深度学习框架学习

tensorflow

1、设置GPU占用量

import tensorflow as tf
import keras.backend.tensorflow_backend as KTF

os.environ["CUDA_VISIBLE_DEVICES"] = "0"
config = tf.ConfigProto()  
config.gpu_options.allow_growth=True
sess = tf.Session(config=config)
KTF.set_session(sess)

pytorch

1、类似keras model.summary()

from torchsummary import summary

summary(model, (1, 28, 28))

keras

1、使用tensorboard

 

from keras.callbacks import TensorBoard
model.fit(x_train,y_train,batch_size,epoch,verbose=1,callbacks=[TensorBoard(log_dir='./logs')])
!tensorboard --logdir='logs'

将owen改成localhost

2、类别不平衡

在fit设置class_weight参数