from flask import Flask, jsonify, request
from keras.models import load_model
from keras.preprocessing.text import Tokenizer

app = Flask(__name__)

def get_model():
    global model
    model = load_model('E:/faculta/Anul 3/LICENTA/SentimentModel/model_weights.h5')

    print("Model loaded!")

def preprocess_text(tweet_list):

    tokenizer = Tokenizer(num_words=3000)
    tokenizer.fit_on_texts(tweet_list)

    tweet_list = tokenizer.texts_to_sequences(tweet_list)
    tweet_list = tokenizer.sequences_to_matrix(tweet_list, mode='binary')

    return tweet_list


print("Loading model...")
get_model()

@app.route('/predict', methods=['POST', 'OPTIONS'])
def post_result():
    message = request.get_json(force=True)
    tweets = []

    for tweet in message:
        tweets.append(tweet)

    processed_tweets = preprocess_text(tweets)

    prediction = model.predict(processed_tweets).tolist()

    avg_neg = 0
    avg_pos = 0
    for pr in prediction:
        avg_neg += pr[0]
        avg_pos += pr[1]

    response = {
        'prediction': {
            'negative': avg_neg / len(prediction),
            'positive': avg_pos / len(prediction)
        }
    }

    return jsonify(response)

if __name__ == '_main_':
   app.run()