Capstone 4 of 4
A model on HuggingFace
A model on the HuggingFace Hub plus a live Space that demos it.
Fine-tune or adapt a small model and share it. Train on your own machine or a free notebook, push the result to the HuggingFace Hub with a short model card, and build a Space that lets anyone try it. This one runs on your own machine and free HuggingFace hosting, not inside PythonCook.
Before you submit
How to ship it
HuggingFace Hub and Spaces
- 1
Sign in from your machine
Install the tools with pip install huggingface_hub, then run huggingface-cli login and paste a token from your HuggingFace account settings.
- 2
Push the model
Call push_to_hub("your-name/your-model") on your trained model and tokenizer. This creates the public model page.
- 3
Write the model card
Edit the README on the model page to say what the model does, what it was trained on, and how to load it.
- 4
Build the Space
Create a public Gradio Space that loads your model and offers one input and one output, then confirm the Space URL loads in a private window.
Submit and verify
Paste your public links. PythonCook checks them and, if anything is not ready, tells you the exact next step.
