Shortcuts

Evaluation

During and after training we need a way to evaluate our models to make sure they are not overfitting while training and generalize well on unseen or real-world data. There are generally 2 stages of evaluation: validation and testing. To some degree they serve the same purpose, to make sure models works on real data but they have some practical differences.

Validation is usually done during training, traditionally after each training epoch. It can be used for hyperparameter optimization or tracking model performance during training. It’s a part of the training process.

Testing is usually done once we are satisfied with the training and only with the best model selected from the validation metrics.

Let’s see how these can be performed with Lightning.

Testing

Lightning allows the user to test their models with any compatible test dataloaders. This can be done before/after training and is completely agnostic to fit() call. The logic used here is defined under test_step().

Testing is performed using the Trainer object’s .test() method.

Trainer.test(model=None, dataloaders=None, ckpt_path=None, verbose=True, datamodule=None)[source]

Perform one evaluation epoch over the test set. It’s separated from fit to make sure you never run on your test set until you want to.

Parameters
Return type

List[Dict[str, float]]

Returns

List of dictionaries with metrics logged during the test phase, e.g., in model- or callback hooks like test_step(), test_epoch_end(), etc. The length of the list corresponds to the number of test dataloaders used.

Test after Fit

To run the test set after training completes, use this method.

# run full training
trainer.fit(model)

# (1) load the best checkpoint automatically (lightning tracks this for you)
trainer.test(ckpt_path="best")

# (2) test using a specific checkpoint
trainer.test(ckpt_path="/path/to/my_checkpoint.ckpt")

# (3) test with an explicit model (will use this model and not load a checkpoint)
trainer.test(model)

Warning

It is recommended to test with Trainer(devices=1) since distributed strategies such as DDP use DistributedSampler internally, which replicates some samples to make sure all devices have same batch size in case of uneven inputs. This is helpful to make sure benchmarking for research papers is done the right way.

Test Multiple Models

You can run the test set on multiple models using the same trainer instance.

model1 = LitModel()
model2 = GANModel()

trainer = Trainer()
trainer.test(model1)
trainer.test(model2)

Test Pre-Trained Model

To run the test set on a pre-trained model, use this method.

model = MyLightningModule.load_from_checkpoint(
    checkpoint_path="/path/to/pytorch_checkpoint.ckpt",
    hparams_file="/path/to/test_tube/experiment/version/hparams.yaml",
    map_location=None,
)

# init trainer with whatever options
trainer = Trainer(...)

# test (pass in the model)
trainer.test(model)

In this case, the options you pass to trainer will be used when running the test set (ie: 16-bit, dp, ddp, etc…)

Test with Additional DataLoaders

You can still run inference on a test dataset even if the test_dataloader() method hasn’t been defined within your lightning module instance. This would be the case when your test data is not available at the time your model was declared.

# setup your data loader
test_dataloader = DataLoader(...)

# test (pass in the loader)
trainer.test(dataloaders=test_dataloader)

You can either pass in a single dataloader or a list of them. This optional named parameter can be used in conjunction with any of the above use cases. Additionally, you can also pass in an datamodules that have overridden the test_dataloader method.

class MyDataModule(pl.LightningDataModule):
    ...

    def test_dataloader(self):
        return DataLoader(...)


# setup your datamodule
dm = MyDataModule(...)

# test (pass in datamodule)
trainer.test(datamodule=dm)

Validation

Lightning allows the user to validate their models with any compatible val dataloaders. This can be done before/after training. The logic associated to the validation is defined within the validation_step().

Apart from this .validate has same API as .test, but would rely respectively on validation_step() and test_step().

Note

.validate method uses the same validation logic being used under validation happening within fit() call.

Warning

When using trainer.validate(), it is recommended to use Trainer(devices=1) since distributed strategies such as DDP uses DistributedSampler internally, which replicates some samples to make sure all devices have same batch size in case of uneven inputs. This is helpful to make sure benchmarking for research papers is done the right way.

Trainer.validate(model=None, dataloaders=None, ckpt_path=None, verbose=True, datamodule=None)[source]

Perform one evaluation epoch over the validation set.

Parameters
Return type

List[Dict[str, float]]

Returns

List of dictionaries with metrics logged during the validation phase, e.g., in model- or callback hooks like validation_step(), validation_epoch_end(), etc. The length of the list corresponds to the number of validation dataloaders used.

Read the Docs v: stable
Versions
latest
stable
1.6.3
1.6.2
1.6.1
1.6.0
1.5.10
1.5.9
1.5.8
1.5.7
1.5.6
1.5.5
1.5.4
1.5.3
1.5.2
1.5.1
1.5.0
1.4.9
1.4.8
1.4.7
1.4.6
1.4.5
1.4.4
1.4.3
1.4.2
1.4.1
1.4.0
1.3.8
1.3.7
1.3.6
1.3.5
1.3.4
1.3.3
1.3.2
1.3.1
1.3.0
1.2.10
1.2.8
1.2.7
1.2.6
1.2.5
1.2.4
1.2.3
1.2.2
1.2.1
1.2.0
1.1.8
1.1.7
1.1.6
1.1.5
1.1.4
1.1.3
1.1.2
1.1.1
1.1.0
1.0.8
1.0.7
1.0.6
1.0.5
1.0.4
1.0.3
1.0.2
1.0.1
1.0.0
0.10.0
0.9.0
0.8.5
0.8.4
0.8.3
0.8.2
0.8.1
0.8.0
0.7.6
0.7.5
0.7.4
0.7.3
0.7.2
0.7.1
0.7.0
0.6.0
0.5.3
0.4.9
docs_2
Downloads
html
On Read the Docs
Project Home
Builds

Free document hosting provided by Read the Docs.