app.py

import gradio as gr
import tensorflow as tf
import joblib
import pickle
from tensorflow.keras.preprocessing.sequence import pad_sequences

# Load your model and tokenizer
model = tf.keras.models.load_model('lstm_model.h5')
tokenizer = joblib.load('tokenizer.pkl')

with open('padding_config.pkl', 'rb') as file:
    padding_config = pickle.load(file)

# Preprocessing function
def preprocess(text):
    tokenized_text = tokenizer.texts_to_sequences([text])
    padded_text = pad_sequences(tokenized_text, **padding_config)
    return padded_text

# Prediction function
def predict(text):
    processed_text = preprocess(text)
    prediction = model.predict(processed_text)
    return prediction.tolist()

# Gradio interface
iface = gr.Interface(fn=predict, inputs="text", outputs="json")

# Launch Gradio app
iface.launch()