提交 0b01bd83 编写于 作者: A Aston Zhang

aug epoch 5

上级 9c7ace08
......@@ -228,7 +228,7 @@ def train_with_data_aug(train_augs, test_augs, lr=0.001):
loss = gloss.SoftmaxCrossEntropyLoss()
train_iter = load_cifar10(True, train_augs, batch_size)
test_iter = load_cifar10(False, test_augs, batch_size)
train(train_iter, test_iter, net, loss, trainer, ctx, num_epochs=6)
train(train_iter, test_iter, net, loss, trainer, ctx, num_epochs=5)
```
### 有关图像增广的对比实验
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册