steps.md

DESI Spectra Zoo Setup Guide

Deploy to HuggingFace Spaces

1. Create a new Space

• Go to huggingface.co/new-space
• Choose a name (e.g. desi-spectra-zoo)
• Select Docker as the SDK
• Set visibility to Public or Private
• Click Create Space

2. Clone the Space and add files

git clone https://huggingface.co/spaces/YOUR_USERNAME/desi-spectra-zoo
cd desi-spectra-zoo
# Copy all project files into this directory:
# app.py, Dockerfile, requirements.txt, dataset_config.yaml, src/
git add .
git commit -m "Initial deploy"
git push

3. Create a HuggingFace dataset for logging

• Go to huggingface.co/new-dataset
• Name it something like desi-spectra-zoo-logs
• This stores annotation state and logs

4. Create a HuggingFace token

• Go to huggingface.co/settings/tokens
• Create a new token with Read & Write access to the dataset repo from step 3

5. Add secrets to the Space

• Go to your Space's Settings page
• Under Repository secrets, add:
  • HF_LOG_REPO_ID = YOUR_USERNAME/desi-spectra-zoo-logs
  • HF_TOKEN = the token from step 4

The Space will automatically rebuild after adding secrets.

6. Local development (optional)

Create a .env file in the project root:

HF_LOG_REPO_ID=YOUR_USERNAME/desi-spectra-zoo-logs
HF_TOKEN=hf_your_token_here

Make sure .env is in .gitignore (it already is by default).

Run locally:

python app.py