Upload wilds_downloader.py with huggingface_hub

wilds_downloader.py

@@ -0,0 +1,40 @@
+import ssl
+import certifi
+import urllib.request
+from wilds import get_dataset
+
+# Fix SSL certificate issues
+ssl._create_default_https_context = ssl._create_unverified_context
+
+# List of datasets to download
+datasets_to_download = [
+    # "camelyon17",
+    "iwildcam", 
+    # "fmow",
+    # "globalwheat",
+    # "rxrx1",
+    # "poverty"  # Note: it's "poverty" not "povertymap" in the API
+]
+
+root_dir = "/hdd/datasets/wilds"
+
+print("Starting WILDS dataset downloads...")
+print(f"Root directory: {root_dir}")
+print(f"Datasets to download: {datasets_to_download}\n")
+
+for dataset_name in datasets_to_download:
+    try:
+        print(f"\n{'='*60}")
+        print(f"Downloading {dataset_name}...")
+        print(f"{'='*60}")
+        dataset = get_dataset(dataset=dataset_name, root_dir=root_dir, download=True)
+        print(f"✓ Successfully downloaded {dataset_name}")
+        print(f"  Dataset size: {len(dataset)} samples")
+    except Exception as e:
+        print(f"✗ Error downloading {dataset_name}: {str(e)}")
+        print(f"  You may need to download this dataset manually from https://wilds.stanford.edu/downloads")
+        continue
+
+print(f"\n{'='*60}")
+print("Download process completed!")
+print(f"{'='*60}")