如何解决Pytorch-lightning 奇怪的错误:实现的方法没有覆盖其父级的方法
我正在尝试使用库 Pytorch Lightning,但遇到了一个我无法解决的奇怪错误。我无法提供最小的工作实现,因为我的代码分散在多个模块中并且不使用许多预构建的组件。此外,我无法共享我正在使用的数据。不过,我希望能找到一些帮助。
基本上,我使用了一个 VGG 网络,它实现了 pytorch-lightning 库所需的方法:
import pytorch_lightning as pl
[...]
class VGG(pl.LightningModule):
[...]
def forward(self,x):
x = self.conv_layers(x)
x = x.reshape(x.shape[0],-1)
x = self.output(x)
return x
def configure_optimizers(self):
# in the docs on the pytorch-lightning website,the self.parameters()
# below appears both with parentheses and without,# i.e.,both self.parameters() and self.paramters
# I tried both with no changes
optimizer = torch.optim.SGD(self.parameters(),lr=self.config["learning_rate"],momentum=self.config["momentum"])
lr_scheduler = ReduceLROnPlateau(optimizer,'min',factor=0.05,patience=5,cooldown=0,verbose=True)
return optimizer,lr_scheduler
def training_step(self,batch,batch_idx):
x,y = batch
y_hat = self.forward(x)
loss = F.cross_entropy(y_hat,y)
return {'loss':loss}
我有一个 DataModule
提供训练、验证和测试数据:
from pytorch_lightning import LightningDataModule
class MyDataModule(LightningDataModule):
def train_dataloader(self):
if self.train_DL is None:
self.train_DL = self._from_indexes_to_dataloader("train_i.tsv")
return self.train_DL
def val_dataloader(self):
if self.valid_DL is None:
self.valid_DL = self._from_indexes_to_dataloader("valid_i.tsv")
return self.valid_DL
def test_dataloader(self):
if self.test_DL is None:
self.test_DL = self._from_indexes_to_dataloader("test_i.tsv")
return self.test_DL
一个外部脚本试图通过以下方式训练模型:
import pytorch_lightning as pl
[...]
model = VGG(config)
trainer = pl.Trainer(gpus=1,max_epochs=config["n_epochs"],progress_bar_refresh_rate=5)
datamodule = MyDataModule(config,logger=log)
trainer.fit(model,datamodule)
但是当我尝试运行它时,pytorch-lightning
告诉我我忘记实现方法 training_step
:
pytorch_lightning.utilities.exceptions.MisconfigurationException:
No `training_step()` method defined.
Lightning `Trainer` expects as minimum a `training_step()`,`train_dataloader()` and `configure_optimizers()` to be defined.
错误是由于他们的检查器 pytorch_lightning.utilities.model_helpers.py
不认为我的 VGG
已经覆盖了 training_step
方法。在其 is_overriden(.)
方法中:
def is_overridden(method_name: str,model: Union[LightningModule,LightningDataModule]) -> bool:
# if you pass DataModule instead of None or a LightningModule,we use LightningDataModule as super
# TODO - refector this function to accept model_name,instance,parent so it makes more sense
super_object = LightningModule if not isinstance(model,LightningDataModule) else LightningDataModule
if not hasattr(model,method_name) or not hasattr(super_object,method_name):
# in case of calling deprecated method
return False
instance_attr = getattr(model,method_name)
if not instance_attr:
return False
super_attr = getattr(super_object,method_name)
# when code pointers are different,it was implemented
if hasattr(instance_attr,'patch_loader_code'):
# cannot pickle __code__ so cannot verify if PatchDataloader
# exists which shows dataloader methods have been overwritten.
# so,we hack it by using the string representation
is_overridden = instance_attr.patch_loader_code != str(super_attr.__code__)
else:
is_overridden = instance_attr.__code__ is not super_attr.__code__
print(method_name)
print(instance_attr.__code__)
print(super_attr.__code__)
print(f'last is_overriden should be True,but it is {is_overridden}')
return is_overridden
最后一次检查失败。我添加的 print
的输出是:
training_step
<code object training_step at 0x7f22e26cc870,file "/opt/conda/lib/python3.8/site-packages/pytorch_lightning/core/lightning.py",line 531>
<code object training_step at 0x7f22e26cc870,line 531>
last is_overriden should be True,but it is False
基本上我的 training_step
类中的 VGG
“指向”其超类的相同代码。有没有人能告诉我为什么它不考虑VGG
来覆盖其父级的方法?
谁能告诉我应该看什么? 我试图在他们的支持论坛上提问,但我的问题和其他许多人一样,仍未得到解答。
VGG
、数据集和 pytorch DataLoader
在没有 pytorch-lightning
和仅 pytorch
的情况下使用时正常工作。
版权声明:本文内容由互联网用户自发贡献,该文观点与技术仅代表作者本人。本站仅提供信息存储空间服务,不拥有所有权,不承担相关法律责任。如发现本站有涉嫌侵权/违法违规的内容, 请发送邮件至 dio@foxmail.com 举报,一经查实,本站将立刻删除。