How bashutosh017/iris-flower-variant-prediction works
This is a small Iris-species classification app with two workflows: a training script loads and splits the Iris dataset, trains and evaluates a random forest, then saves it; at runtime, users enter four measurements in Streamlit, which sends them to a FastAPI endpoint for prediction. The API loads the saved model and maps its output to a species name. The source excerpts support these links; the serialized model is present in the tree but its contents were not inspected.
Main entry points: server.py Prediction endpoint · model.py Train and evaluate · iris_flower.py Measurement form · iris_model.pkl Classifier artifact
Components 8
Model training
- Iris dataset
Iris Flower Model/model.py - Train and evaluate
Iris Flower Model/model.py
Prediction runtime
- Measurement form
Iris Flower Model/iris_flower.py - Prediction endpoint
Iris Flower Model/server.py - Random forest
Iris Flower Model/server.py
Saved model
- Classifier artifact
Iris Flower Model/iris_model.pkl
Other components
- Operator
- Flower user
Connections 9
- Operator → Train and evaluate: runs
Iris Flower Model/model.py - Train and evaluate → Iris dataset: loads
Iris Flower Model/model.py - Train and evaluate → Classifier artifact: saves
Iris Flower Model/model.py - Flower user → Measurement form: enters measurements
Iris Flower Model/iris_flower.py - Measurement form → Prediction endpoint: posts measurements
Iris Flower Model/iris_flower.py - Prediction endpoint → Classifier artifact: loads
Iris Flower Model/server.py - Prediction endpoint → Random forest: predicts with
Iris Flower Model/server.py - Prediction endpoint → Iris dataset: loads class names
Iris Flower Model/server.py - Prediction endpoint → Measurement form: returns species
Iris Flower Model/iris_flower.py