from_pretrained_fastai does not download files

[pablogps]

Describe the bug

Trying to load a learner from Hugging Face does not get all or any files (the process stalls).

Reproduction

pip install huggingface_hub["fastai"]
from huggingface_hub import from_pretrained_fastai
learner = from_pretrained_fastai("espejelomar/identify-my-cat")

Results in

Fetching 4 files: 0% 0/4 [00:00<?, ?it/s]
model.pkl: 100% 87.5M/87.5M [00:02<00:00, 33.9MB/s]

(I sometimes get one file, sometimes none, but the first progress bar never moves from 0%.)

Logs

There's no error message (the cell never finishes execution).

System info

- huggingface_hub version: 0.20.1
- Platform: Linux-6.1.58+-x86_64-with-glibc2.35
- Python version: 3.10.12
- Running in iPython ?: Yes
- iPython shell: Shell
- Running in notebook ?: Yes
- Running in Google Colab ?: Yes
- Token path ?: /root/.cache/huggingface/token
- Has saved token ?: True
- Who am I ?: Pablogps
- Configured git credential helpers: store
- FastAI: 2.7.13
- Tensorflow: 2.15.0
- Torch: 2.1.0+cu121
- Jinja2: 3.1.2
- Graphviz: 0.20.1
- Pydot: 1.4.2
- Pillow: 9.4.0
- hf_transfer: N/A
- gradio: 4.13.0
- tensorboard: N/A
- numpy: 1.23.5
- pydantic: 2.5.3
- aiohttp: 3.9.1
- ENDPOINT: https://huggingface.co
- HF_HUB_CACHE: /root/.cache/huggingface/hub
- HF_ASSETS_CACHE: /root/.cache/huggingface/assets
- HF_TOKEN_PATH: /root/.cache/huggingface/token
- HF_HUB_OFFLINE: False
- HF_HUB_DISABLE_TELEMETRY: False
- HF_HUB_DISABLE_PROGRESS_BARS: None
- HF_HUB_DISABLE_SYMLINKS_WARNING: False
- HF_HUB_DISABLE_EXPERIMENTAL_WARNING: False
- HF_HUB_DISABLE_IMPLICIT_TOKEN: False
- HF_HUB_ENABLE_HF_TRANSFER: False
- HF_HUB_ETAG_TIMEOUT: 10
- HF_HUB_DOWNLOAD_TIMEOUT: 10

{'huggingface_hub version': '0.20.1',
 'Platform': 'Linux-6.1.58+-x86_64-with-glibc2.35',
 'Python version': '3.10.12',
 'Running in iPython ?': 'Yes',
 'iPython shell': 'Shell',
 'Running in notebook ?': 'Yes',
 'Running in Google Colab ?': 'Yes',
 'Token path ?': '/root/.cache/huggingface/token',
 'Has saved token ?': True,
 'Who am I ?': 'Pablogps',
 'Configured git credential helpers': 'store',
 'FastAI': '2.7.13',
 'Tensorflow': '2.15.0',
 'Torch': '2.1.0+cu121',
 'Jinja2': '3.1.2',
 'Graphviz': '0.20.1',
 'Pydot': '1.4.2',
 'Pillow': '9.4.0',
 'hf_transfer': 'N/A',
 'gradio': '4.13.0',
 'tensorboard': 'N/A',
 'numpy': '1.23.5',
 'pydantic': '2.5.3',
 'aiohttp': '3.9.1',
 'ENDPOINT': 'https://huggingface.co',
 'HF_HUB_CACHE': '/root/.cache/huggingface/hub',
 'HF_ASSETS_CACHE': '/root/.cache/huggingface/assets',
 'HF_TOKEN_PATH': '/root/.cache/huggingface/token',
 'HF_HUB_OFFLINE': False,
 'HF_HUB_DISABLE_TELEMETRY': False,
 'HF_HUB_DISABLE_PROGRESS_BARS': None,
 'HF_HUB_DISABLE_SYMLINKS_WARNING': False,
 'HF_HUB_DISABLE_EXPERIMENTAL_WARNING': False,
 'HF_HUB_DISABLE_IMPLICIT_TOKEN': False,
 'HF_HUB_ENABLE_HF_TRANSFER': False,
 'HF_HUB_ETAG_TIMEOUT': 10,
 'HF_HUB_DOWNLOAD_TIMEOUT': 10}

[Wauplin]

Hi @pablogps, thanks for reporting. The issue doesn't come from from_pretrained_fastai but from a bug between huggingface_hub and google colab integration. A fix has been implemented in #1953 and released in huggingface_hub==0.20.2. However the default version in google colab is 0.20.1 (with the bug). The default version will soon be upgraded but in the meantime you can fix your problem by running !pip install -U huggingface_hub. More details in #1953 (comment).
I'm therefore closing this issue. Sorry for the inconvenience, hopefully solved very soon :)