1、环境：Win7x64、python3.7x64、tensorflow1.14、CPU i5-9400F
2、
3、
3.1、cifar10，没有数据，全新下载，下到默认目录（C:\Users\Administrator\tensorflow_datasets），全过程 控制台输出：（20190903）
"C:\Program Files\Python37\python.exe" E:/Project_Py37/cifar/cifar10/cifar10_input.py WARNING: Logging before flag parsing goes to stderr. W0903 08:30:02.193804 4392 lazy_loader.py:50] The TensorFlow contrib module will not be included in TensorFlow 2.0. For more information, please see: * https://github.com/tensorflow/community/blob/master/rfcs/20180907-contrib-sunset.md * https://github.com/tensorflow/addons * https://github.com/tensorflow/io (for I/O related ops) If you depend on functionality not listed there, please file an issue. Downloading and preparing dataset cifar10 (162.17 MiB) to C:\Users\Administrator\tensorflow_datasets\cifar10\1.0.2... Dl Completed...: 0 url [00:00, ? url/s] Dl Size...: 0 MiB [00:00, ? MiB/s] Dl Completed...: 0%| | 0/1 [00:00
3.1.1、当 "C:\Users\Administrator\tensorflow_datasets"中 已经有了 tfrecord文件之后，再次下载的话，输出 类似如下信息：（只要 版本对 就不需要重新下载了）
Downloading and preparing dataset cifar10 (162.17 MiB) to C:\Users\Administrator\tensorflow_datasets\cifar10\1.0.2... Dataset cifar10 downloaded and prepared to C:\Users\Administrator\tensorflow_datasets\cifar10\1.0.2. Subsequent calls will reuse this data.
4、
5、代码 输出
5.1、ZC：代码里面的 注释我去掉了，这样贴的东西可以少一点...
def builder(name, **builder_init_kwargs): name, builder_kwargs = _dataset_name_and_kwargs_from_name_str(name) builder_kwargs.update(builder_init_kwargs) if name in _ABSTRACT_DATASET_REGISTRY: raise DatasetNotFoundError(name, is_abstract=True) if name in _IN_DEVELOPMENT_REGISTRY: raise DatasetNotFoundError(name, in_development=True) if name not in _DATASET_REGISTRY: raise DatasetNotFoundError(name) try: return _DATASET_REGISTRY[name](**builder_kwargs) except BaseException: logging.error("Failed to construct dataset %s", name) raise
其中 打印出 _DATASET_REGISTRY、name、builder_kwargs 的信息，如下：
_DATASET_REGISTRY :
{
'dummy_dataset_shared_generator': ,
'dummy_mnist': ,
'groove': ,
'nsynth': ,
'abstract_reasoning': ,
'aflw2k3d': ,
'bigearthnet': ,
'mnist': ,
'fashion_mnist': ,
'kmnist': ,
'emnist': ,
'binarized_mnist': ,
'binary_alpha_digits': ,
'caltech101': ,
'caltech_birds2010': ,
'caltech_birds2011': ,
'cats_vs_dogs': ,
'curated_breast_imaging_ddsm': ,
'celeb_a': ,
'celeb_a_hq': ,
'chexpert': ,
'cifar10': ,
'cifar100': ,
'cifar10_corrupted': ,
'clevr': ,
'coco': ,
'coco2014': ,
'coil100': ,
'colorectal_histology': ,
'colorectal_histology_large': ,
'cycle_gan': ,
'deep_weeds': ,
'diabetic_retinopathy_detection': ,
'downsampled_imagenet': ,
'dsprites': ,
'dtd': ,
'eurosat': ,
'tf_flowers': ,
'food101': ,
'horses_or_humans': ,
'image_label_folder': ,
'imagenet2012': ,
'imagenet2012_corrupted': ,
'kitti': ,
'lfw': ,
'lsun': ,
'mnist_corrupted': ,
'omniglot': ,
'open_images_v4': ,
'oxford_flowers102': ,
'oxford_iiit_pet': ,
'patch_camelyon': ,
'pet_finder': ,
'quickdraw_bitmap': ,
'resisc45': ,
'rock_paper_scissors': ,
'scene_parse150': ,
'shapes3d': ,
'smallnorb': ,
'so2sat': ,
'stanford_dogs': ,
'stanford_online_products': ,
'sun397': ,
'svhn_cropped': ,
'uc_merced': ,
'visual_domain_decathlon': ,
'voc2007': ,
'amazon_us_reviews': ,
'higgs': ,
'iris': ,
'rock_you': ,
'titanic': ,
'cnn_dailymail': ,
'definite_pronoun_resolution': ,
'gap': ,
'glue': ,
'imdb_reviews': ,
'lm1b': ,
'multi_nli': ,
'snli': ,
'squad': ,
'super_glue': ,
'trivia_qa': ,
'wikipedia': ,
'xnli': ,
'flores': ,
'para_crawl': ,
'ted_hrlr_translate': ,
'ted_multi_translate': ,
'wmt_translate': ,
'wmt14_translate': ,
'wmt15_translate': ,
'wmt16_translate': ,
'wmt17_translate': ,
'wmt18_translate': ,
'wmt19_translate': ,
'wmt_t2t_translate': ,
'bair_robot_pushing_small': ,
'moving_mnist': ,
'starcraft_video': ,
'ucf101': }
name : cifar10
builder_kwargs : {'data_dir': '~\\tensorflow_datasets'}
6、
6.1、cifar100，没有数据，全新下载，下到默认目录（C:\Users\Administrator\tensorflow_datasets），全过程 控制台输出：（20190904）
"C:\Program Files\Python37\python.exe" E:/Project_Py37/cifar/cifar10/cifar10_input.py WARNING: Logging before flag parsing goes to stderr. W0904 08:07:41.751157 6092 lazy_loader.py:50] The TensorFlow contrib module will not be included in TensorFlow 2.0. For more information, please see: * https://github.com/tensorflow/community/blob/master/rfcs/20180907-contrib-sunset.md * https://github.com/tensorflow/addons * https://github.com/tensorflow/io (for I/O related ops) If you depend on functionality not listed there, please file an issue. W0904 08:07:42.109957 6092 deprecation_wrapper.py:119] From E:/Project_Py37/cifar/cifar10/cifar10_input.py:117: The name tf.Session is deprecated. Please use tf.compat.v1.Session instead. 2019-09-04 08:07:42.109957: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 W0904 08:07:42.109957 6092 deprecation_wrapper.py:119] From E:/Project_Py37/cifar/cifar10/cifar10_input.py:118: The name tf.global_variables_initializer is deprecated. Please use tf.compat.v1.global_variables_initializer instead. Downloading and preparing dataset cifar100 (160.71 MiB) to C:\Users\Administrator\tensorflow_datasets\cifar100\1.3.1... Dl Completed...: 0 url [00:00, ? url/s] Dl Size...: 0 MiB [00:00, ? MiB/s] Dl Completed...: 0%| | 0/1 [00:00
7、
8、cifar10 训练过程。20190903下班开始，到20190904上班，貌似 这个训练的速度很慢啊... 我没改训练相关的代码，是不是有哪些地方可以改进？
ZC：CPU占用 基本保持 99%左右，这是我 自己手动停止的程序
ZC：看到 下面 每10step，差不多就要耗时4+分钟，代码中 max_steps的值为100000，那么就要 10000个4分钟（方便点按4分钟计算）-->40000分钟--> 666.666667小时--> 27.777778天，那就要 差不多 一个月了... 处理方法？升级机子？用GPU？修改训练代码（还要保证准确率啊）？
"C:\Program Files\Python37\python.exe" E:/Project_Py37/cifar/cifar10/cifar10_train.py WARNING: Logging before flag parsing goes to stderr. W0903 17:01:04.763609 3520 lazy_loader.py:50] The TensorFlow contrib module will not be included in TensorFlow 2.0. For more information, please see: * https://github.com/tensorflow/community/blob/master/rfcs/20180907-contrib-sunset.md * https://github.com/tensorflow/addons * https://github.com/tensorflow/io (for I/O related ops) If you depend on functionality not listed there, please file an issue. W0903 17:01:05.115609 3520 deprecation_wrapper.py:119] From E:/Project_Py37/cifar/cifar10/cifar10_train.py:127: The name tf.app.run is deprecated. Please use tf.compat.v1.app.run instead. W0903 17:01:05.116609 3520 deprecation_wrapper.py:119] From E:/Project_Py37/cifar/cifar10/cifar10_train.py:120: The name tf.gfile.Exists is deprecated. Please use tf.io.gfile.exists instead. W0903 17:01:05.116609 3520 deprecation_wrapper.py:119] From E:/Project_Py37/cifar/cifar10/cifar10_train.py:121: The name tf.gfile.DeleteRecursively is deprecated. Please use tf.io.gfile.rmtree instead. W0903 17:01:05.116609 3520 deprecation_wrapper.py:119] From E:/Project_Py37/cifar/cifar10/cifar10_train.py:122: The name tf.gfile.MakeDirs is deprecated. Please use tf.io.gfile.makedirs instead. W0903 17:01:05.117609 3520 deprecation_wrapper.py:119] From E:/Project_Py37/cifar/cifar10/cifar10_train.py:62: The name tf.train.get_or_create_global_step is deprecated. Please use tf.compat.v1.train.get_or_create_global_step instead. main(1) : ./tmp/cifar10_train I0903 17:01:05.120609 3520 dataset_builder.py:184] Overwrite dataset info from restored data version. I0903 17:01:05.122608 3520 dataset_builder.py:253] Reusing dataset cifar10 (C:\Users\Administrator\tensorflow_datasets\cifar10\1.0.2) I0903 17:01:05.122608 3520 dataset_builder.py:399] Constructing tf.data.Dataset for split train, from C:\Users\Administrator\tensorflow_datasets\cifar10\1.0.2 W0903 17:01:05.122608 3520 dataset_builder.py:439] Warning: Setting shuffle_files=True because split=TRAIN and shuffle_files=None. This behavior will be deprecated on 2019-08-06, at which point shuffle_files=False will be the default for all splits. W0903 17:01:05.187609 3520 deprecation_wrapper.py:119] From E:\Project_Py37\cifar\cifar10\cifar10_input.py:65: The name tf.random_crop is deprecated. Please use tf.image.random_crop instead. W0903 17:01:05.219609 3520 deprecation.py:323] From C:\Program Files\Python37\lib\site-packages\tensorflow\python\ops\image_ops_impl.py:1514: div (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version. Instructions for updating: Deprecated in favor of operator or tf.math.divide. W0903 17:01:05.228609 3520 deprecation.py:323] From E:\Project_Py37\cifar\cifar10\cifar10_input.py:46: DatasetV1.make_one_shot_iterator (from tensorflow.python.data.ops.dataset_ops) is deprecated and will be removed in a future version. Instructions for updating: Use `for ... in dataset:` to iterate over a dataset. If using `tf.estimator`, return the `Dataset` object directly from your input function. As a last resort, you can use `tf.compat.v1.data.make_one_shot_iterator(dataset)`. W0903 17:01:05.240608 3520 deprecation_wrapper.py:119] From E:\Project_Py37\cifar\cifar10\cifar10_input.py:49: The name tf.summary.image is deprecated. Please use tf.compat.v1.summary.image instead. W0903 17:01:05.242608 3520 deprecation_wrapper.py:119] From E:\Project_Py37\cifar\cifar10\cifar10.py:178: The name tf.variable_scope is deprecated. Please use tf.compat.v1.variable_scope instead. W0903 17:01:05.242608 3520 deprecation.py:506] From E:\Project_Py37\cifar\cifar10\cifar10.py:126: calling TruncatedNormal.__init__ (from tensorflow.python.ops.init_ops) with dtype is deprecated and will be removed in a future version. Instructions for updating: Call initializer instance with the dtype argument instead of passing it to the constructor W0903 17:01:05.242608 3520 deprecation_wrapper.py:119] From E:\Project_Py37\cifar\cifar10\cifar10.py:102: The name tf.get_variable is deprecated. Please use tf.compat.v1.get_variable instead. W0903 17:01:05.250608 3520 deprecation_wrapper.py:119] From E:\Project_Py37\cifar\cifar10\cifar10.py:85: The name tf.summary.histogram is deprecated. Please use tf.compat.v1.summary.histogram instead. W0903 17:01:05.251609 3520 deprecation_wrapper.py:119] From E:\Project_Py37\cifar\cifar10\cifar10.py:86: The name tf.summary.scalar is deprecated. Please use tf.compat.v1.summary.scalar instead. W0903 17:01:05.262609 3520 deprecation_wrapper.py:119] From E:\Project_Py37\cifar\cifar10\cifar10.py:190: The name tf.nn.max_pool is deprecated. Please use tf.nn.max_pool2d instead. W0903 17:01:05.293608 3520 deprecation_wrapper.py:119] From E:\Project_Py37\cifar\cifar10\cifar10.py:129: The name tf.add_to_collection is deprecated. Please use tf.compat.v1.add_to_collection instead. W0903 17:01:05.350608 3520 deprecation_wrapper.py:119] From E:\Project_Py37\cifar\cifar10\cifar10.py:270: The name tf.get_collection is deprecated. Please use tf.compat.v1.get_collection instead. W0903 17:01:05.350608 3520 deprecation_wrapper.py:119] From E:\Project_Py37\cifar\cifar10\cifar10.py:318: The name tf.train.exponential_decay is deprecated. Please use tf.compat.v1.train.exponential_decay instead. I0903 17:01:05.373208 3520 summary_op_util.py:66] Summary name local3/weight_loss (raw) is illegal; using local3/weight_loss__raw_ instead. I0903 17:01:05.373208 3520 summary_op_util.py:66] Summary name local4/weight_loss (raw) is illegal; using local4/weight_loss__raw_ instead. I0903 17:01:05.373208 3520 summary_op_util.py:66] Summary name cross_entropy (raw) is illegal; using cross_entropy__raw_ instead. I0903 17:01:05.373208 3520 summary_op_util.py:66] Summary name total_loss (raw) is illegal; using total_loss__raw_ instead. W0903 17:01:05.373208 3520 deprecation_wrapper.py:119] From E:\Project_Py37\cifar\cifar10\cifar10.py:330: The name tf.train.GradientDescentOptimizer is deprecated. Please use tf.compat.v1.train.GradientDescentOptimizer instead. W0903 17:01:05.467408 3520 deprecation.py:323] From C:\Program Files\Python37\lib\site-packages\tensorflow\python\training\moving_averages.py:433: Variable.initialized_value (from tensorflow.python.ops.variables) is deprecated and will be removed in a future version. Instructions for updating: Use Variable.read_value. Variables in 2.X are initialized automatically both in eager and graph (inside tf.defun) contexts. W0903 17:01:05.611208 3520 deprecation_wrapper.py:119] From E:/Project_Py37/cifar/cifar10/cifar10_train.py:81: The name tf.train.SessionRunHook is deprecated. Please use tf.estimator.SessionRunHook instead. W0903 17:01:05.611208 3520 deprecation_wrapper.py:119] From E:/Project_Py37/cifar/cifar10/cifar10_train.py:107: The name tf.train.MonitoredTrainingSession is deprecated. Please use tf.compat.v1.train.MonitoredTrainingSession instead. I0903 17:01:05.611208 3520 basic_session_run_hooks.py:541] Create CheckpointSaverHook. W0903 17:01:05.810209 3520 deprecation.py:323] From C:\Program Files\Python37\lib\site-packages\tensorflow\python\ops\array_ops.py:1354: add_dispatch_support..wrapper (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version. Instructions for updating: Use tf.where in 2.0, which has the same broadcast rule as np.where I0903 17:01:05.862209 3520 monitored_session.py:240] Graph was finalized. 2019-09-03 17:01:05.863209: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 I0903 17:01:05.951209 3520 session_manager.py:500] Running local_init_op. I0903 17:01:05.961209 3520 session_manager.py:502] Done running local_init_op. I0903 17:01:06.295208 3520 basic_session_run_hooks.py:606] Saving checkpoints for 0 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 17:01:33.559809: step 0, loss = 4.68 (46.1 examples/sec; 2.777 sec/batch) 2019-09-03 17:06:00.632410: step 10, loss = 4.59 (4.8 examples/sec; 26.707 sec/batch) 2019-09-03 17:10:33.134212: step 20, loss = 4.48 (4.7 examples/sec; 27.250 sec/batch) I0903 17:11:28.001412 3520 basic_session_run_hooks.py:606] Saving checkpoints for 23 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 17:15:07.789814: step 30, loss = 4.39 (4.7 examples/sec; 27.466 sec/batch) 2019-09-03 17:19:44.396016: step 40, loss = 4.20 (4.6 examples/sec; 27.661 sec/batch) I0903 17:21:34.979816 3520 basic_session_run_hooks.py:606] Saving checkpoints for 45 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 17:24:21.915417: step 50, loss = 4.28 (4.6 examples/sec; 27.752 sec/batch) 2019-09-03 17:29:00.172619: step 60, loss = 4.24 (4.6 examples/sec; 27.826 sec/batch) I0903 17:31:47.024620 3520 basic_session_run_hooks.py:606] Saving checkpoints for 67 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 17:33:39.251021: step 70, loss = 4.14 (4.6 examples/sec; 27.908 sec/batch) 2019-09-03 17:38:17.761823: step 80, loss = 4.18 (4.6 examples/sec; 27.851 sec/batch) I0903 17:42:01.403424 3520 basic_session_run_hooks.py:606] Saving checkpoints for 89 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 17:42:58.203025: step 90, loss = 4.08 (4.6 examples/sec; 28.044 sec/batch) I0903 17:47:38.394626 3520 basic_session_run_hooks.py:692] global_step/sec: 0.0361685 2019-09-03 17:47:38.831426: step 100, loss = 4.02 (4.6 examples/sec; 28.063 sec/batch) I0903 17:52:18.321028 3520 basic_session_run_hooks.py:606] Saving checkpoints for 111 into ./tmp/cifar10_train\model.ckpt. W0903 17:52:18.539428 3520 deprecation.py:323] From C:\Program Files\Python37\lib\site-packages\tensorflow\python\training\saver.py:960: remove_checkpoint (from tensorflow.python.training.checkpoint_management) is deprecated and will be removed in a future version. Instructions for updating: Use standard file APIs to delete files with this prefix. 2019-09-03 17:52:18.679828: step 110, loss = 4.06 (4.6 examples/sec; 27.985 sec/batch) 2019-09-03 17:56:58.434630: step 120, loss = 3.97 (4.6 examples/sec; 27.975 sec/batch) 2019-09-03 18:01:38.793432: step 130, loss = 4.00 (4.6 examples/sec; 28.036 sec/batch) I0903 18:02:35.000232 3520 basic_session_run_hooks.py:606] Saving checkpoints for 133 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 18:06:19.767034: step 140, loss = 3.96 (4.6 examples/sec; 28.097 sec/batch) 2019-09-03 18:11:00.177035: step 150, loss = 3.88 (4.6 examples/sec; 28.041 sec/batch) I0903 18:12:52.122636 3520 basic_session_run_hooks.py:606] Saving checkpoints for 155 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 18:15:40.680637: step 160, loss = 3.96 (4.6 examples/sec; 28.050 sec/batch) 2019-09-03 18:20:21.527439: step 170, loss = 3.96 (4.6 examples/sec; 28.085 sec/batch) I0903 18:23:09.913840 3520 basic_session_run_hooks.py:606] Saving checkpoints for 177 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 18:25:03.076241: step 180, loss = 3.81 (4.5 examples/sec; 28.155 sec/batch) 2019-09-03 18:29:44.063443: step 190, loss = 3.83 (4.6 examples/sec; 28.099 sec/batch) I0903 18:33:28.831244 3520 basic_session_run_hooks.py:606] Saving checkpoints for 199 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 18:34:25.820044: step 200, loss = 3.82 (4.5 examples/sec; 28.176 sec/batch) I0903 18:34:25.820044 3520 basic_session_run_hooks.py:692] global_step/sec: 0.0356196 2019-09-03 18:39:06.978846: step 210, loss = 3.74 (4.6 examples/sec; 28.116 sec/batch) I0903 18:43:47.950448 3520 basic_session_run_hooks.py:606] Saving checkpoints for 221 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 18:43:48.792848: step 220, loss = 3.73 (4.5 examples/sec; 28.181 sec/batch) 2019-09-03 18:48:29.982850: step 230, loss = 3.71 (4.6 examples/sec; 28.119 sec/batch) 2019-09-03 18:53:10.892052: step 240, loss = 3.67 (4.6 examples/sec; 28.091 sec/batch) I0903 18:54:07.052052 3520 basic_session_run_hooks.py:606] Saving checkpoints for 243 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 18:57:52.628053: step 250, loss = 3.77 (4.5 examples/sec; 28.174 sec/batch) 2019-09-03 19:02:33.374655: step 260, loss = 3.62 (4.6 examples/sec; 28.075 sec/batch) I0903 19:04:25.771056 3520 basic_session_run_hooks.py:606] Saving checkpoints for 265 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 19:07:15.031057: step 270, loss = 3.66 (4.5 examples/sec; 28.166 sec/batch) 2019-09-03 19:11:55.784259: step 280, loss = 3.72 (4.6 examples/sec; 28.075 sec/batch) I0903 19:14:44.248660 3520 basic_session_run_hooks.py:606] Saving checkpoints for 287 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 19:16:37.021061: step 290, loss = 3.68 (4.6 examples/sec; 28.124 sec/batch) I0903 19:21:17.789862 3520 basic_session_run_hooks.py:692] global_step/sec: 0.0355623 2019-09-03 19:21:18.304662: step 300, loss = 3.36 (4.6 examples/sec; 28.128 sec/batch) I0903 19:25:03.209864 3520 basic_session_run_hooks.py:606] Saving checkpoints for 309 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 19:25:59.697464: step 310, loss = 3.52 (4.5 examples/sec; 28.139 sec/batch) 2019-09-03 19:30:40.216666: step 320, loss = 3.55 (4.6 examples/sec; 28.052 sec/batch) I0903 19:35:20.828468 3520 basic_session_run_hooks.py:606] Saving checkpoints for 331 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 19:35:21.733268: step 330, loss = 3.53 (4.5 examples/sec; 28.152 sec/batch) 2019-09-03 19:40:01.659670: step 340, loss = 3.38 (4.6 examples/sec; 27.993 sec/batch) 2019-09-03 19:44:41.929271: step 350, loss = 3.37 (4.6 examples/sec; 28.027 sec/batch) I0903 19:45:37.933272 3520 basic_session_run_hooks.py:606] Saving checkpoints for 353 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 19:49:22.760473: step 360, loss = 3.36 (4.6 examples/sec; 28.083 sec/batch) 2019-09-03 19:54:03.201675: step 370, loss = 3.26 (4.6 examples/sec; 28.044 sec/batch) I0903 19:55:55.630876 3520 basic_session_run_hooks.py:606] Saving checkpoints for 375 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 19:58:44.859677: step 380, loss = 3.28 (4.5 examples/sec; 28.166 sec/batch) 2019-09-03 20:03:25.464679: step 390, loss = 3.30 (4.6 examples/sec; 28.061 sec/batch) I0903 20:06:13.765280 3520 basic_session_run_hooks.py:606] Saving checkpoints for 397 into ./tmp/cifar10_train\model.ckpt. I0903 20:08:06.756080 3520 basic_session_run_hooks.py:692] global_step/sec: 0.0356003 2019-09-03 20:08:06.756080: step 400, loss = 3.46 (4.6 examples/sec; 28.129 sec/batch) 2019-09-03 20:12:47.181682: step 410, loss = 3.34 (4.6 examples/sec; 28.043 sec/batch) I0903 20:16:31.416084 3520 basic_session_run_hooks.py:606] Saving checkpoints for 419 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 20:17:28.137684: step 420, loss = 3.11 (4.6 examples/sec; 28.096 sec/batch) 2019-09-03 20:22:08.376086: step 430, loss = 3.26 (4.6 examples/sec; 28.024 sec/batch) I0903 20:26:48.864088 3520 basic_session_run_hooks.py:606] Saving checkpoints for 441 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 20:26:49.737688: step 440, loss = 3.24 (4.5 examples/sec; 28.136 sec/batch) 2019-09-03 20:31:30.386689: step 450, loss = 3.16 (4.6 examples/sec; 28.065 sec/batch) 2019-09-03 20:36:10.937091: step 460, loss = 3.33 (4.6 examples/sec; 28.055 sec/batch) I0903 20:37:06.691492 3520 basic_session_run_hooks.py:606] Saving checkpoints for 463 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 20:40:51.159893: step 470, loss = 3.14 (4.6 examples/sec; 28.022 sec/batch) 2019-09-03 20:45:31.211095: step 480, loss = 3.37 (4.6 examples/sec; 28.005 sec/batch) I0903 20:47:22.829095 3520 basic_session_run_hooks.py:606] Saving checkpoints for 485 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 20:50:11.231097: step 490, loss = 3.01 (4.6 examples/sec; 28.002 sec/batch) I0903 20:54:51.282298 3520 basic_session_run_hooks.py:692] global_step/sec: 0.0356566 2019-09-03 20:54:51.812698: step 500, loss = 3.00 (4.6 examples/sec; 28.058 sec/batch) I0903 20:57:39.465899 3520 basic_session_run_hooks.py:606] Saving checkpoints for 507 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 20:59:31.489500: step 510, loss = 3.28 (4.6 examples/sec; 27.968 sec/batch) 2019-09-03 21:04:11.330902: step 520, loss = 3.46 (4.6 examples/sec; 27.984 sec/batch) I0903 21:07:54.941303 3520 basic_session_run_hooks.py:606] Saving checkpoints for 529 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 21:08:51.509904: step 530, loss = 2.99 (4.6 examples/sec; 28.018 sec/batch) 2019-09-03 21:13:30.999506: step 540, loss = 3.04 (4.6 examples/sec; 27.949 sec/batch) I0903 21:18:10.520307 3520 basic_session_run_hooks.py:606] Saving checkpoints for 551 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 21:18:11.347107: step 550, loss = 3.13 (4.6 examples/sec; 28.035 sec/batch) 2019-09-03 21:22:50.992709: step 560, loss = 3.13 (4.6 examples/sec; 27.965 sec/batch) 2019-09-03 21:27:30.435511: step 570, loss = 2.99 (4.6 examples/sec; 27.944 sec/batch) I0903 21:28:26.595511 3520 basic_session_run_hooks.py:606] Saving checkpoints for 573 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 21:32:11.034713: step 580, loss = 2.98 (4.6 examples/sec; 28.060 sec/batch) 2019-09-03 21:36:50.978715: step 590, loss = 3.24 (4.6 examples/sec; 27.994 sec/batch) I0903 21:38:42.877515 3520 basic_session_run_hooks.py:606] Saving checkpoints for 595 into ./tmp/cifar10_train\model.ckpt. I0903 21:41:31.529116 3520 basic_session_run_hooks.py:692] global_step/sec: 0.0357111 2019-09-03 21:41:31.575916: step 600, loss = 2.94 (4.6 examples/sec; 28.060 sec/batch) 2019-09-03 21:46:11.268318: step 610, loss = 2.89 (4.6 examples/sec; 27.969 sec/batch) I0903 21:48:59.233519 3520 basic_session_run_hooks.py:606] Saving checkpoints for 617 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 21:50:51.569120: step 620, loss = 2.95 (4.6 examples/sec; 28.030 sec/batch) 2019-09-03 21:55:31.043122: step 630, loss = 3.00 (4.6 examples/sec; 27.947 sec/batch) I0903 21:59:15.137123 3520 basic_session_run_hooks.py:606] Saving checkpoints for 639 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 22:00:11.858724: step 640, loss = 3.14 (4.6 examples/sec; 28.082 sec/batch) 2019-09-03 22:04:51.063925: step 650, loss = 3.06 (4.6 examples/sec; 27.921 sec/batch) I0903 22:09:30.773927 3520 basic_session_run_hooks.py:606] Saving checkpoints for 661 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 22:09:31.678727: step 660, loss = 2.79 (4.6 examples/sec; 28.061 sec/batch) 2019-09-03 22:14:11.261929: step 670, loss = 2.70 (4.6 examples/sec; 27.958 sec/batch) 2019-09-03 22:18:50.799331: step 680, loss = 2.83 (4.6 examples/sec; 27.954 sec/batch) I0903 22:19:46.943731 3520 basic_session_run_hooks.py:606] Saving checkpoints for 683 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 22:23:31.302932: step 690, loss = 2.98 (4.6 examples/sec; 28.050 sec/batch) I0903 22:28:10.979734 3520 basic_session_run_hooks.py:692] global_step/sec: 0.0357213 2019-09-03 22:28:11.510134: step 700, loss = 3.04 (4.6 examples/sec; 28.021 sec/batch) I0903 22:30:03.050135 3520 basic_session_run_hooks.py:606] Saving checkpoints for 705 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 22:32:51.191936: step 710, loss = 2.73 (4.6 examples/sec; 27.968 sec/batch) 2019-09-03 22:37:30.619138: step 720, loss = 3.09 (4.6 examples/sec; 27.943 sec/batch) I0903 22:40:18.053939 3520 basic_session_run_hooks.py:606] Saving checkpoints for 727 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 22:42:10.701540: step 730, loss = 2.78 (4.6 examples/sec; 28.008 sec/batch) 2019-09-03 22:46:49.769941: step 740, loss = 2.71 (4.6 examples/sec; 27.907 sec/batch) I0903 22:50:32.662743 3520 basic_session_run_hooks.py:606] Saving checkpoints for 749 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 22:51:29.306343: step 750, loss = 3.05 (4.6 examples/sec; 27.954 sec/batch) 2019-09-03 22:56:08.047145: step 760, loss = 2.59 (4.6 examples/sec; 27.874 sec/batch) I0903 23:00:46.959547 3520 basic_session_run_hooks.py:606] Saving checkpoints for 771 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 23:00:47.770747: step 770, loss = 2.48 (4.6 examples/sec; 27.972 sec/batch) 2019-09-03 23:05:27.002749: step 780, loss = 2.69 (4.6 examples/sec; 27.923 sec/batch) 2019-09-03 23:10:06.258350: step 790, loss = 2.73 (4.6 examples/sec; 27.926 sec/batch) I0903 23:11:02.121951 3520 basic_session_run_hooks.py:606] Saving checkpoints for 793 into ./tmp/cifar10_train\model.ckpt. I0903 23:14:46.200352 3520 basic_session_run_hooks.py:692] global_step/sec: 0.0357754 2019-09-03 23:14:46.247152: step 800, loss = 2.72 (4.6 examples/sec; 27.999 sec/batch) 2019-09-03 23:19:25.315554: step 810, loss = 2.91 (4.6 examples/sec; 27.907 sec/batch) I0903 23:21:16.902355 3520 basic_session_run_hooks.py:606] Saving checkpoints for 815 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 23:24:04.789556: step 820, loss = 2.87 (4.6 examples/sec; 27.947 sec/batch) 2019-09-03 23:28:43.533358: step 830, loss = 2.57 (4.6 examples/sec; 27.874 sec/batch) I0903 23:31:30.992959 3520 basic_session_run_hooks.py:606] Saving checkpoints for 837 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 23:33:23.127159: step 840, loss = 2.73 (4.6 examples/sec; 27.959 sec/batch) 2019-09-03 23:38:02.117561: step 850, loss = 2.59 (4.6 examples/sec; 27.899 sec/batch) I0903 23:41:45.634362 3520 basic_session_run_hooks.py:606] Saving checkpoints for 859 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 23:42:42.075163: step 860, loss = 2.67 (4.6 examples/sec; 27.996 sec/batch) 2019-09-03 23:47:20.940765: step 870, loss = 2.48 (4.6 examples/sec; 27.887 sec/batch) I0903 23:51:59.900966 3520 basic_session_run_hooks.py:606] Saving checkpoints for 881 into ./tmp/cifar10_train\model.ckpt. 2019-09-03 23:52:00.743367: step 880, loss = 2.70 (4.6 examples/sec; 27.980 sec/batch) 2019-09-03 23:56:40.264168: step 890, loss = 2.47 (4.6 examples/sec; 27.952 sec/batch) I0904 00:01:18.887770 3520 basic_session_run_hooks.py:692] global_step/sec: 0.0358078 2019-09-04 00:01:19.402570: step 900, loss = 2.42 (4.6 examples/sec; 27.914 sec/batch) I0904 00:02:14.907371 3520 basic_session_run_hooks.py:606] Saving checkpoints for 903 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 00:05:58.034172: step 910, loss = 2.73 (4.6 examples/sec; 27.863 sec/batch) 2019-09-04 00:10:36.338174: step 920, loss = 2.45 (4.6 examples/sec; 27.830 sec/batch) I0904 00:12:27.675374 3520 basic_session_run_hooks.py:606] Saving checkpoints for 925 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 00:15:15.281775: step 930, loss = 2.52 (4.6 examples/sec; 27.894 sec/batch) 2019-09-04 00:19:53.884177: step 940, loss = 2.45 (4.6 examples/sec; 27.860 sec/batch) I0904 00:22:40.850978 3520 basic_session_run_hooks.py:606] Saving checkpoints for 947 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 00:24:33.030579: step 950, loss = 2.52 (4.6 examples/sec; 27.915 sec/batch) 2019-09-04 00:29:11.724581: step 960, loss = 2.30 (4.6 examples/sec; 27.869 sec/batch) I0904 00:32:54.471182 3520 basic_session_run_hooks.py:606] Saving checkpoints for 969 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 00:33:50.903983: step 970, loss = 2.42 (4.6 examples/sec; 27.918 sec/batch) 2019-09-04 00:38:29.083184: step 980, loss = 2.35 (4.6 examples/sec; 27.818 sec/batch) I0904 00:43:07.823986 3520 basic_session_run_hooks.py:606] Saving checkpoints for 991 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 00:43:08.681986: step 990, loss = 2.42 (4.6 examples/sec; 27.960 sec/batch) I0904 00:47:47.266788 3520 basic_session_run_hooks.py:692] global_step/sec: 0.0358631 2019-09-04 00:47:47.313588: step 1000, loss = 2.43 (4.6 examples/sec; 27.863 sec/batch) 2019-09-04 00:52:25.461590: step 1010, loss = 2.50 (4.6 examples/sec; 27.815 sec/batch) I0904 00:53:21.013190 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1013 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 00:57:04.139992: step 1020, loss = 2.38 (4.6 examples/sec; 27.868 sec/batch) 2019-09-04 01:01:42.901393: step 1030, loss = 2.63 (4.6 examples/sec; 27.876 sec/batch) I0904 01:03:34.381994 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1035 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 01:06:22.144395: step 1040, loss = 2.33 (4.6 examples/sec; 27.924 sec/batch) 2019-09-04 01:11:00.729197: step 1050, loss = 2.33 (4.6 examples/sec; 27.858 sec/batch) I0904 01:13:47.820798 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1057 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 01:15:39.563599: step 1060, loss = 2.27 (4.6 examples/sec; 27.883 sec/batch) 2019-09-04 01:20:17.758400: step 1070, loss = 2.20 (4.6 examples/sec; 27.819 sec/batch) I0904 01:24:00.261202 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1079 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 01:24:56.608402: step 1080, loss = 2.03 (4.6 examples/sec; 27.885 sec/batch) 2019-09-04 01:29:34.803204: step 1090, loss = 2.29 (4.6 examples/sec; 27.819 sec/batch) I0904 01:34:13.603406 3520 basic_session_run_hooks.py:692] global_step/sec: 0.0358894 I0904 01:34:14.133806 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1101 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 01:34:14.508206: step 1100, loss = 2.28 (4.6 examples/sec; 27.971 sec/batch) 2019-09-04 01:38:52.297408: step 1110, loss = 2.40 (4.6 examples/sec; 27.779 sec/batch) 2019-09-04 01:43:30.429809: step 1120, loss = 2.34 (4.6 examples/sec; 27.813 sec/batch) I0904 01:44:25.997010 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1123 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 01:48:08.983411: step 1130, loss = 2.25 (4.6 examples/sec; 27.855 sec/batch) 2019-09-04 01:52:47.256213: step 1140, loss = 2.17 (4.6 examples/sec; 27.827 sec/batch) I0904 01:54:38.468614 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1145 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 01:57:25.872215: step 1150, loss = 2.55 (4.6 examples/sec; 27.862 sec/batch) 2019-09-04 02:02:03.978417: step 1160, loss = 2.37 (4.6 examples/sec; 27.811 sec/batch) I0904 02:04:50.761018 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1167 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 02:06:42.815818: step 1170, loss = 2.32 (4.6 examples/sec; 27.884 sec/batch) 2019-09-04 02:11:20.729820: step 1180, loss = 2.21 (4.6 examples/sec; 27.791 sec/batch) I0904 02:15:03.201421 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1189 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 02:15:59.486222: step 1190, loss = 2.27 (4.6 examples/sec; 27.876 sec/batch) I0904 02:20:37.665424 3520 basic_session_run_hooks.py:692] global_step/sec: 0.0359187 2019-09-04 02:20:38.117824: step 1200, loss = 2.16 (4.6 examples/sec; 27.863 sec/batch) I0904 02:25:16.375025 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1211 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 02:25:16.702625: step 1210, loss = 2.12 (4.6 examples/sec; 27.858 sec/batch) 2019-09-04 02:29:54.554227: step 1220, loss = 2.19 (4.6 examples/sec; 27.785 sec/batch) 2019-09-04 02:34:32.342229: step 1230, loss = 2.22 (4.6 examples/sec; 27.779 sec/batch) I0904 02:35:27.862629 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1233 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 02:39:11.207831: step 1240, loss = 2.17 (4.6 examples/sec; 27.887 sec/batch) 2019-09-04 02:43:49.184233: step 1250, loss = 2.40 (4.6 examples/sec; 27.798 sec/batch) I0904 02:45:40.240633 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1255 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 02:48:27.472634: step 1260, loss = 2.22 (4.6 examples/sec; 27.829 sec/batch) 2019-09-04 02:53:05.184836: step 1270, loss = 2.31 (4.6 examples/sec; 27.771 sec/batch) I0904 02:55:52.136037 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1277 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 02:57:44.206438: step 1280, loss = 2.21 (4.6 examples/sec; 27.902 sec/batch) 2019-09-04 03:02:22.206040: step 1290, loss = 2.14 (4.6 examples/sec; 27.800 sec/batch) I0904 03:06:04.240841 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1299 into ./tmp/cifar10_train\model.ckpt. I0904 03:07:00.590042 3520 basic_session_run_hooks.py:692] global_step/sec: 0.0359334 2019-09-04 03:07:00.590042: step 1300, loss = 2.16 (4.6 examples/sec; 27.838 sec/batch) 2019-09-04 03:11:38.441643: step 1310, loss = 2.13 (4.6 examples/sec; 27.785 sec/batch) I0904 03:16:16.355645 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1321 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 03:16:17.182445: step 1320, loss = 2.09 (4.6 examples/sec; 27.874 sec/batch) 2019-09-04 03:20:54.566047: step 1330, loss = 2.13 (4.6 examples/sec; 27.738 sec/batch) 2019-09-04 03:25:32.152449: step 1340, loss = 2.14 (4.6 examples/sec; 27.759 sec/batch) I0904 03:26:27.782049 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1343 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 03:30:10.955650: step 1350, loss = 2.35 (4.6 examples/sec; 27.880 sec/batch) 2019-09-04 03:34:48.907852: step 1360, loss = 2.09 (4.6 examples/sec; 27.795 sec/batch) I0904 03:36:40.089053 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1365 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 03:39:27.695454: step 1370, loss = 2.14 (4.6 examples/sec; 27.879 sec/batch) 2019-09-04 03:44:05.223856: step 1380, loss = 1.82 (4.6 examples/sec; 27.753 sec/batch) I0904 03:46:51.925457 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1387 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 03:48:43.730658: step 1390, loss = 1.94 (4.6 examples/sec; 27.851 sec/batch) I0904 03:53:21.863059 3520 basic_session_run_hooks.py:692] global_step/sec: 0.0359548 2019-09-04 03:53:22.362259: step 1400, loss = 1.89 (4.6 examples/sec; 27.863 sec/batch) I0904 03:57:04.116261 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1409 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 03:57:59.979861: step 1410, loss = 2.06 (4.6 examples/sec; 27.762 sec/batch) 2019-09-04 04:02:37.306063: step 1420, loss = 2.11 (4.6 examples/sec; 27.733 sec/batch) I0904 04:07:14.863665 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1431 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 04:07:15.721665: step 1430, loss = 2.09 (4.6 examples/sec; 27.842 sec/batch) 2019-09-04 04:11:53.042866: step 1440, loss = 2.14 (4.6 examples/sec; 27.732 sec/batch) 2019-09-04 04:16:30.364068: step 1450, loss = 1.88 (4.6 examples/sec; 27.732 sec/batch) I0904 04:17:25.744069 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1453 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 04:21:08.059670: step 1460, loss = 1.84 (4.6 examples/sec; 27.770 sec/batch) 2019-09-04 04:25:45.131272: step 1470, loss = 2.13 (4.6 examples/sec; 27.707 sec/batch) I0904 04:27:36.078472 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1475 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 04:30:23.107674: step 1480, loss = 1.97 (4.6 examples/sec; 27.798 sec/batch) 2019-09-04 04:35:00.045875: step 1490, loss = 1.95 (4.6 examples/sec; 27.694 sec/batch) I0904 04:37:46.435477 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1497 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 04:39:38.225077: step 1500, loss = 1.93 (4.6 examples/sec; 27.818 sec/batch) I0904 04:39:38.225077 3520 basic_session_run_hooks.py:692] global_step/sec: 0.0360184 2019-09-04 04:44:15.780279: step 1510, loss = 2.29 (4.6 examples/sec; 27.756 sec/batch) I0904 04:47:58.345480 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1519 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 04:48:55.004681: step 1520, loss = 2.00 (4.6 examples/sec; 27.922 sec/batch) 2019-09-04 04:53:32.201082: step 1530, loss = 1.98 (4.6 examples/sec; 27.720 sec/batch) I0904 04:58:09.381884 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1541 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 04:58:10.177484: step 1540, loss = 1.83 (4.6 examples/sec; 27.798 sec/batch) 2019-09-04 05:02:47.689886: step 1550, loss = 2.03 (4.6 examples/sec; 27.751 sec/batch) 2019-09-04 05:07:25.073488: step 1560, loss = 1.93 (4.6 examples/sec; 27.738 sec/batch) I0904 05:08:20.656288 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1563 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 05:12:03.283890: step 1570, loss = 1.84 (4.6 examples/sec; 27.821 sec/batch) 2019-09-04 05:16:40.449091: step 1580, loss = 1.93 (4.6 examples/sec; 27.717 sec/batch) I0904 05:18:31.318292 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1585 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 05:21:18.519093: step 1590, loss = 2.13 (4.6 examples/sec; 27.807 sec/batch) I0904 05:25:55.855895 3520 basic_session_run_hooks.py:692] global_step/sec: 0.0360019 2019-09-04 05:25:56.417495: step 1600, loss = 1.81 (4.6 examples/sec; 27.790 sec/batch) I0904 05:28:42.604296 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1607 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 05:30:34.471897: step 1610, loss = 1.96 (4.6 examples/sec; 27.805 sec/batch) 2019-09-04 05:35:11.756299: step 1620, loss = 1.80 (4.6 examples/sec; 27.728 sec/batch) I0904 05:38:53.557100 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1629 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 05:39:49.654700: step 1630, loss = 1.91 (4.6 examples/sec; 27.790 sec/batch) 2019-09-04 05:44:26.601502: step 1640, loss = 1.92 (4.6 examples/sec; 27.695 sec/batch) I0904 05:49:03.610704 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1651 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 05:49:04.437504: step 1650, loss = 1.84 (4.6 examples/sec; 27.784 sec/batch) 2019-09-04 05:53:41.337506: step 1660, loss = 2.00 (4.6 examples/sec; 27.690 sec/batch) 2019-09-04 05:58:18.393507: step 1670, loss = 1.73 (4.6 examples/sec; 27.706 sec/batch) I0904 05:59:13.960708 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1673 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 06:02:56.860509: step 1680, loss = 1.73 (4.6 examples/sec; 27.847 sec/batch) 2019-09-04 06:07:34.696511: step 1690, loss = 1.85 (4.6 examples/sec; 27.784 sec/batch) I0904 06:09:25.924511 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1695 into ./tmp/cifar10_train\model.ckpt. I0904 06:12:13.218913 3520 basic_session_run_hooks.py:692] global_step/sec: 0.0360054 2019-09-04 06:12:13.250113: step 1700, loss = 1.90 (4.6 examples/sec; 27.855 sec/batch) 2019-09-04 06:16:50.290515: step 1710, loss = 1.83 (4.6 examples/sec; 27.704 sec/batch) I0904 06:19:36.477316 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1717 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 06:21:28.360516: step 1720, loss = 1.90 (4.6 examples/sec; 27.807 sec/batch) 2019-09-04 06:26:05.091918: step 1730, loss = 2.02 (4.6 examples/sec; 27.673 sec/batch) I0904 06:29:46.502720 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1739 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 06:30:42.569120: step 1740, loss = 1.65 (4.6 examples/sec; 27.748 sec/batch) 2019-09-04 06:35:21.255522: step 1750, loss = 1.84 (4.6 examples/sec; 27.869 sec/batch) I0904 06:39:59.356724 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1761 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 06:39:59.746723: step 1760, loss = 1.76 (4.6 examples/sec; 27.849 sec/batch) 2019-09-04 06:44:38.347125: step 1770, loss = 1.92 (4.6 examples/sec; 27.860 sec/batch) 2019-09-04 06:49:16.947527: step 1780, loss = 1.73 (4.6 examples/sec; 27.860 sec/batch) I0904 06:50:12.639527 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1783 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 06:53:56.265529: step 1790, loss = 1.71 (4.6 examples/sec; 27.932 sec/batch) I0904 06:58:34.803530 3520 basic_session_run_hooks.py:692] global_step/sec: 0.0359507 2019-09-04 06:58:35.333931: step 1800, loss = 1.55 (4.6 examples/sec; 27.907 sec/batch) I0904 07:00:26.234331 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1805 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 07:03:13.367132: step 1810, loss = 1.75 (4.6 examples/sec; 27.803 sec/batch) 2019-09-04 07:07:51.125134: step 1820, loss = 1.91 (4.6 examples/sec; 27.776 sec/batch) I0904 07:10:37.405535 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1827 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 07:12:28.851936: step 1830, loss = 1.85 (4.6 examples/sec; 27.773 sec/batch) 2019-09-04 07:17:07.655138: step 1840, loss = 1.59 (4.6 examples/sec; 27.880 sec/batch) I0904 07:20:51.015939 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1849 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 07:21:47.300739: step 1850, loss = 1.72 (4.6 examples/sec; 27.965 sec/batch) 2019-09-04 07:26:26.649941: step 1860, loss = 1.75 (4.6 examples/sec; 27.935 sec/batch) I0904 07:31:06.128943 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1871 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 07:31:06.924543: step 1870, loss = 1.71 (4.6 examples/sec; 28.027 sec/batch) 2019-09-04 07:35:46.576145: step 1880, loss = 1.90 (4.6 examples/sec; 27.965 sec/batch) 2019-09-04 07:40:26.362147: step 1890, loss = 1.78 (4.6 examples/sec; 27.979 sec/batch) I0904 07:41:22.241347 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1893 into ./tmp/cifar10_train\model.ckpt. I0904 07:45:06.257348 3520 basic_session_run_hooks.py:692] global_step/sec: 0.0358236 2019-09-04 07:45:06.304148: step 1900, loss = 1.63 (4.6 examples/sec; 27.994 sec/batch) 2019-09-04 07:49:45.700150: step 1910, loss = 1.62 (4.6 examples/sec; 27.940 sec/batch) I0904 07:51:37.630151 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1915 into ./tmp/cifar10_train\model.ckpt. 2019-09-04 07:54:25.673352: step 1920, loss = 1.67 (4.6 examples/sec; 27.997 sec/batch) 2019-09-04 07:59:05.396754: step 1930, loss = 1.80 (4.6 examples/sec; 27.972 sec/batch) I0904 08:01:53.514355 3520 basic_session_run_hooks.py:606] Saving checkpoints for 1937 into ./tmp/cifar10_train\model.ckpt. Process finished with exit code -1
9、