tensorboard在colab中的实现

colab现在自带tensorboard的魔术方法了，用这个命令就能展示tensorboard

%reload_ext tensorboard
%tensorboard --logdir './callsbacks'

而在代码中则需要写明以下的部分

log_dir= './callsbacks'

if not os.path.exists(log_dir):

os.mkdir(log_dir)

output_model_file = os.path.join(log_dir,"best.h5")

callbacks = [

tf.keras.callbacks.TensorBoard(log_dir),

tf.keras.callbacks.ModelCheckpoint(output_model_file,save_beat_only = True),

tf.keras.callbacks.EarlyStopping(patience=5,min_delta=1e-3),

]

 

 

model.fit(trainX, trainY, batch_size=256, epochs=100, verbose=2, validation_split=0.1,callbacks = callbacks)