keras使用tensorboard

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15

# 导入库
from keras.callbacks import TensorBoard
# 创建tensorboard对象， 结果保存在logs目录下
tensorboard = TensorBoard(log_dir='logs/{}'.format(NAME),histogram_freq=1,write_grads=True)
# 在模型生成器函数作为回调参数

model.fit_generator(
        generator=train_generator,
        epochs=100,
        steps_per_epoch=2200 // BATCH_SIZE,
        validation_data=validation_generator,
        validation_steps=200 // BATCH_SIZE,
        callbacks=[tensorboard]
    )

1

tensorboard --logdir=logs