How garcia-786/isro_launch_success_predictor works
This is a small Streamlit launch-success prediction app. A user selects launch details and a success threshold; the app loads and preprocesses launch data, splits it, trains a RandomForest classifier, and displays a prediction. It also evaluates the model on held-out data and shows outcome visualizations. The README documents a separate EDA helper. The tree places the Python modules at the repository root, despite the README describing them under scripts/.
Main entry points: app.py Streamlit app · preprocess.py Load and clean · model.py RandomForest model · eda.py EDA plot helper
Components 13
User interaction
- Streamlit app
app.py - Interactive prediction
app.py
Data and modeling
- Launch CSV
- Load and clean
preprocess.py - One-hot encoding
preprocess.py - Train/test split
app.py - Model training
model.py - RandomForest model
model.py
Evaluation and insights
- Test-set evaluation
app.py - Metrics and matrix
app.py - Outcome histogram
app.py - EDA plot helper
eda.py
Other components
- User
Connections 15
- User → Streamlit app: uses
README.md - Streamlit app → Launch CSV: loads
app.py - Streamlit app → Load and clean: calls
app.py - Load and clean → Launch CSV: reads
preprocess.py - Load and clean → One-hot encoding: cleans then encodes
preprocess.py - Streamlit app → Train/test split: splits data
app.py - Streamlit app → Model training: calls
app.py - Model training → RandomForest model: fits
model.py - Streamlit app → Interactive prediction: handles inputs
app.py - Interactive prediction → RandomForest model: predicts with
app.py - Streamlit app → Test-set evaluation: evaluates
app.py - Test-set evaluation → RandomForest model: predicts with
app.py - Test-set evaluation → Metrics and matrix: computes
app.py - Test-set evaluation → Outcome histogram: plots
app.py - User → EDA plot helper: can use
README.md