Merge f435f8f into 592b580

flash/data/data_utils.py

@@ -1,9 +1,15 @@
-from typing import Dict, List, Union
+from typing import Any, Dict, List, Union
 
 import pandas as pd
 
 
-def labels_from_categorical_csv(csv: str, index_col: str, return_dict: bool = True) -> Union[Dict, List]:
+def labels_from_categorical_csv(
+        csv: str,
+        index_col: str,
+        feature_cols: List,
+        return_dict: bool = True,
+        index_col_collate_fn: Any = None
+) -> Union[Dict, List]:
     """
     Returns a dictionary with {index_col: label} for each entry in the csv.
 
@@ -17,10 +23,15 @@ def labels_from_categorical_csv(csv: str, index_col: str, return_dict: bool = Tr
     df = pd.read_csv(csv)
     # get names
     names = df[index_col].to_list()
-    del df[index_col]
+
+    # apply colate fn to index_col
+    if index_col_collate_fn:
+        for i in range(len(names)):
+            names[i] = index_col_collate_fn(names[i])
 
     # everything else is binary
-    labels = df.to_numpy().argmax(1).tolist()
+    feature_df = df[feature_cols]
+    labels = feature_df.to_numpy().argmax(1).tolist()
 
     if return_dict:
         labels = {name: label for name, label in zip(names, labels)}

flash/vision/classification/data.py

@@ -77,7 +77,8 @@ def __getitem__(self, index: int) -> Tuple[Any, Optional[int]]:
             img = self.transform(img)
         label = None
         if self.has_dict_labels:
-            name = os.path.basename(filename)
+            name = os.path.splitext(filename)[0]
+            name = os.path.basename(name)
             label = self.labels[name]
 
         elif self.has_labels:
@@ -256,6 +257,7 @@ def from_filepaths(
         train_filepaths: Union[str, Optional[Sequence[Union[str, pathlib.Path]]]] = None,
         train_labels: Optional[Sequence] = None,
         train_transform: Optional[Callable] = _default_train_transforms,
+        valid_split: Union[None, float] = None,
         valid_filepaths: Union[str, Optional[Sequence[Union[str, pathlib.Path]]]] = None,
         valid_labels: Optional[Sequence] = None,
         valid_transform: Optional[Callable] = _default_valid_transforms,
@@ -264,6 +266,7 @@ def from_filepaths(
         loader: Callable = _pil_loader,
         batch_size: int = 64,
         num_workers: Optional[int] = None,
+        seed: int = 1234,
         **kwargs
     ):
         """Creates a ImageClassificationData object from lists of image filepaths and labels
@@ -272,6 +275,7 @@ def from_filepaths(
             train_filepaths: string or sequence of file paths for training dataset. Defaults to ``None``.
             train_labels: sequence of labels for training dataset. Defaults to ``None``.
             train_transform: transforms for training dataset. Defaults to ``None``.
+            valid_split: if not None, generates val split from train dataloader using this value.
             valid_filepaths: string or sequence of file paths for validation dataset. Defaults to ``None``.
             valid_labels: sequence of labels for validation dataset. Defaults to ``None``.
             valid_transform: transforms for validation and testing dataset. Defaults to ``None``.
@@ -281,6 +285,7 @@ def from_filepaths(
             batch_size: the batchsize to use for parallel loading. Defaults to ``64``.
             num_workers: The number of workers to use for parallelized loading.
                 Defaults to ``None`` which equals the number of available CPU threads.
+            seed: Used for the train/val splits when valid_split is not None
 
         Returns:
             ImageClassificationData: The constructed data module.
@@ -319,14 +324,25 @@ def from_filepaths(
             loader=loader,
             transform=train_transform,
         )
-        valid_ds = (
-            FilepathDataset(
-                filepaths=valid_filepaths,
-                labels=valid_labels,
-                loader=loader,
-                transform=valid_transform,
-            ) if valid_filepaths is not None else None
-        )
+
+        if valid_split:
+            full_length = len(train_ds)
+            train_split = int((1.0 - valid_split) * full_length)
+            valid_split = full_length - train_split
+            train_ds, valid_ds = torch.utils.data.random_split(
+                train_ds,
+                [train_split, valid_split],
+                generator=torch.Generator().manual_seed(seed)
+            )
+        else:
+            valid_ds = (
+                FilepathDataset(
+                    filepaths=valid_filepaths,
+                    labels=valid_labels,
+                    loader=loader,
+                    transform=valid_transform,
+                ) if valid_filepaths is not None else None
+            )
 
         test_ds = (
             FilepathDataset(

tests/vision/classification/test_data.py

@@ -93,27 +93,33 @@ def test_categorical_csv_labels(tmpdir):
     train_csv = os.path.join(tmpdir, 'some_dataset', 'train.csv')
     text_file = open(train_csv, 'w')
     text_file.write(
-        'my_id, label_a, label_b, label_c\n"train_1.png", 0, 1, 0\n"train_2.png", 0, 0, 1\n"train_2.png", 1, 0, 0\n'
+        'my_id,label_a,label_b,label_c\n"train_1.png", 0, 1, 0\n"train_2.png", 0, 0, 1\n"train_2.png", 1, 0, 0\n'
     )
     text_file.close()
 
     valid_csv = os.path.join(tmpdir, 'some_dataset', 'valid.csv')
     text_file = open(valid_csv, 'w')
     text_file.write(
-        'my_id, label_a, label_b, label_c\n"valid_1.png", 0, 1, 0\n"valid_2.png", 0, 0, 1\n"valid_3.png", 1, 0, 0\n'
+        'my_id,label_a,label_b,label_c\n"valid_1.png", 0, 1, 0\n"valid_2.png", 0, 0, 1\n"valid_3.png", 1, 0, 0\n'
     )
     text_file.close()
 
     test_csv = os.path.join(tmpdir, 'some_dataset', 'test.csv')
     text_file = open(test_csv, 'w')
     text_file.write(
-        'my_id, label_a, label_b, label_c\n"test_1.png", 0, 1, 0\n"test_2.png", 0, 0, 1\n"test_3.png", 1, 0, 0\n'
+        'my_id,label_a,label_b,label_c\n"test_1.png", 0, 1, 0\n"test_2.png", 0, 0, 1\n"test_3.png", 1, 0, 0\n'
     )
     text_file.close()
 
-    train_labels = labels_from_categorical_csv(train_csv, 'my_id')
-    valid_labels = labels_from_categorical_csv(valid_csv, 'my_id')
-    test_labels = labels_from_categorical_csv(test_csv, 'my_id')
+    def index_col_collate_fn(x):
+        return os.path.splitext(x)[0]
+
+    train_labels = labels_from_categorical_csv(
+        train_csv, 'my_id', feature_cols=['label_a', 'label_b', 'label_c'], index_col_collate_fn=index_col_collate_fn)
+    valid_labels = labels_from_categorical_csv(
+        valid_csv, 'my_id', feature_cols=['label_a', 'label_b', 'label_c'], index_col_collate_fn=index_col_collate_fn)
+    test_labels = labels_from_categorical_csv(
+        test_csv, 'my_id', feature_cols=['label_a', 'label_b', 'label_c'], index_col_collate_fn=index_col_collate_fn)
 
     data = ImageClassificationData.from_filepaths(
         batch_size=2,
@@ -134,6 +140,16 @@ def test_categorical_csv_labels(tmpdir):
     for (x, y) in data.test_dataloader():
         assert len(x) == 2
 
+    data = ImageClassificationData.from_filepaths(
+        batch_size=2,
+        train_filepaths=os.path.join(tmpdir, 'some_dataset', 'train'),
+        train_labels=train_labels,
+        valid_split=0.5
+    )
+
+    for (x, y) in data.val_dataloader():
+        assert len(x) == 1
+
 
 def test_from_folders(tmpdir):
     train_dir = Path(tmpdir / "train")