app.py

import tensorflow as tf
import gradio as gr
import numpy as np

# Load the saved model
model = load_model('sentimentality.h5')

# Define the function to preprocess the text
def preprocess(text):
    # Tokenize the text
    tokenizer = tf.keras.preprocessing.text.Tokenizer(num_words=10000)
    tokenizer.fit_on_texts(text)
    sequences = tokenizer.texts_to_sequences(text)

    # Pad the sequences to a fixed length of 30
    padded_sequences = tf.keras.preprocessing.sequence.pad_sequences(sequences, maxlen=30, padding='post')

    return np.array(padded_sequences)

# Define the function to make predictions on the text
def sentiment_analysis(text):
    preprocessed_text = preprocess([text])
    prediction = model.predict(preprocessed_text)[0][0]
    
    if prediction >= 0.5:
        return "positive"
    else:
        return "negative"

# Define the Gradio interface
interface = gr.Interface(
    fn=get_sentiment_scores,
    inputs=gr.inputs.Textbox(placeholder="Enter a positive or negative sentence here..."),
    outputs=gr.outputs.Textbox(label="Sentiment Label"),
    examples=[["This is wonderful!"], ["I hate this product."]]
)

# Launch the interface
interface.launch()