modif of predict_model

src/models/predict_model.py

@@ -1,14 +1,35 @@
+
+import click
 import joblib
-from imblearn.metrics import classification_report_imbalanced
+import pandas as pd
+
+# Load your saved model
+loaded_model = joblib.load("src/models/trained_model.joblib")
+
+@click.command()
+def main():
+    # Load your training dataset (replace 'train_data.csv' with your file)
+    X_train = pd.read_csv("data/preprocessed/X_train.csv")
+
+    # Get feature names from X_train columns
+    feature_names = X_train.columns.tolist()
+
+    features = {}
 
-# Load the trained model from the file
-model_filename = 'trained_model.joblib'
-loaded_model = joblib.load(model_filename)
+    # Get user input for each feature
+    for feature_name in feature_names:
+        feature_value = click.prompt(f"Enter value for {feature_name}", type=float)
+        features[feature_name] = feature_value
 
-y_pred = loaded_model.predict(X_test)
+    # Predict using the model
+    result = predict_model(features)
+    print("Prediction:", result)
 
+def predict_model(features):
+    input_df = pd.DataFrame([features])
+    prediction = loaded_model.predict(input_df)
+    return prediction
 
+if __name__ == "__main__":
+    main()
 
-print(f1_score(y_test, y_pred_rf))
-pd.crosstab(y_test, y_pred, rownames=['Classe réelle'], colnames=['Classe prédite'])
-print(classification_report_imbalanced(y_test, y_pred))