Fix bug when passing metrics as empty list (#660)

CHANGELOG.md

@@ -60,6 +60,8 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 - Fixed a bug where custom samplers would not be properly forwarded to the data loader ([#651](https://github.com/PyTorchLightning/lightning-flash/pull/651))
 
+- Fixed a bug where it was not possible to pass no metrics to the `ImageClassifier` or `TestClassifier` ([#660](https://github.com/PyTorchLightning/lightning-flash/pull/660))
+
 ## [0.4.0] - 2021-06-22
 
 ### Added

flash/core/classification.py

@@ -41,14 +41,15 @@ class ClassificationTask(Task):
     def __init__(
         self,
         *args,
+        num_classes: Optional[int] = None,
         loss_fn: Optional[Callable] = None,
         metrics: Union[torchmetrics.Metric, Mapping, Sequence, None] = None,
         multi_label: bool = False,
         serializer: Optional[Union[Serializer, Mapping[str, Serializer]]] = None,
         **kwargs,
     ) -> None:
         if metrics is None:
-            metrics = torchmetrics.Accuracy(subset_accuracy=multi_label)
+            metrics = torchmetrics.F1(num_classes) if (multi_label and num_classes) else torchmetrics.Accuracy()
 
         if loss_fn is None:
             loss_fn = binary_cross_entropy_with_logits if multi_label else F.cross_entropy

flash/image/classification/model.py

@@ -17,7 +17,7 @@
 import torch
 from torch import nn
 from torch.optim.lr_scheduler import _LRScheduler
-from torchmetrics import Accuracy, F1, Metric
+from torchmetrics import Metric
 
 from flash.core.classification import ClassificationTask, Labels
 from flash.core.data.data_source import DefaultDataKeys
@@ -86,13 +86,14 @@ def __init__(
         serializer: Optional[Union[Serializer, Mapping[str, Serializer]]] = None,
     ):
         super().__init__(
+            num_classes=num_classes,
             model=None,
             loss_fn=loss_fn,
             optimizer=optimizer,
             optimizer_kwargs=optimizer_kwargs,
             scheduler=scheduler,
             scheduler_kwargs=scheduler_kwargs,
-            metrics=metrics or (F1(num_classes) if multi_label else Accuracy()),
+            metrics=metrics,
             learning_rate=learning_rate,
             multi_label=multi_label,
             serializer=serializer or Labels(multi_label=multi_label),

flash/text/classification/model.py

@@ -16,7 +16,7 @@
 from typing import Any, Callable, Dict, List, Mapping, Optional, Sequence, Type, Union
 
 import torch
-from torchmetrics import Accuracy, F1, Metric
+from torchmetrics import Metric
 
 from flash.core.classification import ClassificationTask, Labels
 from flash.core.data.process import Serializer
@@ -67,10 +67,11 @@ def __init__(
         os.environ["PYTHONWARNINGS"] = "ignore"
 
         super().__init__(
+            num_classes=num_classes,
             model=None,
             loss_fn=loss_fn,
             optimizer=optimizer,
-            metrics=metrics or (F1(num_classes) if multi_label else Accuracy()),
+            metrics=metrics,
             learning_rate=learning_rate,
             multi_label=multi_label,
             serializer=serializer or Labels(multi_label=multi_label),

tests/image/classification/test_model.py

@@ -60,17 +60,18 @@ def __len__(self) -> int:
 
 @pytest.mark.skipif(not _IMAGE_TESTING, reason="image libraries aren't installed.")
 @pytest.mark.parametrize(
-    "backbone",
+    "backbone,metrics",
     [
-        "resnet18",
+        ("resnet18", None),
+        ("resnet18", []),
         # "resnet34",
         # "resnet50",
         # "resnet101",
         # "resnet152",
     ],
 )
-def test_init_train(tmpdir, backbone):
-    model = ImageClassifier(10, backbone=backbone)
+def test_init_train(tmpdir, backbone, metrics):
+    model = ImageClassifier(10, backbone=backbone, metrics=metrics)
     train_dl = torch.utils.data.DataLoader(DummyDataset())
     trainer = Trainer(default_root_dir=tmpdir, fast_dev_run=True)
     trainer.finetune(model, train_dl, strategy="freeze_unfreeze")