Question

起初看起来像是 Pycharm 错误，但即使从 shell 运行，它也会突然死亡。

如何解决这个问题？

import tensorflow as tf
import matplotlib.pyplot as plt
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Conv2D, Dropout, Flatten, MaxPooling2D

(x_train, y_train), (x_test, y_test) = tf.keras.datasets.mnist.load_data()


image_index = 3535
print(y_train[image_index])
plt.imshow(x_train[image_index], cmap='Greys')

x_train = x_train.reshape(x_train.shape[0], 28, 28, 1)
x_test = x_test.reshape(x_test.shape[0], 28, 28, 1)
input_shape = (28, 28, 1)
x_train = x_train.astype('float32')
x_test = x_test.astype('float32')
x_train /= 255
x_test /= 255
print('x_train shape:', x_train.shape)
print('Number of images in x_train', x_train.shape[0])
print('Number of images in x_test', x_test.shape[0])

# Creating a Sequential Model and adding the layers
model = Sequential()
model.add(Conv2D(28, kernel_size=(3,3), input_shape=input_shape))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Flatten()) # Flattening the 2D arrays for fully connected layers
model.add(Dense(128, activation=tf.nn.relu))
model.add(Dropout(0.2))
model.add(Dense(10,activation=tf.nn.softmax))

model.compile(optimizer='adam',
              loss='sparse_categorical_crossentropy',
              metrics=['accuracy'])
model.fit(x=x_train, y=y_train, epochs=10)
model.evaluate(x_test, y_test)
image_index = 4444
plt.imshow(x_test[image_index].reshape(28, 28),cmap='Greys')
pred = model.predict(x_test[image_index].reshape(1, 28, 28, 1))
print(pred.argmax())

我使用的是 NVIDIA RTX 2080ti，最后一行日志是

2021-06-07 18:29:36.621016: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library cudnn64_8.dll

更新

这适用于 tensorflow-cpu。问题可能是我的 cuDNN 库

Answer 1

对我来说，它在 Jupyter Lab 中工作得很好： tensorflow-gpu 2.4.1

在运行代码之前尝试清除会话：

from tensorflow.keras import backend
backend.clear_session()

Keras 模型退出，代码为 -1073740791 (0xC0000409)

1 个答案: