Add basic image folder style dataset to read directly out of tar files, example in validate.py

timm/data/__init__.py

@@ -1,6 +1,6 @@
 from .constants import *
 from .config import resolve_data_config
-from .dataset import Dataset
+from .dataset import Dataset, DatasetTar
 from .transforms import *
 from .loader import create_loader
 from .mixup import mixup_target, FastCollateMixup

timm/data/dataset.py

@@ -7,6 +7,7 @@ import torch.utils.data as data
 import os
 import re
 import torch
+import tarfile
 from PIL import Image
 
 
@@ -89,3 +90,53 @@ class Dataset(data.Dataset):
                 return [os.path.basename(x[0]) for x in self.imgs]
             else:
                 return [x[0] for x in self.imgs]
+
+
+def _extract_tar_info(tarfile):
+    class_to_idx = {}
+    files = []
+    labels = []
+    for ti in tarfile.getmembers():
+        if not ti.isfile():
+            continue
+        dirname, basename = os.path.split(ti.path)
+        label = os.path.basename(dirname)
+        class_to_idx[label] = None
+        ext = os.path.splitext(basename)[1]
+        if ext.lower() in IMG_EXTENSIONS:
+            files.append(ti)
+            labels.append(label)
+    for idx, c in enumerate(sorted(class_to_idx.keys(), key=natural_key)):
+        class_to_idx[c] = idx
+    tarinfo_and_targets = zip(files, [class_to_idx[l] for l in labels])
+    tarinfo_and_targets = sorted(tarinfo_and_targets, key=lambda k: natural_key(k[0].path))
+    return tarinfo_and_targets
+
+
+class DatasetTar(data.Dataset):
+
+    def __init__(self, root, load_bytes=False, transform=None):
+
+        assert os.path.isfile(root)
+        self.root = root
+        with tarfile.open(root) as tf:  # cannot keep this open across processes, reopen later
+            self.imgs = _extract_tar_info(tf)
+        self.tarfile = None  # lazy init in __getitem__
+        self.load_bytes = load_bytes
+        self.transform = transform
+
+    def __getitem__(self, index):
+        if self.tarfile is None:
+            self.tarfile = tarfile.open(self.root)
+        tarinfo, target = self.imgs[index]
+        iob = self.tarfile.extractfile(tarinfo)
+        img = iob.read() if self.load_bytes else Image.open(iob).convert('RGB')
+        if self.transform is not None:
+            img = self.transform(img)
+        if target is None:
+            target = torch.zeros(1).long()
+        return img, target
+
+    def __len__(self):
+        return len(self.imgs)
+

validate.py

@@ -14,7 +14,7 @@ import torch.nn.parallel
 from collections import OrderedDict
 
 from timm.models import create_model, apply_test_time_pool, load_checkpoint, is_model, list_models
-from timm.data import Dataset, create_loader, resolve_data_config
+from timm.data import Dataset, DatasetTar, create_loader, resolve_data_config
 from timm.utils import accuracy, AverageMeter, natural_key, setup_default_logging
 
 torch.backends.cudnn.benchmark = True
@@ -24,7 +24,7 @@ parser.add_argument('data', metavar='DIR',
                     help='path to dataset')
 parser.add_argument('--model', '-m', metavar='MODEL', default='dpn92',
                     help='model architecture (default: dpn92)')
-parser.add_argument('-j', '--workers', default=2, type=int, metavar='N',
+parser.add_argument('-j', '--workers', default=4, type=int, metavar='N',
                     help='number of data loading workers (default: 2)')
 parser.add_argument('-b', '--batch-size', default=256, type=int,
                     metavar='N', help='mini-batch size (default: 256)')
@@ -91,9 +91,14 @@ def validate(args):
 
     criterion = nn.CrossEntropyLoss().cuda()
 
+    if os.path.splitext(args.data)[1] == '.tar' and os.path.isfile(args.data):
+        dataset = DatasetTar(args.data, load_bytes=args.tf_preprocessing)
+    else:
+        dataset = Dataset(args.data, load_bytes=args.tf_preprocessing)
+
     crop_pct = 1.0 if test_time_pool else data_config['crop_pct']
     loader = create_loader(
-        Dataset(args.data, load_bytes=args.tf_preprocessing),
+        dataset,
         input_size=data_config['input_size'],
         batch_size=args.batch_size,
         use_prefetcher=args.prefetcher,