How greco1899/scrape_ufc_stats works
Project type and purpose
• This is a data‑pipeline / ETL‑style Python scraper project whose sole purpose is to fetch historical and incremental UFC events, fights, stats, and fighter “tale of the tape” data from ufcstats.com and store it in CSV files for downstream analysis or ML.
• There is no traditional frontend; it is purely a backend/scraper service, orchestrated via notebooks and scripts, with scheduled runs on PythonAnywhere.
File structure
• Top‑level files:
• scrape_ufc_stats_library.py: core scraping and parsing logic encapsulated as reusable functions
• multiple Jupyter notebooks (*.ipynb) illustrating full‑historical and single‑run workflows
• scrape_ufc_stats_unparsed_data.py: incremental refresh script
• scrape_ufc_stats_config.yaml: centralized configuration for URLs, output directories, etc.
• requirements.txt: Python dependencies
• Data output CSVs live at the repo root after each run.
• No database—flat files only.
• Scheduler: described in the README as running daily on PythonAnywhere.
README for additional insights
• Describes two modes:
a) Full‑historical scrape (hours‑long notebooks)
b) Incremental daily refresh (script runs in minutes)
• Notes use of PythonAnywhere’s scheduler to automatically execute scrape_ufc_stats_unparsed_data.py and push updated CSVs back to GitHub.
• Future roadmap: cleaning, exploration, ML prediction—but out of scope for now.