Closed xiacoder closed 2 years ago
模型训练结束后,进行测试时候报了cuda out of memory model, optimizer, epoch, loss = self.load_ckp(model, optimizer, checkpoint)这一行代码
那是因为GPU显存不够,调小测试时的batch_size
模型训练结束后,进行测试时候报了cuda out of memory model, optimizer, epoch, loss = self.load_ckp(model, optimizer, checkpoint)这一行代码