chiaseacc's picture
Upload folder using huggingface_hub
c67e924 verified
|
Raw History Blame Contribute Delete
1.16 kB
# Uploading to Hugging Face
1. Create a **write** token at https://huggingface.co/settings/tokens.
2. Replace `<you>` (GitHub) and `<user>` (Hugging Face) placeholders in `README.md`.
## Option A: script
```bash
export HF_TOKEN=hf_xxx
./upload.sh <your-hf-username>
```
## Option B: Python
```python
import os
from huggingface_hub import HfApi
api = HfApi(token=os.environ["HF_TOKEN"])
repo = "<user>/teencode-normalizer-v3"
api.create_repo(repo, repo_type="model", exist_ok=True) # add private=True to start private
api.upload_folder(folder_path=".", repo_id=repo, repo_type="model",
ignore_patterns=["upload.sh", "UPLOAD.md"], commit_message="Upload teencode-normalizer v3")
```
## Option C: web UI
1. https://huggingface.co/new → name `teencode-normalizer-v3`, license MIT, Public or Private → Create.
2. Files and versions → Add file → Upload files. Drag everything in this folder **keeping the `onnx/` folder** (drag the folder itself). Files over 10 MB are stored via LFS automatically.
3. Commit. The model card renders from `README.md`.
Check: the model page should show tags, the MIT license and the CER metrics.