app.py

from flask import Flask, request, jsonify
import joblib
import nltk
from nltk.corpus import stopwords
from nltk.tokenize import word_tokenize
from nltk.stem import WordNetLemmatizer
import numpy as np
import traceback

# Download necessary NLTK data files
nltk.download('punkt')
nltk.download('stopwords')
nltk.download('wordnet')

app = Flask(__name__)

# Load models and vectorizer
try:
    model_impactf = joblib.load('model_impactf.pkl')
    model_probf = joblib.load('model_probf.pkl')
    vectorizer = joblib.load('vectorizer.pkl')
except Exception as e:
    print(f"Error loading models or vectorizer: {str(e)}")
    raise

# Define stop words and lemmatizer
stop_words = set(stopwords.words('french'))
lemmatizer = WordNetLemmatizer()

def preprocess_text(text):
    try:
        # Convert the text to string (in case it's not)
        text = str(text)
        # Tokenize the text
        tokens = word_tokenize(text)
        # Convert to lower case
        tokens = [word.lower() for word in tokens]
        # Remove punctuation and stop words
        tokens = [word for word in tokens if word.isalnum() and word not in stop_words]
        # Lemmatize the words
        tokens = [lemmatizer.lemmatize(word) for word in tokens]
        # Join the tokens back into a string
        return ' '.join(tokens)
    except Exception as e:
        print(f"Error in text preprocessing: {str(e)}")
        raise

@app.route('/predict', methods=['POST'])
def predict():
    try:
        data = request.get_json(force=True)
        print("Received data:", data)  # Debugging

        if 'Risques' not in data:
            return jsonify({'error': 'Key "Risques" is missing in the input data'}), 400

        risques = data['Risques']

        # Preprocess the text
        preprocessed_text = preprocess_text(risques)
        print("Preprocessed text:", preprocessed_text)  # Debugging

        # Vectorize the input text
        input_features = vectorizer.transform([preprocessed_text])

        # Predict impact and probability
        predicted_impact = model_impactf.predict(input_features)[0]
        predicted_probability = model_probf.predict(input_features)[0]

        # Convert predictions to standard Python integers
        predicted_impact = int(predicted_impact.item())
        predicted_probability = int(predicted_probability.item())

        response = {
            'Predicted Impact': predicted_impact,
            'Predicted Probability': predicted_probability
        }

        print("Response:", response)  # Debugging
        return jsonify(response)

    except Exception as e:
        print("Error occurred:", traceback.format_exc())
        response = {
            'error': str(e)
        }
        return jsonify(response), 500

if __name__ == '__main__':
    app.run(debug=True)