【keras】python mnist_mlp.py下载数据集mnist.npz失败的解决

最新推荐文章于 2023-10-25 10:51:03 发布

CV_ML_DP

最新推荐文章于 2023-10-25 10:51:03 发布

阅读量5.5k

点赞数 3

分类专栏：人工智能 keras 机器学习跟我一起学机器学习 Machine Learning

本文链接：https://blog.csdn.net/u012421852/article/details/79520098

版权

本文档描述了在Ubuntu 16.04上运行keras的mnist_mlp.py示例时遇到的数据集下载失败问题。由于mnist.npz的下载链接不可用，导致程序终止。解决方案是修改mnist.py，使其从本地读取数据集。通过这种方式，mnist_mlp.py成功运行并提供了训练的详细日志。

摘要由CSDN通过智能技术生成

环境

环境：ubuntu 16.04 LTS（ubuntu-16.04.4-server-amd64）

已安装：tensorflow, theano, keras, anaconda

要解决的问题

#python keras-master/examples/mnist_mlp.py

因下载数据集失败而终止运行

keras开发包下载

1）浏览器直接下载，keras-master.zip下载地址：

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

2）也可直接从我的百度网盘下载（永久有效）：

链接：https://pan.baidu.com/share/init?surl=1seQqX6r1Jlr8zLq6x3O_A 密码：fkw1

3）linux环境下直接git命令下载

git clone https://github.com/fchollet/keras.git

开始训练python mnist_mlp.py

get_file方式成功下载mnist_npz时

执行成功，执行过程如下所示：

get_file下载mnist_npz数据集失败时

1）浏览器直接下载，keras-master.zip下载地址

https://s3.amazonaws.com/img-datasets/mnist.npz

2）也可直接从我的百度网盘下载(永久有效)：

链接：https://pan.baidu.com/s/13FhV67vf8umg261a29Eg2Q 密码：yp6h

3）keras-master/keras/datasets/mnist.py中代码执行下载

mnist.py代码：

"""MNIST handwritten digits dataset.
"""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function

from ..utils.data_utils import get_file
import numpy as np


def load_data(path='mnist.npz'):
    """Loads the MNIST dataset.

    # Arguments
        path: path where to cache the dataset locally
            (relative to ~/.keras/datasets).

    # Returns
        Tuple of Numpy arrays: `(x_train, y_train), (x_test, y_test)`.
    """
    path = get_file(path,
                    origin='https://s3.amazonaws.com/img-datasets/mnist.npz',
                    file_hash='8a61469f7ea1b51cbae51d4f78837e45')
    f = np.load(path)
    x_train, y_train = f['x_train'], f['y_train']
    x_test, y_test = f['x_test'], f['y_test']
    f.close()
    return (x_train, y_train), (x_test, y_test)

问题：mnist.py代码下载mnist.npz数据集时失败

分析mnist_mlp.py下载数据集的代码流程

从上面看出来，是因为mnist.npz数据集链接无法访问，导致数据集下载失败。

下面看看mnist_mlp.py下载数据集的代码流程(step1, step2)。

step1:

keras-master\examples\mnist_mlp.py通过keras-master\keras\datasets\mnist.py定义的load_data()函数获取到数据集

from keras.datasets import mnist

# the data, split between train and test sets
(x_train, y_train), (x_test, y_test) = mnist.load_data()

step2：

keras-master\keras\datasets\mnist.py定义的load_data()函数通过get_file(path, origin=数据集链接，file_hash=hash值)从服务器下载数据集mnist.npz

def load_data(path='mnist.npz'):
    """Loads the MNIST dataset.

    # Arguments
        path: path where to cache the dataset locally
            (relative to ~/.keras/datasets).

    # Returns
        Tuple of Numpy arrays: `(x_train, y_train), (x_test, y_test)`.
    """
    path = get_file(path,
                    origin='https://s3.amazonaws.com/img-datasets/mnist.npz',
                    file_hash='8a61469f7ea1b51cbae51d4f78837e45')
    f = np.load(path)
    x_train, y_train = f['x_train'], f['y_train']
    x_test, y_test = f['x_test'], f['y_test']
    f.close()
    return (x_train, y_train), (x_test, y_test)

解决数据集下载失败导致mnist_mlp.py程序退出的问题

改成读本地数据集

step1

keras-master\examples\mnist_mlp.py通过keras-master\keras\datasets\mnist.py定义的load_data()函数获取到数据集

from keras.datasets import mnist

# the data, split between train and test sets
(x_train, y_train), (x_test, y_test) = mnist.load_data()

step2

keras-master\keras\datasets\mnist.py定义的load_data()函数改成读取本地数据集，然后提供给mnist_mlp.py。

mnist_mlp.py运行

先给出训练的简洁结果，后面是运行过程的详细日志。

60000 train samples
10000 test samples
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
dense_1 (Dense)              (None, 512)               401920    
_________________________________________________________________
dropout_1 (Dropout)          (None, 512)               0         
_________________________________________________________________
dense_2 (Dense)              (None, 512)               262656    
_________________________________________________________________
dropout_2 (Dropout)          (None, 512)               0         
_________________________________________________________________
dense_3 (Dense)              (None, 10)                5130      
=================================================================
Total params: 669,706
Trainable params: 669,706
Non-trainable params: 0
_________________________________________________________________
Train on 60000 samples, validate on 10000 samples
Epoch 1/20
60000/60000 [==============================] - 6s 104us/step - loss: 0.2471 - acc: 0.9232 - val_loss: 0.1094 - val_acc: 0.9664

Epoch 2/20
60000/60000 [==============================] - 6s 101us/step - loss: 0.1024 - acc: 0.9698 - val_loss: 0.0726 - val_acc: 0.9776

Epoch 3/20
60000/60000 [==============================] - 6s 99us/step - loss: 0.0754 - acc: 0.9778 - val_loss: 0.0767 - val_acc: 0.9780

Epoch 4/20
60000/60000 [==============================] - 5s 89us/step - loss: 0.0596 - acc: 0.9822 - val_loss: 0.0709 - val_acc: 0.9803

Epoch 5/20
60000/60000 [==============================] - 6s 95us/step - loss: 0.0481 - acc: 0.9853 - val_loss: 0.0877 - val_acc: 0.9789

Epoch 6/20
60000/60000 [==============================] - 6s 100us/step - loss: 0.0432 - acc: 0.9871 - val_loss: 0.0765 - val_acc: 0.9798

Epoch 7/20
60000/60000 [==============================] - 7s 114us/step - loss: 0.0367 - acc: 0.9880 - val_loss: 0.0813 - val_acc: 0.9805

Epoch 8/20
60000/60000 [==============================] - 6s 95us/step - loss: 0.0346 - acc: 0.9897 - val_loss: 0.0867 - val_acc: 0.9815

Epoch 9/20
60000/60000 [==============================] - 6s 94us/step - loss: 0.0313 - acc: 0.9903 - val_loss: 0.0916 - val_acc: 0.9823

Epoch 10/20
60000/60000 [==============================] - 6s 101us/step - loss: 0.0291 - acc: 0.9916 - val_loss: 0.0944 - val_acc: 0.9823

Epoch 11/20
60000/60000 [==============================] - 7s 112us/step - loss: 0.0270 - acc: 0.9923 - val_loss: 0.0810 - val_acc: 0.9832

Epoch 12/20
60000/60000 [==============================] - 6s 101us/step - loss: 0.0260 - acc: 0.9929 - val_loss: 0.0867 - val_acc: 0.9823

Epoch 13/20
60000/60000 [==============================] - 6s 92us/step - loss: 0.0237 - acc: 0.9934 - val_loss: 0.0805 - val_acc: 0.9842

Epoch 14/20
60000/60000 [==============================] - 6s 95us/step - loss: 0.0211 - acc: 0.9943 - val_loss: 0.1062 - val_acc: 0.9807

Epoch 15/20
60000/60000 [==============================] - 6s 101us/step - loss: 0.0209 - acc: 0.9942 - val_loss: 0.1007 - val_acc: 0.9835

Epoch 16/20
60000/60000 [==============================] - 7s 112us/step - loss: 0.0204 - acc: 0.9945 - val_loss: 0.0952 - val_acc: 0.9842

Epoch 17/20
60000/60000 [==============================] - 6s 96us/step - loss: 0.0209 - acc: 0.9944 - val_loss: 0.1140 - val_acc: 0.9821

Epoch 18/20
60000/60000 [==============================] - 8s 130us/step - loss: 0.0199 - acc: 0.9948 - val_loss: 0.1190 - val_acc: 0.9804

Epoch 19/20
60000/60000 [==============================] - 5s 90us/step - loss: 0.0154 - acc: 0.9957 - val_loss: 0.1004 - val_acc: 0.9856

Epoch 20/20
60000/60000 [==============================] - 7s 110us/step - loss: 0.0170 - acc: 0.9955 - val_loss: 0.1028 - val_acc: 0.9835

Test loss: 0.102794561023
Test accuracy: 0.9835

详细训练日志

60000 train samples
10000 test samples
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
dense_1 (Dense)              (None, 512)               401920    
_________________________________________________________________
dropout_1 (Dropout)          (None, 512)               0         
_________________________________________________________________
dense_2 (Dense)              (None, 512)               262656    
_________________________________________________________________
dropout_2 (Dropout)          (None, 512)               0         
_________________________________________________________________
dense_3 (Dense)              (None, 10)                5130      
=================================================================
Total params: 669,706
Trainable params: 669,706
Non-trainable params: 0
_________________________________________________________________
Train on 60000 samples, validate on 10000 samples
Epoch 1/20

  128/60000 [..............................] - ETA: 1:00 - loss: 2.3504 - acc: 0.0703
  512/60000 [..............................] - ETA: 22s - loss: 1.9503 - acc: 0.2812 
 1152/60000 [..............................] - ETA: 12s - loss: 1.4677 - acc: 0.4983
 1664/60000 [..............................] - ETA: 10s - loss: 1.2394 - acc: 0.5811
 2048/60000 [>.............................] - ETA: 11s - loss: 1.1236 - acc: 0.6294
 2560/60000 [>.............................] - ETA: 10s - loss: 1.0245 - acc: 0.6621
 3200/60000 [>.............................] - ETA: 9s - loss: 0.9193 - acc: 0.7016 
 3840/60000 [>.............................] - ETA: 8s - loss: 0.8495 - acc: 0.7255
 4480/60000 [=>............................] - ETA: 8s - loss: 0.8051 - acc: 0.7417
 4992/60000 [=>............................] - ETA: 7s - loss: 0.7731 - acc: 0.7518
 5376/60000 [=>............................] - ETA: 7s - loss: 0.7389 - acc: 0.7641
 5504/60000 [=>............................] - ETA: 8s - loss: 0.7306 - acc: 0.7671
 6144/60000 [==>...........................] - ETA: 7s - loss: 0.7006 - acc: 0.7760
 6784/60000 [==>...........................] - ETA: 7s - loss: 0.6652 - acc: 0.7873
 7424/60000 [==>...........................] - ETA: 7s - loss: 0.6430 - acc: 0.7953
 8192/60000 [===>..........................] - ETA: 6s - loss: 0.6175 - acc: 0.8035
 8960/60000 [===>..........................] - ETA: 6s - loss: 0.5914 - acc: 0.8135
 9472/60000 [===>..........................] - ETA: 6s - loss: 0.5820 - acc: 0.8161
10112/60000 [====>.........................] - ETA: 6s - loss: 0.5660 - acc: 0.8216
10880/60000 [====>.........................] - ETA: 5s - loss: 0.5457 - acc: 0.8280
11648/60000 [====>.........................] - ETA: 5s - loss: 0.5292 - acc: 0.8329
12416/60000 [=====>........................] - ETA: 5s - loss: 0.5155 - acc: 0.8372
13056/60000 [=====>........................] - ETA: 5s - loss: 0.5052 - acc: 0.8408
13440/60000 [=====>........................] - ETA: 5s - loss: 0.4973 - acc: 0.8432
14080/60000 [======>.......................] - ETA: 5s - loss: 0.4879 - acc: 0.8455
14848/60000 [======>.......................] - ETA: 4s - loss: 0.4751 - acc: 0.8501
15616/60000 [======>.......................] - ETA: 4s - loss: 0.4636 - acc: 0.8539
16384/60000 [=======>......................] - ETA: 4s - loss: 0.4545 - acc: 0.8569
17024/60000 [=======>......................] - ETA: 4s - loss: 0.4471 - acc: 0.8598
17792/60000 [=======>......................] - ETA: 4s - loss: 0.4374 - acc: 0.8630
18560/60000 [========>.....................] - ETA: 4s - loss: 0.4288 - acc: 0.8657
19200/60000 [========>.....................] - ETA: 4s - loss: 0.4225 - acc: 0.8678
19968/60000 [========>.....................] - ETA: 4s - loss: 0.4135 - acc: 0.8708
20480/60000 [=========>....................] - ETA: 4s - loss: 0.4085 - acc: 0.8720
21248/60000 [=========>....................] - ETA: 3s - loss: 0.4012 - acc: 0.8742
21760/60000 [=========>....................] - ETA: 3s - loss: 0.3969 - acc: 0.8756
22528/60000 [==========>...................] - ETA: 3s - loss: 0.3905 - acc: 0.8774
23168/60000 [==========>...................] - ETA: 3s - loss: 0.3867 - acc: 0.8790
23808/60000 [==========>...................] - ETA: 3s - loss: 0.3814 - acc: 0.8806
24448/60000 [===========>..................] - ETA: 3s - loss: 0.3760 - acc: 0.8823
25088/60000 [===========>..................] - ETA: 3s - loss: 0.3705 - acc: 0.8839
25728/60000 [===========>..................] - ETA: 3s - loss: 0.3660 - acc: 0.8854
26368/60000 [============>.................] - ETA: 3s - loss: 0.3614 - acc: 0.8869
27136/60000 [============>.................] - ETA: 3s - loss: 0.3576 - acc: 0.8883
27904/60000 [============>.................] - ETA: 3s - loss: 0.3539 - acc: 0.8895
28544/60000 [=============>................] - ETA: 3s - loss: 0.3502 - acc: 0.8907
29184/60000 [=============>................] - ETA: 2s - loss: 0.3459 - acc: 0.8921
29824/60000 [=============>................] - ETA: 2s - loss: 0.3427 - acc: 0.8928
30464/60000 [==============>...............] - ETA: 2s - loss: 0.3398 - acc: 0.8936
31104/60000 [==============>...............] - ETA: 2s - loss: 0.3364 - acc: 0.8945
31744/60000 [==============>...............] - ETA: 2s - loss: 0.3327 - acc: 0.8957
32384/60000 [===============>..............] - ETA: 2s - loss: 0.3299 - acc: 0.8966
33024/60000 [===============>..............] - ETA: 2s - loss: 0.3267 - acc: 0.8979
33664/60000 [===============>..............] - ETA: 2s - loss: 0.3237 - acc: 0.8987
34304/60000 [================>.............] - ETA: 2s - loss: 0.3207 - acc: 0.8997
35072/60000 [================>.............] - ETA: 2s - loss: 0.3174 - acc: 0.9007
35712/60000 [================>.............] - ETA: 2s - loss: 0.3148 - acc: 0.9016
36352/60000 [=================>............] - ETA: 2s - loss: 0.3130 - acc: 0.9020
36992/60000 [=================>............] - ETA: 2s - loss: 0.3095 - acc: 0.9030
37632/60000 [=================>............] - ETA: 2s - loss: 0.3067 - acc: 0.9041
38144/60000 [==================>...........] - ETA: 2s - loss: 0.3054 - acc: 0.9044
38784/60000 [==================>...........] - ETA: 1s - loss: 0.3030 - acc: 0.9051
39424/60000 [==================>...........] - ETA: 1s - loss: 0.3009 - acc: 0.9059
40064/60000 [===================>..........] - ETA: 1s - loss: 0.2985 - acc: 0.9068
40704/60000 [===================>..........] - ETA: 1s - loss: 0.2963 - acc: 0.9077
41344/60000 [===================>..........] - ETA: 1s - loss: 0.2941 - acc: 0.9084
41984/60000 [===================>..........] - ETA: 1s - loss: 0.2916 - acc: 0.9091
42624/60000 [====================>.........] - ETA: 1s - loss: 0.2898 - acc: 0.9096
43008/60000 [====================>.........] - ETA: 1s - loss: 0.2887 - acc: 0.9100
43392/60000 [====================>.........] - ETA: 1s - loss: 0.2876 - acc: 0.9103
44032/60000 [=====================>........] - ETA: 1s - loss: 0.2857 - acc: 0.9108
44672/60000 [=====================>........] - ETA: 1s - loss: 0.2837 - acc: 0.9115
45312/60000 [=====================>........] - ETA: 1s - loss: 0.2817 - acc: 0.9121
45952/60000 [=====================>........] - ETA: 1s - loss: 0.2796 - acc: 0.9127
46464/60000 [======================>.......] - ETA: 1s - loss: 0.2790 - acc: 0.9131
47104/60000 [======================>.......] - ETA: 1s - loss: 0.2770 - acc: 0.9138
47744/60000 [======================>.......] - ETA: 1s - loss: 0.2751 - acc: 0.9144
48384/60000 [=======================>......] - ETA: 1s - loss: 0.2739 - acc: 0.9148
48640/60000 [=======================>......] - ETA: 1s - loss: 0.2728 - acc: 0.9152
49280/60000 [=======================>......] - ETA: 1s - loss: 0.2705 - acc: 0.9159
49920/60000 [=======================>......] - ETA: 0s - loss: 0.2690 - acc: 0.9163
50176/60000 [========================>.....] - ETA: 0s - loss: 0.2683 - acc: 0.9165
50816/60000 [========================>.....] - ETA: 0s - loss: 0.2669 - acc: 0.9171
51456/60000 [========================>.....] - ETA: 0s - loss: 0.2653 - acc: 0.9176
51840/60000 [========================>.....] - ETA: 0s - loss: 0.2645 - acc: 0.9179
52480/60000 [=========================>....] - ETA: 0s - loss: 0.2623 - acc: 0.9186
52992/60000 [=========================>....] - ETA: 0s - loss: 0.2609 - acc: 0.9190
53376/60000 [=========================>....] - ETA: 0s - loss: 0.2602 - acc: 0.9192
53888/60000 [=========================>....] - ETA: 0s - loss: 0.2592 - acc: 0.9195
54400/60000 [==========================>...] - ETA: 0s - loss: 0.2580 - acc: 0.9198
54912/60000 [==========================>...] - ETA: 0s - loss: 0.2571 - acc: 0.9202
55552/60000 [==========================>...] - ETA: 0s - loss: 0.2560 - acc: 0.9206
55936/60000 [==========================>...] - ETA: 0s - loss: 0.2554 - acc: 0.9207
56576/60000 [===========================>..] - ETA: 0s - loss: 0.2539 - acc: 0.9212
57216/60000 [===========================>..] - ETA: 0s - loss: 0.2528 - acc: 0.9214
57856/60000 [===========================>..] - ETA: 0s - loss: 0.2516 - acc: 0.9218
58496/60000 [============================>.] - ETA: 0s - loss: 0.2503 - acc: 0.9222
59008/60000 [============================>.] - ETA: 0s - loss: 0.2489 - acc: 0.9226
59648/60000 [============================>.] - ETA: 0s - loss: 0.2479 - acc: 0.9229
60000/60000 [==============================] - 6s 104us/step - loss: 0.2471 - acc: 0.9232 - val_loss: 0.1094 - val_acc: 0.9664
Epoch 2/20

  128/60000 [..............................] - ETA: 5s - loss: 0.1950 - acc: 0.9531
  512/60000 [..............................] - ETA: 7s - loss: 0.1482 - acc: 0.9551
 1152/60000 [..............................] - ETA: 6s - loss: 0.1353 - acc: 0.9566
 1792/60000 [..............................] - ETA: 5s - loss: 0.1247 - acc: 0.9570
 2432/60000 [>.............................] - ETA: 5s - loss: 0.1116 - acc: 0.9630
 2688/60000 [>.............................] - ETA: 5s - loss: 0.1149 - acc: 0.9632
 2944/60000 [>.............................] - ETA: 6s - loss: 0.1128 - acc: 0.9643
 3200/60000 [>.............................] - ETA: 7s - loss: 0.1131 - acc: 0.9650
 3840/60000 [>.............................] - ETA: 7s - loss: 0.1162 - acc: 0.9648
 4480/60000 [=>............................] - ETA: 6s - loss: 0.1107 - acc: 0.9667
 4608/60000 [=>............................] - ETA: 7s - loss: 0.1114 - acc: 0.9659
 5120/60000 [=>............................] - ETA: 7s - loss: 0.1128 - acc: 0.9654
 5760/60000 [=>............................] - ETA: 7s - loss: 0.1116 - acc: 0.9655
 6272/60000 [==>...........................] - ETA: 6s - loss: 0.1093 - acc: 0.9660
 6912/60000 [==>...........................] - ETA: 6s - loss: 0.1104 - acc: 0.9651
 7296/60000 [==>...........................] - ETA: 6s - loss: 0.1113 - acc: 0.9645
 7808/60000 [==>...........................] - ETA: 6s - loss: 0.1122 - acc: 0.9645
 8320/60000 [===>..........................] - ETA: 6s - loss: 0.1131 - acc: 0.9641
 8960/60000 [===>..........................] - ETA: 6s - loss: 0.1136 - acc: 0.9642
 9472/60000 [===>..........................] - ETA: 6s - loss: 0.1151 - acc: 0.9638
 9728/60000 [===>..........................] - ETA: 6s - loss: 0.1149 - acc: 0.9640
10240/60000 [====>.........................] - ETA: 6s - loss: 0.1135 - acc: 0.9645
10624/60000 [====>.........................] - ETA: 6s - loss: 0.1125 - acc: 0.9648
11008/60000 [====>.........................] - ETA: 6s - loss: 0.1114 - acc: 0.9654
11520/60000 [====>.........................] - ETA: 6s - loss: 0.1100 - acc: 0.9657
11776/60000 [====>.........................] - ETA: 6s - loss: 0.1099 - acc: 0.9657
12288/60000 [=====>........................] - ETA: 6s - loss: 0.1109 - acc: 0.9654
12928/60000 [=====>........................] - ETA: 6s - loss: 0.1112 - acc: 0.9649
13568/60000 [=====>........................] - ETA: 5s - loss: 0.1098 - acc: 0.9654
14080/60000 [======>.......................] - ETA: 5s - loss: 0.1096 - acc: 0.9656
14592/60000 [======>.......................] - ETA: 5s - loss: 0.1089 - acc: 0.9655
15232/60000 [======>.......................] - ETA: 5s - loss: 0.1077 - acc: 0.9662
15872/60000 [======>.......................] - ETA: 5s - loss: 0.1055 - acc: 0.9669
16512/60000 [=======>......................] - ETA: 5s - loss: 0.1062 - acc: 0.9669
17152/60000 [=======>......................] - ETA: 5s - loss: 0.1089 - acc: 0.9662
17792/60000 [=======>......................] - ETA: 5s - loss: 0.1093 - acc: 0.9663
18432/60000 [========>.....................] - ETA: 4s - loss: 0.1091 - acc: 0.9663
19072/60000 [========>.....................] - ETA: 4s - loss: 0.1089 - acc: 0.9664
19584/60000 [========>.....................] - ETA: 4s - loss: 0.1090 - acc: 0.9666
20096/60000 [=========>....................] - ETA: 4s - loss: 0.1095 - acc: 0.9662
20736/60000 [=========>....................] - ETA: 4s - loss: 0.1102 - acc: 0.9661
21376/60000 [=========>....................] - ETA: 4s - loss: 0.1114 - acc: 0.9659
22016/60000 [==========>...................] - ETA: 4s - loss: 0.1106 - acc: 0.9660
22656/60000 [==========>...................] - ETA: 4s - loss: 0.1091 - acc: 0.9664
23296/60000 [==========>...................] - ETA: 4s - loss: 0.1093 - acc: 0.9666
23936/60000 [==========>...................] - ETA: 4s - loss: 0.1087 - acc: 0.9666
24576/60000 [===========>..................] - ETA: 3s - loss: 0.1075 - acc: 0.9670
24832/60000 [===========>..................] - ETA: 3s - loss: 0.1078 - acc: 0.9670
25472/60000 [===========>..................] - ETA: 3s - loss: 0.1068 - acc: 0.9673
26112/60000 [============>.................] - ETA: 3s - loss: 0.1062 - acc: 0.9676
26752/60000 [============>.................] - ETA: 3s - loss: 0.1066 - acc: 0.9676
27392/60000 [============>.................] - ETA: 3s - loss: 0.1067 - acc: 0.9676
28160/60000 [=============>................] - ETA: 3s - loss: 0.1069 - acc: 0.9676
28672/60000 [=============>................] - ETA: 3s - loss: 0.1064 - acc: 0.9679
29440/60000 [=============>................] - ETA: 3s - loss: 0.1071 - acc: 0.9677
30080/60000 [==============>...............] - ETA: 3s - loss: 0.1078 - acc: 0.9676
30720/60000 [==============>...............] - ETA: 3s - loss: 0.1076 - acc: 0.9676
31360/60000 [==============>...............] - ETA: 3s - loss: 0.1074 - acc: 0.9677
32000/60000 [===============>..............] - ETA: 2s - loss: 0.1072 - acc: 0.9677
32640/60000 [===============>..............] - ETA: 2s - loss: 0.1075 - acc: 0.9677
33280/60000 [===============>..............] - ETA: 2s - loss: 0.1067 - acc: 0.9680
34048/60000 [================>.............] - ETA: 2s - loss: 0.1058 - acc: 0.9682
34560/60000 [================>.............] - ETA: 2s - loss: 0.1052 - acc: 0.9683
35200/60000 [================>.............] - ETA: 2s - loss: 0.1068 - acc: 0.9683
35840/60000 [================>.............] - ETA: 2s - loss: 0.1068 - acc: 0.9683
36480/60000 [=================>............] - ETA: 2s - loss: 0.1067 - acc: 0.9683
37120/60000 [=================>............] - ETA: 2s - loss: 0.1069 - acc: 0.9683
37760/60000 [=================>............] - ETA: 2s - loss: 0.1062 - acc: 0.9685
38400/60000 [==================>...........] - ETA: 2s - loss: 0.1063 - acc: 0.9686
39040/60000 [==================>...........] - ETA: 2s - loss: 0.1059 - acc: 0.9687
39680/60000 [==================>...........] - ETA: 2s - loss: 0.1060 - acc: 0.9687
40320/60000 [===================>..........] - ETA: 1s - loss: 0.1062 - acc: 0.9688
40960/60000 [===================>..........] - ETA: 1s - loss: 0.1064 - acc: 0.9686
41600/60000 [===================>..........] - ETA: 1s - loss: 0.1060 - acc: 0.9686
42240/60000 [====================>.........] - ETA: 1s - loss: 0.1055 - acc: 0.9688
42880/60000 [====================>.........] - ETA: 1s - loss: 0.1061 - acc: 0.9686
43392/60000 [====================>.........] - ETA: 1s - loss: 0.1062 - acc: 0.9687
44032/60000 [=====================>........] - ETA: 1s - loss: 0.1054 - acc: 0.9689
44672/60000 [=====================>........] - ETA: 1s - loss: 0.1057 - acc: 0.9688
45312/60000 [=====================>........] - ETA: 1s - loss: 0.1051 - acc: 0.9690
45952/60000 [=====================>........] - ETA: 1s - loss: 0.1046 - acc: 0.9691
46464/60000 [======================>.......] - ETA: 1s - loss: 0.1041 - acc: 0.9691
47104/60000 [======================>.......] - ETA: 1s - loss: 0.1042 - acc: 0.9690
47744/60000 [======================>.......] - ETA: 1s - loss: 0.1035 - acc: 0.9692
48384/60000 [=======================>......] - ETA: 1s - loss: 0.1038 - acc: 0.9691
49024/60000 [=======================>......] - ETA: 1s - loss: 0.1036 - acc: 0.9692
49664/60000 [=======================>......] - ETA: 1s - loss: 0.1036 - acc: 0.9692
50304/60000 [========================>.....] - ETA: 0s - loss: 0.1042 - acc: 0.9692
50944/60000 [========================>.....] - ETA: 0s - loss: 0.1039 - acc: 0.9693
51584/60000 [========================>.....] - ETA: 0s - loss: 0.1045 - acc: 0.9692
52224/60000 [=========================>....] - ETA: 0s - loss: 0.1040 - acc: 0.9693
52864/60000 [=========================>....] - ETA: 0s - loss: 0.1039 - acc: 0.9693
53504/60000 [=========================>....] - ETA: 0s - loss: 0.1036 - acc: 0.9694
54144/60000 [==========================>...] - ETA: 0s - loss: 0.1033 - acc: 0.9695
54528/60000 [==========================>...] - ETA: 0s - loss: 0.1036 - acc: 0.9694
55168/60000 [==========================>...] - ETA: 0s - loss: 0.1032 - acc: 0.9695
55680/60000 [==========================>...] - ETA: 0s - loss: 0.1029 - acc: 0.9696
56320/60000 [===========================>..] - ETA: 0s - loss: 0.1028 - acc: 0.9696
56960/60000 [===========================>..] - ETA: 0s - loss: 0.1027 - acc: 0.9696
57472/60000 [===========================>..] - ETA: 0s - loss: 0.1023 - acc: 0.9697
58112/60000 [============================>.] - ETA: 0s - loss: 0.1023 - acc: 0.9698
58752/60000 [============================>.] - ETA: 0s - loss: 0.1024 - acc: 0.9698
59392/60000 [============================>.] - ETA: 0s - loss: 0.1025 - acc: 0.9698
60000/60000 [==============================] - 6s 101us/step - loss: 0.1024 - acc: 0.9698 - val_loss: 0.0726 - val_acc: 0.9776
Epoch 3/20

  128/60000 [..............................] - ETA: 9s - loss: 0.1333 - acc: 0.9688
  640/60000 [..............................] - ETA: 8s - loss: 0.0780 - acc: 0.9797
  896/60000 [..............................] - ETA: 9s - loss: 0.0655 - acc: 0.9821
 1408/60000 [..............................] - ETA: 8s - loss: 0.0605 - acc: 0.9822
 2048/60000 [>.............................] - ETA: 7s - loss: 0.0650 - acc: 0.9805
 2688/60000 [>.............................] - ETA: 6s - loss: 0.0601 - acc: 0.9807
 3456/60000 [>.............................] - ETA: 6s - loss: 0.0668 - acc: 0.9797
 4224/60000 [=>............................] - ETA: 5s - loss: 0.0658 - acc: 0.9799
 4864/60000 [=>............................] - ETA: 5s - loss: 0.0673 - acc: 0.9799
 5504/60000 [=>............................] - ETA: 5s - loss: 0.0689 - acc: 0.9791
 6144/60000 [==>...........................] - ETA: 5s - loss: 0.0730 - acc: 0.9787
 6784/60000 [==>...........................] - ETA: 5s - loss: 0.0758 - acc: 0.9776
 7424/60000 [==>...........................] - ETA: 4s - loss: 0.0751 - acc: 0.9782
 8192/60000 [===>..........................] - ETA: 4s - loss: 0.0732 - acc: 0.9785
 8832/60000 [===>..........................] - ETA: 4s - loss: 0.0721 - acc: 0.9787
 9472/60000 [===>..........................] - ETA: 4s - loss: 0.0720 - acc: 0.9787
10112/60000 [====>.........................] - ETA: 4s - loss: 0.0703 - acc: 0.9793
10880/60000 [====>.........................] - ETA: 4s - loss: 0.0720 - acc: 0.9790
11648/60000 [====>.........................] - ETA: 4s - loss: 0.0703 - acc: 0.9795
12288/60000 [=====>........................] - ETA: 4s - loss: 0.0711 - acc: 0.9791
12544/60000 [=====>........................] - ETA: 4s - loss: 0.0708 - acc: 0.9792
12800/60000 [=====>........................] - ETA: 4s - loss: 0.0713 - acc: 0.9791
13440/60000 [=====>........................] - ETA: 4s - loss: 0.0703 - acc: 0.9789
14080/60000 [======>.......................] - ETA: 4s - loss: 0.0711 - acc: 0.9790
14720/60000 [======>.......................] - ETA: 4s - loss: 0.0711 - acc: 0.9790
15360/60000 [======>.......................] - ETA: 4s - loss: 0.0715 - acc: 0.9790
16000/60000 [=======>......................] - ETA: 4s - loss: 0.0724 - acc: 0.9788
16384/60000 [=======>......................] - ETA: 4s - loss: 0.0730 - acc: 0.9785
17024/60000 [=======>......................] - ETA: 4s - loss: 0.0721 - acc: 0.9786
17536/60000 [=======>......................] - ETA: 4s - loss: 0.0722 - acc: 0.9784
18048/60000 [========>.....................] - ETA: 3s - loss: 0.0722 - acc: 0.9784
18432/60000 [========>.....................] - ETA: 3s - loss: 0.0719 - acc: 0.9785
19072/60000 [========>.....................] - ETA: 3s - loss: 0.0730 - acc: 0.9783
19712/60000 [========>.....................] - ETA: 3s - loss: 0.0732 - acc: 0.9779
19968/60000 [========>.....................] - ETA: 3s - loss: 0.0730 - acc: 0.9779
20608/60000 [=========>....................] - ETA: 3s - loss: 0.0723 - acc: 0.9782
21248/60000 [=========>....................] - ETA: 3s - loss: 0.0722 - acc: 0.9781
21888/60000 [=========>....................] - ETA: 3s - loss: 0.0720 - acc: 0.9779
22528/60000 [==========>...................] - ETA: 3s - loss: 0.0716 - acc: 0.9779
23168/60000 [==========>...................] - ETA: 3s - loss: 0.0716 - acc: 0.9779
23808/60000 [==========>...................] - ETA: 3s - loss: 0.0718 - acc: 0.9779
24448/60000 [===========>..................] - ETA: 3s - loss: 0.0728 - acc: 0.9777
25088/60000 [===========>..................] - ETA: 3s - loss: 0.0732 - acc: 0.9777
25856/60000 [===========>..................] - ETA: 3s - loss: 0.0742 - acc: 0.9777
26496/60000 [============>.................] - ETA: 3s - loss: 0.0736 - acc: 0.9778
27136/60000 [============>.................] - ETA: 3s - loss: 0.0736 - acc: 0.9778
27776/60000 [============>.................] - ETA: 2s - loss: 0.0736 - acc: 0.9778
28416/60000 [=============>................] - ETA: 2s - loss: 0.0737 - acc: 0.9777
29056/60000 [=============>................] - ETA: 2s - loss: 0.0735 - acc: 0.9778
29824/60000 [=============>................] - ETA: 2s - loss: 0.0740 - acc: 0.9775
30464/60000 [==============>...............] - ETA: 2s - loss: 0.0740 - acc: 0.9775
31104/60000 [==============>...............] - ETA: 2s - loss: 0.0736 - acc: 0.9778
31616/60000 [==============>...............] - ETA: 2s - loss: 0.0732 - acc: 0.9779
32256/60000 [===============>..............] - ETA: 2s - loss: 0.0729 - acc: 0.9778
33024/60000 [===============>..............] - ETA: 2s - loss: 0.0732 - acc: 0.9778
33536/60000 [===============>..............] - ETA: 2s - loss: 0.0734 - acc: 0.9777
34176/60000 [================>.............] - ETA: 2s - loss: 0.0742 - acc: 0.9776
34688/60000 [================>.............] - ETA: 2s - loss: 0.0740 - acc: 0.9777
35328/60000 [================>.............] - ETA: 2s - loss: 0.0742 - acc: 0.9776
35968/60000 [================>.............] - ETA: 2s - loss: 0.0738 - acc: 0.9777
36608/60000 [=================>............] - ETA: 2s - loss: 0.0743 - acc: 0.9776
37248/60000 [=================>............] - ETA: 2s - loss: 0.0739 - acc: 0.9778
37888/60000 [=================>............] - ETA: 2s - loss: 0.0742 - acc: 0.9778
38656/60000 [==================>...........] - ETA: 1s - loss: 0.0742 - acc: 0.9779
39296/60000 [==================>...........] - ETA: 1s - loss: 0.0742 - acc: 0.9779
39936/60000 [==================>...........] - ETA: 1s - loss: 0.0743 - acc: 0.9779
40576/60000 [===================>..........] - ETA: 1s - loss: 0.0738 - acc: 0.9780
41216/60000 [===================>..........] - ETA: 1s - loss: 0.0740 - acc: 0.9780
41728/60000 [===================>..........] - ETA: 1s - loss: 0.0740 - acc: 0.9780
42368/60000 [====================>.........] - ETA: 1s - loss: 0.0740 - acc: 0.9780
43008/60000 [====================>.........] - ETA: 1s - loss: 0.0745 - acc: 0.9781
43520/60000 [====================>.........] - ETA: 1s - loss: 0.0746 - acc: 0.9780
44160/60000 [=====================>........] - ETA: 1s - loss: 0.0744 - acc: 0.9781
44544/60000 [=====================>........] - ETA: 1s - loss: 0.0749 - acc: 0.9780
44928/60000 [=====================>........] - ETA: 1s - loss: 0.0752 - acc: 0.9779
45568/60000 [=====================>........] - ETA: 1s - loss: 0.0750 - acc: 0.9779
45952/60000 [=====================>........] - ETA: 1s - loss: 0.0750 - acc: 0.9779
46592/60000 [======================>.......] - ETA: 1s - loss: 0.0750 - acc: 0.9778
47232/60000 [======================>.......] - ETA: 1s - loss: 0.0751 - acc: 0.9778
47616/60000 [======================>.......] - ETA: 1s - loss: 0.0751 - acc: 0.9778
48128/60000 [=======================>......] - ETA: 1s - loss: 0.0756 - acc: 0.9778
48640/60000 [=======================>......] - ETA: 1s - loss: 0.0757 - acc: 0.9778
49280/60000 [=======================>......] - ETA: 1s - loss: 0.0758 - acc: 0.9777
49408/60000 [=======================>......] - ETA: 1s - loss: 0.0758 - acc: 0.9777
50048/60000 [========================>.....] - ETA: 0s - loss: 0.0758 - acc: 0.9777
50688/60000 [========================>.....] - ETA: 0s - loss: 0.0757 - acc: 0.9777
51328/60000 [========================>.....] - ETA: 0s - loss: 0.0757 - acc: 0.9777
51584/60000 [========================>.....] - ETA: 0s - loss: 0.0754 - acc: 0.9778
52224/60000 [=========================>....] - ETA: 0s - loss: 0.0753 - acc: 0.9778
52608/60000 [=========================>....] - ETA: 0s - loss: 0.0755 - acc: 0.9777
53248/60000 [=========================>....] - ETA: 0s - loss: 0.0751 - acc: 0.9778
53888/60000 [=========================>....] - ETA: 0s - loss: 0.0753 - acc: 0.9779
54400/60000 [==========================>...] - ETA: 0s - loss: 0.0753 - acc: 0.9778
54784/60000 [==========================>...] - ETA: 0s - loss: 0.0753 - acc: 0.9778
55424/60000 [==========================>...] - ETA: 0s - loss: 0.0754 - acc: 0.9777
56064/60000 [===========================>..] - ETA: 0s - loss: 0.0757 - acc: 0.9777
56704/60000 [===========================>..] - ETA: 0s - loss: 0.0760 - acc: 0.9777
57344/60000 [===========================>..] - ETA: 0s - loss: 0.0764 - acc: 0.9776
58112/60000 [============================>.] - ETA: 0s - loss: 0.0758 - acc: 0.9777
58752/60000 [============================>.] - ETA: 0s - loss: 0.0755 - acc: 0.9778
59392/60000 [============================>.] - ETA: 0s - loss: 0.0755 - acc: 0.9778
60000/60000 [==============================] - 6s 99us/step - loss: 0.0754 - acc: 0.9778 - val_loss: 0.0767 - val_acc: 0.9780
Epoch 4/20

  128/60000 [..............................] - ETA: 4s - loss: 0.0358 - acc: 0.9844
  768/60000 [..............................] - ETA: 4s - loss: 0.0566 - acc: 0.9818
 1408/60000 [..............................] - ETA: 4s - loss: 0.0516 - acc: 0.9830
 1920/60000 [..............................] - ETA: 4s - loss: 0.0472 - acc: 0.9833
 2560/60000 [>.............................] - ETA: 4s - loss: 0.0504 - acc: 0.9836
 3200/60000 [>.............................] - ETA: 4s - loss: 0.0485 - acc: 0.9844
 3968/60000 [>.............................] - ETA: 4s - loss: 0.0455 - acc: 0.9851
 4480/60000 [=>............................] - ETA: 4s - loss: 0.0448 - acc: 0.9857
 5120/60000 [=>............................] - ETA: 4s - loss: 0.0473 - acc: 0.9850
 5760/60000 [=>............................] - ETA: 4s - loss: 0.0454 - acc: 0.9852
 6400/60000 [==>...........................] - ETA: 4s - loss: 0.0471 - acc: 0.9848
 7040/60000 [==>...........................] - ETA: 4s - loss: 0.0474 - acc: 0.9848
 7680/60000 [==>...........................] - ETA: 4s - loss: 0.0484 - acc: 0.9844
 8320/60000 [===>..........................] - ETA: 4s - loss: 0.0494 - acc: 0.9839
 8832/60000 [===>..........................] - ETA: 4s - loss: 0.0507 - acc: 0.9835
 9344/60000 [===>..........................] - ETA: 4s - loss: 0.0508 - acc: 0.9836
 9984/60000 [===>..........................] - ETA: 4s - loss: 0.0503 - acc: 0.9842
10624/60000 [====>.........................] - ETA: 4s - loss: 0.0496 - acc: 0.9845
11264/60000 [====>.........................] - ETA: 4s - loss: 0.0501 - acc: 0.9842
11904/60000 [====>.........................] - ETA: 4s - loss: 0.0494 - acc: 0.9845
12672/60000 [=====>........................] - ETA: 4s - loss: 0.0505 - acc: 0.9845
13312/60000 [=====>........................] - ETA: 4s - loss: 0.0516 - acc: 0.9839
13824/60000 [=====>........................] - ETA: 4s - loss: 0.0524 - acc: 0.9839
14208/60000 [======>.......................] - ETA: 4s - loss: 0.0527 - acc: 0.9837
14848/60000 [======>.......................] - ETA: 4s - loss: 0.0551 - acc: 0.9834
15488/60000 [======>.......................] - ETA: 3s - loss: 0.0551 - acc: 0.9834
16128/60000 [=======>......................] - ETA: 3s - loss: 0.0561 - acc: 0.9830
16768/60000 [=======>......................] - ETA: 3s - loss: 0.0561 - acc: 0.9828
17408/60000 [=======>......................] - ETA: 3s - loss: 0.0560 - acc: 0.9828
18048/60000 [========>.....................] - ETA: 3s - loss: 0.0561 - acc: 0.9828
18432/60000 [========>.....................] - ETA: 3s - loss: 0.0559 - acc: 0.9827
19072/60000 [========>.....................] - ETA: 3s - loss: 0.0557 - acc: 0.9830
19712/60000 [========>.....................] - ETA: 3s - loss: 0.0560 - acc: 0.9828
20352/60000 [=========>....................] - ETA: 3s - loss: 0.0555 - acc: 0.9828
20864/60000 [=====

最低0.47元/天解锁文章

CV_ML_DP

关注

3
点赞
踩
3

收藏

觉得还不错? 一键收藏
0
评论
【keras】python mnist_mlp.py下载数据集mnist.npz失败的解决

环境环境：ubuntu 16.04 LTS（ubuntu-16.04.4-server-amd64）已安装：tensorflow, theano, keras, anaconda要解决的问题 #python keras-master/examples/mnist_mlp.py 因下载数据集失败而终止运行keras开发包下载 1）浏览器直接下载，keras-master.z...
复制链接

扫一扫