Will I Catch That Train

◉ A Real-Time J/Z/M NYC Subway Tracker & Commute Optimizer

◉ A Streamlit web app that helps you decide when to leave home for the J, Z, M subway lines in NYC.

◉ It combines static GTFS schedules with real-time GTFS-RT feeds to draw trains moving on the map and to compute a “Leave-Now” suggestion.

---

◉ Features

• Static Map Plots only J/Z/M routes on a dark map (via Folium / Pydeck).
• Live Positions Interpolates train positions between stops using live arrival times (no GPS required).
• Leave-Now Assistant Calculates when to leave home based on walking time + buffer vs upcoming train arrivals.
• Platform-specific Stop IDs Supports separate northbound / southbound platforms (e.g., M11S).
• Feed-agnostic Works without API keys — uses the public MTA GTFS-RT endpoints.
• Deploy-ready Runs locally or on Hugging Face Spaces as a Streamlit app.

◉ Data Sources

• Static GTFS → schedules, stops, shapes (downloaded zip you placed in data/).

• Realtime feeds → public MTA endpoints.

• J/Z → https://api-endpoint.mta.info/Dataservice/mtagtfsfeeds/nyct/gtfs-jz.

• B/D/F/M (for M) → https://api-endpoint.mta.info/Dataservice/mtagtfsfeeds/nyct/gtfs-bdfm.

• Some CDN edges still return 403 for the slash-based URL; the code automatically retries with the %2F-encoded URL.

◉ Tech Stack

• Frontend / UI: Streamlit, Pydeck (or Folium)
• Realtime data: MTA GTFS-RT (protobuf via gtfs-realtime-bindings)
• Static data: MTA GTFS static bundle (CSV in zip)
• Computation: Python 3.9+, pandas, numpy, shapely
• Hosting: HuggingFace Spaces
• Version control: Git + GitHub / HF Repo

◉ Workflow

### 1. Clone & enter project
git clone https://github.com/<your-username>/will-i-catch-that-train.git
cd will-i-catch-that-train

### 2. Set up Python env
python -m venv venv
venv\Scripts\activate       
pip install -r requirements.txt

### 3. Place the static GTFS bundle
# Download the latest MTA Subway static GTFS zip from https://new.mta.info/developers
# Put the untouched zip file as:
data/nyc_gtfs_static.zip

### 4. Run the app
streamlit run streamlit_app.py