import pandas as pd
import numpy as np
import string
import flask
from tensorflow.keras.preprocessing.text import Tokenizer
from tensorflow.keras.preprocessing.sequence import pad_sequences
from flask import Flask, request, jsonify, render_template
import nltk
import json
from tensorflow.keras.models import load_model
import tensorflow as tf

model = load_model("_best_model_LSTM.hdf5")

app = Flask(__name__)


from tensorflow.keras.preprocessing.text import tokenizer_from_json

import json

with open("tokenizer.json", 'r') as f:
    tokenizer_data = f.read()


tokenizer_config = json.loads(tokenizer_data)
tokenizer = tokenizer_from_json(tokenizer_config)

print(tokenizer)
@app.route('/')
def home():
    return render_template('index.html')

@app.route('/predict',methods=['POST'])
def predict():
    new_review = [str(x) for x in request.form.values()]
    print(new_review)
    # data = pd.DataFrame(new_review)
    # data.columns = ['new_review']
    sequences = tokenizer.texts_to_sequences(new_review)
    print(sequences)
    padded_sequences = pad_sequences(sequences, maxlen=200,padding='post', value=0)

    predictions = model.predict(padded_sequences)
    print(predictions)
    if predictions<.5:
        return render_template('index.html', prediction_text='Negative')
    else:
        return render_template('index.html', prediction_text='Positive')


if __name__ == "__main__":
    app.run(host='0.0.0.0',port=8080)