TensorFlow 2.0 版本將 keras 做爲高級 API,對於 keras boy/girl 來講,這就很友好了。tf.keras 從 1.x 版本遷移到 2.0 版本,須要修改幾個地方。python
import tensorflow as tf # TF 1.x tf.set_random_seed(args.seed) # TF 2.0 tf.random.set_seed(args.seed)
import tensorflow as tf from tensorflow.python.keras import backend as K # TF 1.x config = tf.ConfigProto(intra_op_parallelism_threads=1, inter_op_parallelism_threads=1) config.gpu_options.allow_growth = True # 不所有佔滿顯存, 按需分配 K.set_session(tf.Session(config=config)) # TF 2.0 config = tf.compat.v1.ConfigProto(intra_op_parallelism_threads=1, inter_op_parallelism_threads=1) config.gpu_options.allow_growth = True # 不所有佔滿顯存, 按需分配 K.set_session(tf.compat.v1.Session(config=config))
from tensorflow.python.keras import callbacks # TF 1.x ck_callback = callbacks.ModelCheckpoint('./model.h5', monitor='val_acc', mode='max', verbose=1, save_best_only=True, save_weights_only=True) # TF 2.0 ck_callback = callbacks.ModelCheckpoint('./model.h5', monitor='val_accuracy', mode='max', verbose=1, save_best_only=True, save_weights_only=True)