Fork自 PaddlePaddle / Paddle
* change learning_rate_decay to learning_rate_scheduler * optimize code * change nn.cast to tensor.cast
拖放文件到此处或点击上传