CNN 的加载精度和损失时期

Question

我已经使用以下代码保存了构建 CNN 的历史时期

history=classifier.fit_generator(training_set,
                        steps_per_epoch = 3194 // batchsize,
                        epochs = 100,
                        validation_data =test_set,
                        validation_steps = 1020 // batchsize)
with open('32_With_Dropout_rl_001_1_layer', 'wb') as file_pi:
        pickle.dump(history.history, file_pi)

plt.plot(history.history['val_accuracy'])
plt.title('model accuracy using 32 filters, dropout and .001 Adam learning rate')
plt.ylabel('accuracy')
plt.xlabel('epoch')
plt.legend(['test'], loc='upper left')
plt.show()
# summarize history for loss
plt.plot(history.history['val_loss'])
plt.title('model loss using 32 filters, dropout and .001 Adam learning rate')
plt.ylabel('loss')
plt.xlabel('epoch')
plt.legend(['test'], loc='upper left')
plt.show()

我正在尝试加载我使用以下代码保存的同一图，但它给我 AttributeError: 'dict' object has no attribute 'history'


f = open('32_With_Dropout_rl_001_1_layer', 'rb')
history = pickle.load(f)
f.close()

# summarize history for accuracy
plt.plot(history.history['val_accuracy'])
plt.title('model accuracy using 32 filters, dropout and .001 Adam learning rate')
plt.ylabel('accuracy')
plt.xlabel('epoch')
plt.legend(['test'], loc='upper left')
plt.show()
# summarize history for loss
plt.plot(history.history['val_loss'])
plt.title('model loss using 32 filters, dropout and .001 Adam learning rate')
plt.ylabel('loss')
plt.xlabel('epoch')
plt.legend(['test'], loc='upper left')
plt.show()

Answer 1

您正在保存 history.histroy 字典而不是 history。尝试通过 history['val_loss'] 从加载的泡菜数据中访问您的数据。

CNN 的加载精度和损失时期

loading accuracy and loss epochs of CNN

python

python-3.x

keras

tensor