Remove unnecessary code for loading ImageNet

code/datasets.py

@@ -15,8 +15,6 @@ import torch
 # make sure your val directory is preprocessed to look like the train directory, e.g. by running this script
 # https://raw.githubusercontent.com/soumith/imagenetloader.torch/master/valprep.sh
 IMAGENET_LOC_ENV = "IMAGENET_DIR"
-IMAGENET_ON_PHILLY_DIR = "IMAGENET_DIR_PHILLY"
-IMAGENET_ON_AZURE_ENV = "IMAGENET_DIR_AZURE"
 
 # list of all datasets
 DATASETS = ["imagenet", "imagenet32", "cifar10"]
@@ -25,10 +23,7 @@ DATASETS = ["imagenet", "imagenet32", "cifar10"]
 def get_dataset(dataset: str, split: str) -> Dataset:
     """Return the dataset as a PyTorch Dataset object"""
     if dataset == "imagenet":
-        if "PT_DATA_DIR" in os.environ or IMAGENET_ON_AZURE_ENV in os.environ: #running on Philly
-            return _imagenet_remote(split)
-        else:
-            return _imagenet(split)
+        return _imagenet(split)
     
     elif dataset == "imagenet32":
         return _imagenet32(split)
@@ -84,32 +79,6 @@ def _cifar10(split: str) -> Dataset:
     else:
         raise Exception("Unknown split name.")
 
-def _imagenet_remote(split: str) -> Dataset:
-        if IMAGENET_ON_AZURE_ENV in os.environ:
-            dir = os.environ[IMAGENET_ON_AZURE_ENV]
-        else:
-            dir = os.environ[IMAGENET_ON_PHILLY_DIR]  
-
-        trainpath = os.path.join(dir, 'train.zip')
-        train_map = os.path.join(dir, 'train_map.txt')
-        valpath = os.path.join(dir, 'val.zip')
-        val_map = os.path.join(dir, 'val_map.txt')
-
-        if split == "train":
-            return ZipData(trainpath, train_map,
-                            transforms.Compose([
-                            transforms.RandomResizedCrop(224),
-                            transforms.RandomHorizontalFlip(),
-                            transforms.ToTensor(),
-                            ]))
-        elif split == "test":
-            return ZipData(valpath, val_map, 
-                            transforms.Compose([
-                            transforms.Resize(256),
-                            transforms.CenterCrop(224),
-                            transforms.ToTensor(),
-                            ]))
-
 
 def _imagenet(split: str) -> Dataset:
     if not IMAGENET_LOC_ENV in os.environ: