app.py

import gradio as gr
import numpy as np
from sklearn.datasets import load_iris
from sklearn.ensemble import GradientBoostingClassifier
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score, confusion_matrix

# 1. Load dataset
iris = load_iris()
X, y = iris.data, iris.target
feature_names = iris.feature_names
class_names = iris.target_names

# Split into train/test
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.3, random_state=42
)

# 2. Define a function that takes hyperparameters and returns model accuracy + confusion matrix
def train_and_evaluate(learning_rate, n_estimators, max_depth):
    # Train model
    clf = GradientBoostingClassifier(
        learning_rate=learning_rate,
        n_estimators=n_estimators,
        max_depth=int(max_depth),
        random_state=42
    )
    clf.fit(X_train, y_train)

    # Predict on test data
    y_pred = clf.predict(X_test)

    # Calculate metrics
    accuracy = accuracy_score(y_test, y_pred)
    cm = confusion_matrix(y_test, y_pred)

    # Convert confusion matrix to a more display-friendly format
    cm_display = ""
    for row in cm:
        cm_display += str(row) + "\n"

    return f"Accuracy: {accuracy:.3f}\nConfusion Matrix:\n{cm_display}"

# 3. Define a prediction function for user-supplied feature values
def predict_species(sepal_length, sepal_width, petal_length, petal_width,
                    learning_rate, n_estimators, max_depth):
    # Train a new model using same hyperparams
    clf = GradientBoostingClassifier(
        learning_rate=learning_rate,
        n_estimators=n_estimators,
        max_depth=int(max_depth),
        random_state=42
    )
    clf.fit(X_train, y_train)

    # Predict species
    user_sample = np.array([[sepal_length, sepal_width, petal_length, petal_width]])
    prediction = clf.predict(user_sample)[0]
    return f"Predicted species: {class_names[prediction]}"

# 4. Build the Gradio interface

# Inputs to tune hyperparameters
hyperparam_inputs = [
    gr.inputs.Slider(0.01, 1.0, step=0.01, default=0.1, label="learning_rate"),
    gr.inputs.Slider(50, 300, step=50, default=100, label="n_estimators"),
    gr.inputs.Slider(1, 10, step=1, default=3, label="max_depth")
]

# Button or automatic “live” updates
training_interface = gr.Interface(
    fn=train_and_evaluate,
    inputs=hyperparam_inputs,
    outputs="text",
    title="Gradient Boosting Training and Evaluation",
    description="Train a GradientBoostingClassifier on the Iris dataset with different hyperparameters."
)

# Inputs for real-time prediction
feature_inputs = [
    gr.inputs.Number(default=5.1, label=feature_names[0]),
    gr.inputs.Number(default=3.5, label=feature_names[1]),
    gr.inputs.Number(default=1.4, label=feature_names[2]),
    gr.inputs.Number(default=0.2, label=feature_names[3])
] + hyperparam_inputs

prediction_interface = gr.Interface(
    fn=predict_species,
    inputs=feature_inputs,
    outputs="text",
    title="Iris Species Prediction",
    description="Use a GradientBoostingClassifier to predict Iris species from user input."
)

demo = gr.TabbedInterface([training_interface, prediction_interface],
                          ["Train & Evaluate", "Predict"])

# Launch the Gradio app
demo.launch()