: 19s - loss: 0.4665 - acc: 0.7774
22176/25000 [=========================>....] - ETA: 19s - loss: 0.4665 - acc: 0.7774
22208/25000 [=========================>....] - ETA: 19s - loss: 0.4662 - acc: 0.7776
22240/25000 [=========================>....] - ETA: 19s - loss: 0.4662 - acc: 0.7775
22272/25000 [=========================>....] - ETA: 19s - loss: 0.4659 - acc: 0.7777
22304/25000 [=========================>....] - ETA: 18s - loss: 0.4659 - acc: 0.7777
22336/25000 [=========================>....] - ETA: 18s - loss: 0.4660 - acc: 0.7776
22368/25000 [=========================>....] - ETA: 18s - loss: 0.4662 - acc: 0.7774
22400/25000 [=========================>....] - ETA: 18s - loss: 0.4661 - acc: 0.7775
22432/25000 [=========================>....] - ETA: 17s - loss: 0.4660 - acc: 0.7775
22464/25000 [=========================>....] - ETA: 17s - loss: 0.4659 - acc: 0.7777
22496/25000 [=========================>....] - ETA: 17s - loss: 0.4659 - acc: 0.7778
22528/25000 [==========================>...] - ETA: 17s - loss: 0.4664 - acc: 0.7777
22560/25000 [==========================>...] - ETA: 17s - loss: 0.4663 - acc: 0.7778
22592/25000 [==========================>...] - ETA: 16s - loss: 0.4663 - acc: 0.7778
22624/25000 [==========================>...] - ETA: 16s - loss: 0.4664 - acc: 0.7777
22656/25000 [==========================>...] - ETA: 16s - loss: 0.4662 - acc: 0.7778
22688/25000 [==========================>...] - ETA: 16s - loss: 0.4662 - acc: 0.7779
22720/25000 [==========================>...] - ETA: 15s - loss: 0.4661 - acc: 0.7779
22752/25000 [==========================>...] - ETA: 15s - loss: 0.4658 - acc: 0.7780
22784/25000 [==========================>...] - ETA: 15s - loss: 0.4657 - acc: 0.7782
22816/25000 [==========================>...] - ETA: 15s - loss: 0.4654 - acc: 0.7783
22848/25000 [==========================>...] - ETA: 15s - loss: 0.4652 - acc: 0.7785
22880/25000 [==========================>...] - ETA: 14s - loss: 0.4655 - acc: 0.7784
22912/25000 [==========================>...] - ETA: 14s - loss: 0.4654 - acc: 0.7784
22944/25000 [==========================>...] - ETA: 14s - loss: 0.4652 - acc: 0.7786
22976/25000 [==========================>...] - ETA: 14s - loss: 0.4651 - acc: 0.7787
23008/25000 [==========================>...] - ETA: 13s - loss: 0.4650 - acc: 0.7787
23040/25000 [==========================>...] - ETA: 13s - loss: 0.4649 - acc: 0.7789
23072/25000 [==========================>...] - ETA: 13s - loss: 0.4648 - acc: 0.7790
23104/25000 [==========================>...] - ETA: 13s - loss: 0.4646 - acc: 0.7791
23136/25000 [==========================>...] - ETA: 12s - loss: 0.4644 - acc: 0.7793
23168/25000 [==========================>...] - ETA: 12s - loss: 0.4643 - acc: 0.7794
23200/25000 [==========================>...] - ETA: 12s - loss: 0.4644 - acc: 0.7794
23232/25000 [==========================>...] - ETA: 12s - loss: 0.4643 - acc: 0.7794
23264/25000 [==========================>...] - ETA: 12s - loss: 0.4642 - acc: 0.7795
23296/25000 [==========================>...] - ETA: 11s - loss: 0.4640 - acc: 0.7797
23328/25000 [==========================>...] - ETA: 11s - loss: 0.4637 - acc: 0.7798
23360/25000 [===========================>..] - ETA: 11s - loss: 0.4635 - acc: 0.7800
23392/25000 [===========================>..] - ETA: 11s - loss: 0.4634 - acc: 0.7801
23424/25000 [===========================>..] - ETA: 10s - loss: 0.4634 - acc: 0.7801
23456/25000 [===========================>..] - ETA: 10s - loss: 0.4633 - acc: 0.7802
23488/25000 [===========================>..] - ETA: 10s - loss: 0.4633 - acc: 0.7802
23520/25000 [===========================>..] - ETA: 10s - loss: 0.4633 - acc: 0.7801
23552/25000 [===========================>..] - ETA: 10s - loss: 0.4630 - acc: 0.7804
23584/25000 [===========================>..] - ETA: 9s - loss: 0.4630 - acc: 0.7804 
23616/25000 [===========================>..] - ETA: 9s - loss: 0.4628 - acc: 0.7805
23648/25000 [===========================>..] - ETA: 9s - loss: 0.4626 - acc: 0.7806
23680/25000 [===========================>..] - ETA: 9s - loss: 0.4625 - acc: 0.7808
23712/25000 [===========================>..] - ETA: 8s - loss: 0.4622 - acc: 0.7810
23744/25000 [===========================>..] - ETA: 8s - loss: 0.4621 - acc: 0.7810
23776/25000 [===========================>..] - ETA: 8s - loss: 0.4621 - acc: 0.7810
23808/25000 [===========================>..] - ETA: 8s - loss: 0.4621 - acc: 0.7810
23840/25000 [===========================>..] - ETA: 8s - loss: 0.4621 - acc: 0.7811
23872/25000 [===========================>..] - ETA: 7s - loss: 0.4622 - acc: 0.7811
23904/25000 [===========================>..] - ETA: 7s - loss: 0.4621 - acc: 0.7810
23936/25000 [===========================>..] - ETA: 7s - loss: 0.4621 - acc: 0.7810
23968/25000 [===========================>..] - ETA: 7s - loss: 0.4620 - acc: 0.7811
24000/25000 [===========================>..] - ETA: 6s - loss: 0.4618 - acc: 0.7812
24032/25000 [===========================>..] - ETA: 6s - loss: 0.4620 - acc: 0.7812
24064/25000 [===========================>..] - ETA: 6s - loss: 0.4619 - acc: 0.7812
24096/25000 [===========================>..] - ETA: 6s - loss: 0.4619 - acc: 0.7813
24128/25000 [===========================>..] - ETA: 6s - loss: 0.4618 - acc: 0.7814
24160/25000 [===========================>..] - ETA: 5s - loss: 0.4615 - acc: 0.7815
24192/25000 [============================>.] - ETA: 5s - loss: 0.4614 - acc: 0.7815
24224/25000 [============================>.] - ETA: 5s - loss: 0.4615 - acc: 0.7814
24256/25000 [============================>.] - ETA: 5s - loss: 0.4613 - acc: 0.7815
24288/25000 [============================>.] - ETA: 4s - loss: 0.4613 - acc: 0.7816
24320/25000 [============================>.] - ETA: 4s - loss: 0.4611 - acc: 0.7817
24352/25000 [============================>.] - ETA: 4s - loss: 0.4611 - acc: 0.7817
24384/25000 [============================>.] - ETA: 4s - loss: 0.4610 - acc: 0.7817
24416/25000 [============================>.] - ETA: 4s - loss: 0.4609 - acc: 0.7817
24448/25000 [============================>.] - ETA: 3s - loss: 0.4608 - acc: 0.7817
24480/25000 [============================>.] - ETA: 3s - loss: 0.4606 - acc: 0.7818
24512/25000 [============================>.] - ETA: 3s - loss: 0.4606 - acc: 0.7817
24544/25000 [============================>.] - ETA: 3s - loss: 0.4606 - acc: 0.7817
24576/25000 [============================>.] - ETA: 2s - loss: 0.4606 - acc: 0.7817
24608/25000 [============================>.] - ETA: 2s - loss: 0.4605 - acc: 0.7818
24640/25000 [============================>.] - ETA: 2s - loss: 0.4604 - acc: 0.7818
24672/25000 [============================>.] - ETA: 2s - loss: 0.4603 - acc: 0.7818
24704/25000 [============================>.] - ETA: 2s - loss: 0.4602 - acc: 0.7819
24736/25000 [============================>.] - ETA: 1s - loss: 0.4603 - acc: 0.7819
24768/25000 [============================>.] - ETA: 1s - loss: 0.4602 - acc: 0.7819
24800/25000 [============================>.] - ETA: 1s - loss: 0.4599 - acc: 0.7820
24832/25000 [============================>.] - ETA: 1s - loss: 0.4598 - acc: 0.7820
24864/25000 [============================>.] - ETA: 0s - loss: 0.4598 - acc: 0.7820
24896/25000 [============================>.] - ETA: 0s - loss: 0.4598 - acc: 0.7820
24928/25000 [============================>.] - ETA: 0s - loss: 0.4598 - acc: 0.7820
24960/25000 [============================>.] - ETA: 0s - loss: 0.4595 - acc: 0.7821
24992/25000 [============================>.] - ETA: 0s - loss: 0.4595 - acc: 0.7822
25000/25000 [==============================] - 209s 8ms/step - loss: 0.4594 - acc: 0.7822 - val_loss: 0.3851 - val_acc: 0.8335
Epoch 2/15

   32/25000 [..............................] - ETA: 2:34 - loss: 0.3305 - acc: 0.8438
   64/25000 [..............................] - ETA: 2:32 - loss: 0.3299 - acc: 0.8438

24576/25000 [============================>.] - ETA: 2s - loss: 0.3040 - acc: 0.8789
24608/25000 [============================>.] - ETA: 2s - loss: 0.3041 - acc: 0.8790

24960/25000 [============================>.] - ETA: 0s - loss: 0.0138 - acc: 0.9955
24992/25000 [============================>.] - ETA: 0s - loss: 0.0138 - acc: 0.9955
25000/25000 [==============================] - 179s 7ms/step - loss: 0.0138 - acc: 0.9955 - val_loss: 1.2954 - val_acc: 0.8048
Epoch 15/15

   32/25000 [..............................] - ETA: 2:25 - loss: 0.0106 - acc: 1.0000

25000/25000 [==============================] - 34s 1ms/step
Test score: 1.0338516357156635
Test accuracy: 0.812

Process finished with exit code 0

关联阅读

http://machinelearningmastery.com/understanding-stateful-lstm-recurrent-neural-networks-python-keras/

https://blog.csdn.net/jiangpeng59/article/details/77533309

https://blog.csdn.net/ma416539432/article/details/53509607

https://blog.csdn.net/Nicholas_Liu2017/article/details/76018432

https://blog.csdn.net/zzukun/article/details/49968129

Keras详细介绍

英文：https://keras.io/

中文：http://keras-cn.readthedocs.io/en/latest/

实例下载

https://github.com/keras-team/keras

https://github.com/keras-team/keras/tree/master/examples

完整项目下载

方便没积分童鞋，请加企鹅452205574，共享文件夹。

包括：代码、数据集合（图片）、已生成model、安装库文件等。

你可能感兴趣的:(python,人工智能)

  • Python 应用打包成 APK【全流程】 今晚务必早点睡 Python运维python开发语言apk
    将Python应用打包成APK。文章目录步骤1:安装Buildozer和其依赖Linux(Ubuntu)环境下安装:步骤2:创建你的Python应用步骤3:配置Buildozer步骤4:打包成APK总结步骤1:安装Buildozer和其依赖首先确保你的系统中已安装Python和pip。接下来，我们需要安装Buildozer以及一些必要的系统依赖。Linux(Ubuntu)环境下安装:安装Pytho
  • 【Python】已完美解决：ERROR: Could not find a version that satisfies the requirement re 屿小夏 python开发语言
    个人简介：某不知名博主，致力于全栈领域的优质博客分享|用最优质的内容带来最舒适的阅读体验！文末获取免费IT学习资料！文末获取更多信息精彩专栏推荐订阅收藏专栏系列直达链接相关介绍书籍分享点我跳转书籍作为获取知识的重要途径，对于IT从业者来说更是不可或缺的资源。不定期更新IT图书，并在评论区抽取随机粉丝，书籍免费包邮到家AI前沿点我跳转探讨人工智能技术领域的最新发展和创新，涵盖机器学习、深度学习、自然
  • 基于Python的三种主流网络爬虫技术 吃肉肉335 python爬虫开发语言
    一、网络爬虫是什么网络爬虫，通常也被称为网络蜘蛛或网络机器人，是一种按照一定方法，获取网络各种信息的自动化脚本程序，也可以将其理解为一个在互联网上自动提取网页信息并进行解析抓取的程序。网络爬虫的功能不仅局限于复制网页内容、下载音视频文件，更包括自动化执行行为链以及模拟用户登录等复杂操作。在当前大数据背景下，无论是人工智能应用还是数据分析工作，均依赖于海量的数据支持。如果仅依赖人工采集这一种方式，不
  • python md5加密 计算机辅助工程 python前端javascript
    在Python中，可以使用hashlib模块来进行MD5加密。以下是一个简单的例子：importhashlibdefmd5_encryption(data):md5=hashlib.md5()#创建一个md5对象md5.update(data.encode('utf-8'))#使用utf-8编码数据returnmd5.hexdigest()#返回加密后的十六进制字符串#使用函数data="Hell
  • Python网络爬虫核心面试题 闲人编程 程序员面试python爬虫开发语言面试网络编程
    网络爬虫1.爬虫项目中如何处理请求失败的问题？2.解释HTTP协议中的持久连接和非持久连接。3.什么是HTTP的持久化Cookie和会话Cookie？4.如何在爬虫项目中检测并处理网络抖动和丢包？5.在爬虫项目中，如何使用HEAD请求提高效率？6.如何在爬虫项目中实现HTTP请求的限速？7.解释HTTP2相对于HTTP1.1的主要改进。8.如何在爬虫项目中模拟HTTP重试和重定向？9.什么是COR
  • Python全局解释器锁GIL与多线程 程序媛一枚~ PythonOpenCV图像处理Python进阶PythonOpenCVpython多线程全局解释器锁GIL
    Python中如果是I/O密集型的操作，用多线程（协程Asyncio、线程Threading），如果I/O操作很慢，需要很多任务/线程协同操作，用Asyncio，如果需要有限数量的任务/线程，那么使用多线程。如果是CPU密集型操作，用多进程（multeprocessing）。一、GILGIL（GlobalInterpreterLock，即全局解释器锁），Python实质上并不存在真正的多线程，只有
  • Python 面试时需要知道的 10 个问题及详解 迪小莫学AI python面试开发语言
    Python面试时需要知道的10个问题及详解在Python的面试中，考察的重点通常是基础知识、编程思维和实际应用能力。掌握Python的核心概念不仅能帮助你在面试中脱颖而出，还能让你在实际工作中高效编写代码。下面我们将详细解答10个Python面试中常见的问题，帮助你更加深入理解Python的基本特性和应用。1.Python中的全局、受保护和私有属性是什么？在Python中，属性的访问权限并不像J
  • python解奥赛运算题 Python数据分析与机器学习 python开发语言算法贪心算法
    有一个以文字代替数字的算术表达式如下图所示，已知4个替代数字的文字中没有重复，编写程序求出文字所替代的数字。按逻辑思维:如果3位数和3位数相加等于4位数，则“青”只能是1:“山”+“青”大于等于10，因此“山”只能是9，得出“龙”是0:个位的两个“山”相加，推得“外”等于8。而按计算思维，则注重于程序的实现，用穷举法设计嵌套的4层循环，把所有的数字都试一遍，找出4个数字不相互重复的合满足加法等式条
  • 《CPython Internals》阅读笔记：p285-p328 codists 读书笔记python
    《CPythonInternals》学习第15天，p285-p328总结，总计44页。一、技术总结1.shallowcomparisonp285,InObjectsobject.c,thebaseimplementationoftheobjecttypeiswritteninpureCcode.Therearesomeconcreteimplementationsofbasiclogic,like
  • 《CPython Internals》阅读笔记：p250-p284 codists 读书笔记python
    《CPythonInternals》学习第14天，250-p284总结，总计25页。一、技术总结介于我觉得作者写得乱七八糟的，读完我已经不想说话了，所以今日无技术总结。二、英语总结(生词：2)1.spawn(1)spawn:来自于词根expandere。(2)expandere:ex-(“out”)+pandere(“tospread”)spawn原来的意思是“spreadingoutoffish
  • 《CPython Internals》阅读笔记：p221-p231 codists 笔记
    《CPythonInternals》学习第12天，p221-p231总结，总计11页。一、技术总结无。二、英语总结(生词：2)1.atatimeidiom.separately(单独地)inthespecifiedgroups(一次)。示例：(1)Icanonlydoonethingatatim(我一次只能做一件事)。(2)Wecarriedtheboxestwoatatimeupthestair
  • 《CPython Internals》阅读笔记：p152-p176 codists 读书笔记python
    《CPythonInternals》学习第10天，p152-p176总结，总计25页。一、技术总结1.addinganitemtoalistmy_list=[]my_list.append(obj)上面的代码涉及两个指令：LOAD_FAST,LIST_APPEND。整章看下来这有这点算是可以记的了，其它的只感觉作者在零零碎碎的罗列内容。二、英语总结(生词：1)无。关于英语的注解同步更新汇总到htt
  • 《CPython Internals》阅读笔记：p1-p19 codists 笔记
    《CPythonInternals》学习第1天，p1-p19总结，总计19页。一、技术总结无。二、英语总结(生词：2)1.humblevshumbled(1)humble:humus(“earth”)adj.字面意思是“ontheground”,后面引申为“lowlyinkind,state,condition(卑微)”,“notproudthatyouareimportant(谦卑)”。(2)h
  • Linux离线安装Python chudu
    #下载openssl，如果不先安装openssl，后续pip、requests库使用会有很多问题wget--no-check-certificatehttps://www.openssl.org/source/openssl-1.1.1g.tar.gztar-zxvfopenssl-1.1.1g.tar.gzcdopenssl-1.1.1g./configshared--openssldir=/u
  • 深入解析：使用 Python 爬虫获取苏宁商品详情 数据小爬虫@ python爬虫开发语言
    在当今数字化时代，电商数据已成为市场分析、用户研究和商业决策的重要依据。苏宁易购作为国内知名的电商平台，其商品详情页包含了丰富的信息，如商品价格、描述、评价等。这些数据对于商家和市场研究者来说具有极高的价值。本文将详细介绍如何使用Python爬虫获取苏宁商品的详细信息，并提供完整的代码示例。一、爬虫简介爬虫是一种自动化程序，用于从互联网上抓取网页内容。Python因其简洁的语法和强大的库支持，成为
  • 【网络安全 | Python爬虫】URL、HTTP基础必知必会 秋说 爬虫http网络安全
    文章目录URL概念及组成结构HTTP概念简述浏览器接收资源HTTP协议的结构请求结构请求行请求头请求体请求差异及参数说明响应结构状态行响应头响应体推广URL概念及组成结构在开始爬虫的开发实战前，需要了解的是URL的概念及组成结构，这具有基础性和必要性。URL（UniformResourceLocator，统一资源定位符）是用于在互联网上定位和标识资源的字符串。它提供了一种标准的方式来指示资源的位置
  • 什么是网络爬虫？Python爬虫到底怎么学？ 糯米导航 文末下载资源python
    最近我在研究Python网络爬虫，发现这玩意儿真是有趣，干脆和大家聊聊我的心得吧！咱们都知道，网络上的信息多得就像大海里的水，而网络爬虫就像一个勤劳的小矿工，能帮我们从这片浩瀚的信息海洋中挖掘出需要的内容。接下来，我就带你们一步步看看该怎么用Python搞定网络爬虫。为啥选择Python写爬虫？说到Python，简直是写爬虫的最佳选择！它有许多现成的库，就像拥有了各种好用的工具，使得我们的工作变得
  • 《CPython Internals》阅读笔记：p329-p335 codists 读书笔记python
    《CPythonInternals》学习第16天，p329-p335总结，总计7页。一、技术总结1.debuggingp331,Therearetwotypesofdebugger,consoleandvisual——作者将debugger分为两类：(1)console：lldb(MAC系统使用),GDB(Linux系统使用))。(2)visual：VisualStudioDebugger,CLi
  • 商汤善惠获金沙江创投领投A轮融资，聚焦零售AI业务 TMT星球 人工智能人工智能零售大数据
    1月20日，商汤善惠宣布完成A轮融资，本轮融资由金沙江创投数千万元领投，微木资本、嘉实基金和金弘基金等知名资管平台和产业资本数千万元跟投，鞍羽资本担任长期财务顾问。此次融资将重点投向零售AI算法研发创新、海外市场拓展战略方向，助力公司全球化布局迈入新阶段。商汤善惠脱胎于全球领先的AI人工智能软件公司商汤集团，聚焦零售领域的商品识别算法与智能运营提效算法，目前，公司已推出引领行业的新一代无人零售智能
  • 大模型：LangChain技术讲解 玉成226 【大模型】langchain
    一、什么是LangChain1、介绍LangChain是一个用于开发由大型语言模型提供支持的Python框架。它提供了一系列工具和组件，帮助我们将语言模型集成到自己的应用程序中。有了它之后，我们可以更轻松地实现对话系统、文本生成、文本分类、问答系统等功能。2、LangChain官网文档官网：https://python.langchain.com/docs/introduction/3、LangC
  • 科技赋能，商贸物流新速度 —— 智慧供应链商城加速企业成长 呆码科技 科技
    科技赋能，商贸物流新速度——智慧供应链商城加速企业成长随着科技的飞速发展，AI（人工智能）、大数据、物联网等先进技术正深刻重塑着商贸物流行业，推动其向更高效、更智能、更环保的方向迈进。这些技术的应用不仅提升了物流效率，降低了运营成本，还增强了供应链的透明度和可控性，为商贸物流行业带来了前所未有的变革。智慧供应链商城是一个集成了AI、大数据、物联网等先进技术的综合服务平台，旨在通过科技手段提升物流效
  • 数据挖掘：定义、挑战与应用 黑色叉腰丶大魔王 数据挖掘人工智能
    一、数据挖掘的定义（一）概念阐述数据挖掘是从大量的、不完全的、有噪声的、模糊的、随机的数据中，提取隐含在其中的、人们事先不知道的、但又是潜在有用的信息和知识的过程。它融合了数据库技术、统计学、机器学习、人工智能等多学科的理论和方法，旨在通过对数据的深入分析和处理，发现有价值的模式、关联、趋势等，从而为决策提供支持。（二）与相关概念的区别与联系数据库管理：数据库管理侧重于数据的存储、组织、检索和维护
  • lisp不是函授型语言_LISP语言 sunlee0520 lisp不是函授型语言
    [拼音]：LISPyuyan[外文]：LISP为非数值符号运算而设计的表处理语言。LISP是英文LISTPROCESSING(表处理)的缩写。LISP语言是1960年J.麦卡锡在递归函数论基础上首先设计出来的。LISP语言的形式化程度高，表达力强，适合于描述各种知识和编写问题求解的程序，因此一直是用来研究人工智能的一种基本语言。自然语言中词可以认为是能单独用来构成句子的最小单元，由词可以构成词组，
  • lisp语言与python_Lisp 语言优点那么多，为什么国内很少运用？ 特殊后勤小干事 lisp语言与python
    为什么Lisp没有流行起来本文探讨的是为什么Lisp语言不再被广泛使用的。很久以前，这种语言站在计算机科学研究的前沿，特别是人工智能的研究方面。现在，它很少被用到，这一切并不是因为古老,类似古老的语言却被广泛应用.其他类似的古老的语言有FORTRAN,COBOL,LISP,BASIC,和ALGOL家族,这些语言的唯一不同之处在于,他们为谁设计,FORTRAN是为科学家和工程师设计的,他们在计算机上
  • CS书籍、代码资源下载网址 sun_kang CSGuide&&Linkslinuxjavatutorialsdebiandocumentationlinux内核
    CS网址收藏啄木鸟Pythonic开源社区：http://www.woodpecker.org.cn/中国IT实验室（游戏开发）：http://game.chinaitlab.com/游戏开发资源网：http://www.gameres.com/国内游戏开发论坛：http://bbs.gameres.com/国外的游戏开发站：http://www.gamedev.net/《问道》游戏开发论坛：ht
  • 探索ChatLiteLLM与Langchain的集成应用 safHTEAHE langchainpython
    在现代AI应用开发中，简化调用多种语言模型的过程显得尤为重要。ChatLiteLLM库应运而生，它为开发者提供了调用如Anthropic、Azure、Huggingface、Replicate等模型的简便方法。这篇文章将带你深入了解如何使用Langchain与LiteLLMI/O库协同工作，实现高效的语言模型交互。1.技术背景介绍随着人工智能技术的飞速发展，多种语言模型应用于不同场景。集成这些模型
  • Python爬虫项目合集：200个Python爬虫项目带你从入门到精通 人工智能_SYBH 爬虫试读2025年爬虫百篇实战宝典:从入门到精通python爬虫数据分析信息可视化爬虫项目大全Python爬虫项目合集爬虫从入门到精通项目
    适合人群无论你是刚接触编程的初学者，还是已经掌握一定Python基础并希望深入了解网络数据采集的开发者，这个专栏都将为你提供系统化的学习路径。通过循序渐进的理论讲解、代码实例和实践项目，你将获得扎实的爬虫开发技能，适应不同场景下的数据采集需求。专栏特色从基础到高级，内容体系全面专栏内容从爬虫的基础知识与工作原理开始讲解，逐渐覆盖静态网页、动态网页、API数据爬取等实用技术。后续还将深入解析反爬机制
  • FunASR语言识别的环境安装、推理 山山而川_R FunASRr语言xcode开发语言
    目录一、环境配置1、创建虚拟环境2、安装环境及pytorch官网：pytorch下载地址3、安装funasr之前，确保已经安装了下面依赖环境:python代码调用（推荐）4、模型下载5、启动funasr服务二、客户端连接2.1html连接三、推理识别模型1、实时语音识别2、非实时语音识别一、环境配置源码地址：FunASRFunASR/README_zh.mdatmain·alibaba-damo-
  • 安装线程自由（无GIL锁）Python及Pytorch方法 为什么每天的风都这么大 pythonpytorch开发语言
    参考资料：https://py-free-threading.github.io/installing_cpython/https://github.com/pytorch/pytorch/issues/130249https://download.pytorch.org/whl/torch/截至当前（2025.1.22），最新的线程自由Python版本为13.1，因此下面的方法适用于该版本安装P
  • 如何使用python技术爬取下载百度文库文档？ 大懒猫软件 python百度开发语言
    使用Python爬取百度文库文档需要通过分析网页结构和接口请求来实现。以下是一个基于搜索结果的实现方法，适用于爬取百度文库中的文档内容：第一部分：获取百度文库文档实现步骤获取文档ID和基本信息通过文档的URL获取文档ID，并解析页面内容以获取文档的类型、标题等信息。请求文档信息接口使用getdocinfo接口获取文档的页数、MD5校验码等参数。请求文档正文内容根据文档类型（如txt、doc等），请
  • 312个免费高速HTTP代理IP（能隐藏自己真实IP地址） yangshangchuan 高速免费superwordHTTP代理
        124.88.67.20:843 190.36.223.93:8080 117.147.221.38:8123 122.228.92.103:3128 183.247.211.159:8123 124.88.67.35:81 112.18.51.167:8123 218.28.96.39:3128 49.94.160.198:3128 183.20
  • pull解析和json编码 百合不是茶 androidpull解析json
    n.json文件: [{name:java,lan:c++,age:17},{name:android,lan:java,age:8}]   pull.xml文件 <?xml version="1.0" encoding="utf-8"?> <stu>     <name>java
  • [能源与矿产]石油与地球生态系统 comsci 能源
          按照苏联的科学界的说法,石油并非是远古的生物残骸的演变产物,而是一种可以由某些特殊地质结构和物理条件生产出来的东西,也就是说,石油是可以自增长的....       那么我们做一个猜想: 石油好像是地球的体液,我们地球具有自动产生石油的某种机制,只要我们不过量开采石油,并保护好
  • 类与对象浅谈 沐刃青蛟 java基础
           类，字面理解，便是同一种事物的总称，比如人类，是对世界上所有人的一个总称。而对象，便是类的具体化，实例化，是一个具体事物，比如张飞这个人，就是人类的一个对象。但要注意的是：张飞这个人是对象，而不是张飞，张飞只是他这个人的名字，是他的属性而已。而一个类中包含了属性和方法这两兄弟，他们分别用来描述对象的行为和性质（感觉应该是
  • 新站开始被收录后，我们应该做什么？ IT独行者 PHPseo
    新站开始被收录后，我们应该做什么？         百度终于开始收录自己的网站了，作为站长，你是不是觉得那一刻很有成就感呢，同时，你是不是又很茫然，不知道下一步该做什么了？至少我当初就是这样，在这里和大家一份分享一下新站收录后，我们要做哪些工作。       至于如何让百度快速收录自己的网站，可以参考我之前的帖子《新站让百
  • oracle 连接碰到的问题 文强chu oracle
    Unable to find a java Virtual Machine－－安装64位版Oracle11gR2后无法启动SQLDeveloper的解决方案 作者：草根IT网 来源：未知 人气：813标签： 导读：安装64位版Oracle11gR2后发现启动SQLDeveloper时弹出配置java.exe的路径，找到Oracle自带java.exe后产生的路径“C:\app\用户名\prod
  • Swing中按ctrl键同时移动鼠标拖动组件（类中多借口共享同一数据） 小桔子 java继承swing接口监听
            都知道java中类只能单继承，但可以实现多个接口，但我发现实现多个接口之后，多个接口却不能共享同一个数据，应用开发中想实现：当用户按着ctrl键时，可以用鼠标点击拖动组件，比如说文本框。 编写一个监听实现KeyListener,NouseListener,MouseMotionListener三个接口，重写方法。定义一个全局变量boolea
  • linux常用的命令 aichenglong linux常用命令
    1 startx切换到图形化界面 2 man命令:查看帮助信息 man 需要查看的命令,man命令提供了大量的帮助信息,一般可以分成4个部分 name:对命令的简单说明 synopsis:命令的使用格式说明 description:命令的详细说明信息 options:命令的各项说明 3 date:显示时间 语法：date [OPTION]... [+FORMAT]
  • eclipse内存优化 AILIKES javaeclipsejvmjdk
    一 基本说明      在JVM中，总体上分2块内存区,默认空余堆内存小于 40%时，JVM就会增大堆直到-Xmx的最大限制；空余堆内存大于70%时，JVM会减少堆直到-Xms的最小限制。     1)堆内存(Heap memory):堆是运行时数据区域，所有类实例和数组的内存均从此处分配,是Java代码可及的内存，是留给开发人
  • 关键字的使用探讨 百合不是茶 关键字
    //关键字的使用探讨/*访问关键词private 只能在本类中访问public 只能在本工程中访问protected 只能在包中和子类中访问默认的 只能在包中访问*//*final   类 方法 变量 final 类 不能被继承 final 方法 不能被子类覆盖，但可以继承 final 变量 只能有一次赋值，赋值后不能改变 final 不能用来修饰构造方法*///this()
  • JS中定义对象的几种方式 bijian1013 js
        1. 基于已有对象扩充其对象和方法(只适合于临时的生成一个对象)： <html> <head> <title>基于已有对象扩充其对象和方法(只适合于临时的生成一个对象)</title> </head> <script> var obj = new Object();
  • 表驱动法实例 bijian1013 java表驱动法TDD
    获得月的天数是典型的直接访问驱动表方式的实例，下面我们来展示一下： MonthDaysTest.java package com.study.test; import org.junit.Assert; import org.junit.Test; import com.study.MonthDays; public class MonthDaysTest { @T
  • LInux启停重启常用服务器的脚本 bit1129 linux
    启动，停止和重启常用服务器的Bash脚本，对于每个服务器，需要根据实际的安装路径做相应的修改   #! /bin/bash Servers=(Apache2, Nginx, Resin, Tomcat, Couchbase, SVN, ActiveMQ, Mongo); Ops=(Start, Stop, Restart); currentDir=$(pwd); echo
  • 【HBase六】REST操作HBase bit1129 hbase
    HBase提供了REST风格的服务方便查看HBase集群的信息，以及执行增删改查操作   1. 启动和停止HBase REST 服务 1.1 启动REST服务 前台启动（默认端口号8080） [hadoop@hadoop bin]$ ./hbase rest start   后台启动 hbase-daemon.sh start rest   启动时指定
  • 大话zabbix 3.0设计假设 ronin47
    What’s new in Zabbix 2.0? 去年开始使用Zabbix的时候，是1.8.X的版本，今年Zabbix已经跨入了2.0的时代。看了2.0的release notes，和performance相关的有下面几个：          :: Performance improvements::Trigger related da
  • http错误码大全 byalias http协议javaweb
    响应码由三位十进制数字组成，它们出现在由HTTP服务器发送的响应的第一行。 响应码分五种类型，由它们的第一位数字表示： 1）1xx：信息，请求收到，继续处理 2）2xx：成功，行为被成功地接受、理解和采纳 3）3xx：重定向，为了完成请求，必须进一步执行的动作 4）4xx：客户端错误，请求包含语法错误或者请求无法实现 5）5xx：服务器错误，服务器不能实现一种明显无效的请求
  • J2EE设计模式-Intercepting Filter bylijinnan java设计模式数据结构
    Intercepting Filter类似于职责链模式 有两种实现 其中一种是Filter之间没有联系，全部Filter都存放在FilterChain中，由FilterChain来有序或无序地把把所有Filter调用一遍。没有用到链表这种数据结构。示例如下： package com.ljn.filter.custom; import java.util.ArrayList;
  • 修改jboss端口 chicony jboss
    修改jboss端口   %JBOSS_HOME%\server\{服务实例名}\conf\bindingservice.beans\META-INF\bindings-jboss-beans.xml   中找到 <!-- The ports-default bindings are obtained by taking the base bindin
  • c++ 用类模版实现数组类 CrazyMizzz C++
    最近c++学到数组类，写了代码将他实现，基本具有vector类的功能 #include<iostream> #include<string> #include<cassert> using namespace std; template<class T> class Array { public: //构造函数
  • hadoop dfs.datanode.du.reserved 预留空间配置方法 daizj hadoop预留空间
    对于datanode配置预留空间的方法 为：在hdfs-site.xml添加如下配置  <property>     <name>dfs.datanode.du.reserved</name>     <value>10737418240</value>    
  • mysql远程访问的设置 dcj3sjt126com mysql防火墙
    第一步: 激活网络设置 你需要编辑mysql配置文件my.cnf. 通常状况，my.cnf放置于在以下目录： /etc/mysql/my.cnf (Debian linux) /etc/my.cnf （Red Hat Linux/Fedora Linux) /var/db/mysql/my.cnf (FreeBSD) 然后用vi编辑my.cnf，修改内容从以下行： [mysqld] 你所需要: 1
  • ios 使用特定的popToViewController返回到相应的Controller dcj3sjt126com controller
    1、取navigationCtroller中的Controllers NSArray * ctrlArray = self.navigationController.viewControllers; 2、取出后，执行， [self.navigationController popToViewController:[ctrlArray objectAtIndex:0] animated:YES
  • Linux正则表达式和通配符的区别 eksliang 正则表达式通配符和正则表达式的区别通配符
    转载请出自出处：http://eksliang.iteye.com/blog/1976579 首先得明白二者是截然不同的 通配符只能用在shell命令中,用来处理字符串的的匹配。 判断一个命令是否为bash shell(linux 默认的shell)的内置命令 type -t commad 返回结果含义 file  表示为外部命令 alias  表示该
  • Ubuntu Mysql Install and CONF gengzg Install
    http://www.navicat.com.cn/download/navicat-for-mysql Step1: 下载Navicat ，网址：http://www.navicat.com/en/download/download.html Step2：进入下载目录，解压压缩包：tar -zxvf navicat11_mysql_en.tar.gz
  • 批处理，删除文件bat huqiji windowsdos
    @echo off ::演示：删除指定路径下指定天数之前（以文件名中包含的日期字符串为准）的文件。 ::如果演示结果无误，把del前面的echo去掉，即可实现真正删除。 ::本例假设文件名中包含的日期字符串（比如：bak-2009-12-25.log） rem 指定待删除文件的存放路径 set SrcDir=C:/Test/BatHome rem 指定天数 set DaysAgo=1
  • 跨浏览器兼容的HTML5视频音频播放器 天梯梦 html5
    HTML5的video和audio标签是用来在网页中加入视频和音频的标签，在支持html5的浏览器中不需要预先加载Adobe Flash浏览器插件就能轻松快速的播放视频和音频文件。而html5media.js可以在不支持html5的浏览器上使video和audio标签生效。   How to enable <video> and <audio> tags in
  • Bundle自定义数据传递 hm4123660 androidSerializable自定义数据传递BundleParcelable
          我们都知道Bundle可能过put****()方法添加各种基本类型的数据，Intent也可以通过putExtras(Bundle)将数据添加进去，然后通过startActivity()跳到下一下Activity的时候就把数据也传到下一个Activity了。如传递一个字符串到下一个Activity   把数据放到Intent
  • C＃：异步编程和线程的使用（.NET 4.5 ） powertoolsteam .net线程C#异步编程
    异步编程和线程处理是并发或并行编程非常重要的功能特征。为了实现异步编程，可使用线程也可以不用。将异步与线程同时讲，将有助于我们更好的理解它们的特征。 本文中涉及关键知识点 1. 异步编程 2. 线程的使用 3. 基于任务的异步模式 4. 并行编程 5. 总结 异步编程 什么是异步操作？异步操作是指某些操作能够独立运行，不依赖主流程或主其他处理流程。通常情况下，C＃程序
  • spark 查看 job history 日志 Stark_Summer 日志sparkhistoryjob
    SPARK_HOME/conf 下: spark-defaults.conf 增加如下内容 spark.eventLog.enabled true spark.eventLog.dir hdfs://master:8020/var/log/spark spark.eventLog.compress true spark-env.sh 增加如下内容 export SP
  • SSH框架搭建 wangxiukai2015eye springHibernatestruts
    MyEclipse搭建SSH框架 Struts Spring Hibernate 1、new一个web project。 2、右键项目，为项目添加Struts支持。    选择Struts2 Core Libraries -<MyEclipes-Library>      点击Finish。src目录下多了struts
按字母分类： ABCDEFGHIJKLMNOPQRSTUVWXYZ其他