switch to torchmetrics (#169)

* switch to torchmetrics

* 0.2.0

* mnist

* cache

README.md

@@ -244,7 +244,7 @@ To illustrate, say we want to build a model to predict if a passenger survived o
 
 ```python
 # import our libraries
-from pytorch_lightning.metrics.classification import Accuracy, Precision, Recall
+from torchmetrics.classification import Accuracy, Precision, Recall
 import flash
 from flash import download_data
 from flash.tabular import TabularClassifier, TabularData

docs/source/reference/tabular_classification.rst

@@ -47,7 +47,7 @@ Next, we create the :class:`~flash.tabular.TabularClassifier` task, using the Da
     import flash
     from flash import download_data
     from flash.tabular import TabularClassifier, TabularData
-    from pytorch_lightning.metrics.classification import Accuracy, Precision, Recall
+    from torchmetrics.classification import Accuracy, Precision, Recall
 
     # 1. Download the data
     download_data("https://pl-flash-data.s3.amazonaws.com/titanic.zip", 'data/')

flash/tabular/classification/model.py

@@ -14,9 +14,10 @@
 from typing import Any, Callable, List, Optional, Tuple, Type
 
 import torch
-from pytorch_lightning.metrics import Metric
 from pytorch_tabnet.tab_network import TabNet
 from torch.nn import functional as F
+from torch.nn.functional import softmax
+from torchmetrics import Metric
 
 from flash.core.classification import ClassificationTask
 from flash.core.data import DataPipeline
@@ -85,7 +86,7 @@ def predict(
     def forward(self, x_in):
         # TabNet takes single input, x_in is composed of (categorical, numerical)
         x = torch.cat([x for x in x_in if x.numel()], dim=1)
-        return self.model(x)[0]
+        return softmax(self.model(x)[0])
 
     @classmethod
     def from_data(cls, datamodule, **kwargs) -> 'TabularClassifier':

flash/text/classification/model.py

@@ -16,7 +16,7 @@
 from typing import Callable, Mapping, Sequence, Type, Union
 
 import torch
-from pytorch_lightning.metrics.classification import Accuracy
+from torchmetrics.classification import Accuracy
 from transformers import BertForSequenceClassification
 
 from flash.core.classification import ClassificationDataPipeline, ClassificationTask

flash/text/seq2seq/summarization/metric.py

@@ -15,9 +15,9 @@
 
 import numpy as np
 import torch
-from pytorch_lightning.metrics import Metric
 from rouge_score import rouge_scorer, scoring
 from rouge_score.scoring import AggregateScore, Score
+from torchmetrics import Metric
 
 from flash.text.seq2seq.summarization.utils import add_newline_to_end_of_each_sentence
 

flash/text/seq2seq/translation/metric.py

@@ -20,7 +20,7 @@
 from typing import List
 
 import torch
-from pytorch_lightning.metrics import Metric
+from torchmetrics import Metric
 
 
 def _count_ngram(ngram_input_list: List[str], n_gram: int) -> Counter:

flash/vision/classification/model.py

@@ -14,9 +14,10 @@
 from typing import Any, Callable, Mapping, Sequence, Type, Union
 
 import torch
-from pytorch_lightning.metrics import Accuracy
 from torch import nn
 from torch.nn import functional as F
+from torch.nn.functional import softmax
+from torchmetrics import Accuracy
 
 from flash.core.classification import ClassificationTask
 from flash.vision.backbones import backbone_and_num_features
@@ -33,7 +34,7 @@ class ImageClassifier(ClassificationTask):
         loss_fn: Loss function for training, defaults to :func:`torch.nn.functional.cross_entropy`.
         optimizer: Optimizer to use for training, defaults to :class:`torch.optim.SGD`.
         metrics: Metrics to compute for training and evaluation,
-            defaults to :class:`pytorch_lightning.metrics.Accuracy`.
+            defaults to :class:`torchmetrics.Accuracy`.
         learning_rate: Learning rate to use for training, defaults to ``1e-3``.
     """
 
@@ -67,7 +68,7 @@ def __init__(
 
     def forward(self, x) -> Any:
         x = self.backbone(x)
-        return self.head(x)
+        return softmax(self.head(x))
 
     @staticmethod
     def default_pipeline() -> ImageClassificationDataPipeline:

flash/vision/embedding/image_embedder_model.py

@@ -14,11 +14,11 @@
 from typing import Any, Callable, Mapping, Optional, Sequence, Type, Union
 
 import torch
-from pytorch_lightning.metrics import Accuracy
 from pytorch_lightning.utilities.distributed import rank_zero_warn
 from pytorch_lightning.utilities.exceptions import MisconfigurationException
 from torch import nn
 from torch.nn import functional as F
+from torchmetrics import Accuracy
 
 from flash.core import Task
 from flash.core.data import TaskDataPipeline

flash_examples/finetuning/tabular_classification.py

@@ -11,7 +11,7 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-from pytorch_lightning.metrics.classification import Accuracy, Precision, Recall
+from torchmetrics.classification import Accuracy, Precision, Recall
 
 import flash
 from flash.core.data import download_data

flash_examples/generic_task.py

@@ -11,23 +11,33 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
+import os
+import urllib
+
 import pytorch_lightning as pl
 from torch import nn, optim
 from torch.utils.data import DataLoader, random_split
 from torchvision import datasets, transforms
 
 from flash import ClassificationTask
 
+_PATH_ROOT = os.path.dirname(os.path.dirname(__file__))
+# TorchVision hotfix https://github.com/pytorch/vision/issues/1938
+opener = urllib.request.build_opener()
+opener.addheaders = [('User-agent', 'Mozilla/5.0')]
+urllib.request.install_opener(opener)
+
 # 1. Load a basic backbone
 model = nn.Sequential(
     nn.Flatten(),
     nn.Linear(28 * 28, 128),
     nn.ReLU(),
     nn.Linear(128, 10),
+    nn.Softmax(),
 )
 
 # 2. Load a dataset
-dataset = datasets.MNIST('./data', download=True, transform=transforms.ToTensor())
+dataset = datasets.MNIST(os.path.join(_PATH_ROOT, 'data'), download=True, transform=transforms.ToTensor())
 
 # 3. Split the data randomly
 train, val, test = random_split(dataset, [50000, 5000, 5000])  # type: ignore

flash_notebooks/tabular_classification.ipynb

@@ -50,7 +50,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from pytorch_lightning.metrics.classification import Accuracy, Precision, Recall\n",
+    "from torchmetrics.classification import Accuracy, Precision, Recall\n",
     "\n",
     "import flash\n",
     "from flash.core.data import download_data\n",

requirements.txt

@@ -2,6 +2,7 @@ pytorch-lightning==1.2.0rc0  # todo: we shall align with real 1.2
 torch>=1.7  # TODO: regenerate weights with lewer PT version
 PyYAML>=5.1
 Pillow>=7.2
+torchmetrics>=0.2.0
 torchvision>=0.8  # lower to 0.7 after PT 1.6
 transformers>=4.0
 pytorch-tabnet==3.1

tests/__init__.py

@@ -0,0 +1,6 @@
+import urllib
+
+# TorchVision hotfix https://github.com/pytorch/vision/issues/1938
+opener = urllib.request.build_opener()
+opener.addheaders = [('User-agent', 'Mozilla/5.0')]
+urllib.request.install_opener(opener)