win10+keras 跑ResNet 完美解决Failed to get convolution algorithm. This is probably because cuDNN failed to initialize报错

#GPU按需分配,解决
import tensorflow as tf
import keras
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
keras.backend.tensorflow_backend.set_session(tf.Session(config=config))

参考https://blog.csdn.net/qq_41868689/article/details/98503069

原因是显存不足，因此设置成按需分配。