Update README.md

README.md

@@ -60,27 +60,78 @@ All images are 96x96 pixels and are provided in PNG format.
 
 ## 🧼 Clean STL-10 Dataset, 🔧 How to Load?
 
-🎉 We uploaded our cleaned STL-10 dataset to Hugging Face! You can easily load and use it with the 🤗 `datasets` library.
+🎉 We uploaded our cleaned STL-10 dataset to Hugging Face! You can easily load and use it with the 🤗 `webdataset` library.
 
+---
 
-### 🔍 Load with 🤗 `datasets`(Recommended)
-
-You can load the dataset just like any other Hugging Face dataset:
+### 🔧 Load with WebDataset
 
 ```python
-from datasets import load_dataset
+from huggingface_hub import hf_hub_url, HfFileSystem
+import webdataset as wds
+
+# Replace with your actual Hugging Face token
+token = "your_huggingface_token_here"
+fs = HfFileSystem(token=token)
+
+repo_id = "Shu1L0n9/CleanSTL-10"
+splits = {
+    'train': '**/train-*.tar',
+    'test': '**/test-*.tar',
+    'unlabeled': '**/unlabeled-*.tar'
+}
 
-dataset = load_dataset("Shu1L0n9/CleanSTL10-WebDataset", split="train")
+def load_split(split):
+    pattern = f"hf://datasets/{repo_id}/{splits[split]}"
+    files = [fs.resolve_path(p) for p in fs.glob(pattern)]
+    urls = [hf_hub_url(f.repo_id, f.path_in_repo, repo_type="dataset") for f in files]
+    pipe_url = f"pipe: curl -s -L -H 'Authorization:Bearer {token}' {'::'.join(urls)}"
+    return wds.WebDataset(pipe_url, shardshuffle=False).decode("pil")
+
+# Load the splits
+train_ds = load_split("train")
+test_ds = load_split("test")
+unlabeled_ds = load_split("unlabeled")
 ```
 
-### 🚀 Use as WebDataset (e.g. for PyTorch streaming training)
+> ℹ️ Requires: `webdataset`, `huggingface_hub`, `fsspec`
+> Install with:
 
-```python
-import webdataset as wds
+```bash
+pip install webdataset huggingface_hub fsspec
+```
+
+---
 
-train_ds = wds.WebDataset("https://huggingface.co/datasets/Shu1L0n9/CleanSTL10/resolve/main/cleanstl10-train.tar").decode("pil").to_tuple("jpg", "cls")
-test_ds = wds.WebDataset("https://huggingface.co/datasets/Shu1L0n9/CleanSTL10/resolve/main/cleanstl10-test.tar").decode("pil").to_tuple("jpg", "cls")
-unlabeled_ds = wds.WebDataset("https://huggingface.co/datasets/Shu1L0n9/CleanSTL10/resolve/main/cleanstl10-unlabeled.tar").decode("pil").to_tuple("jpg", "cls")
+### 🔑 How to Get Your Hugging Face Token
+
+To download from Hugging Face with authentication, you’ll need a **User Access Token**:
+
+1. Visit [https://huggingface.co/settings/tokens](https://huggingface.co/settings/tokens)
+2. Click **“New token”**
+3. Choose a name and select **“Read”** permission
+4. Click **“Generate”**, then copy the token
+5. Paste it into your script:
+
+   ```python
+   token = "your_token_here"
+   ```
+
+> ⚠️ **Keep your token private** and avoid hardcoding it in shared scripts.
+
+#### 💡 Optional: Use Environment Variable
+
+To avoid hardcoding your token:
+
+```bash
+export HF_TOKEN=your_token_here
+```
+
+Then in your Python script:
+
+```python
+import os
+token = os.getenv("HF_TOKEN")
 ```
 
 ## Citation