app.py

import tensorflow as tf
from tensorflow.keras import layers, models
from tensorflow.keras.preprocessing.text import Tokenizer
from tensorflow.keras.preprocessing.sequence import pad_sequences
import gradio as gr

# Load the IMDb dataset
imdb = tf.keras.datasets.imdb
vocab_size = 10000
maxlen = 100
(X_train, y_train), (X_test, y_test) = imdb.load_data(num_words=vocab_size)
X_train = pad_sequences(X_train, maxlen=maxlen)
X_test = pad_sequences(X_test, maxlen=maxlen)

# Define the model
model = models.Sequential([
    layers.Embedding(vocab_size, 16, input_length=maxlen),
    layers.GlobalAveragePooling1D(),
    layers.Dense(16, activation='relu'),
    layers.Dense(1, activation='sigmoid')
])

model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])
model.fit(X_train, y_train, epochs=10, batch_size=512, validation_data=(X_test, y_test), verbose=1)

# Save the model
model.save("sentiment_analysis_model.h5")

# Function to predict sentiment
def predict_sentiment(text):
    tokenizer = Tokenizer(num_words=vocab_size)
    tokenizer.fit_on_texts([text])
    sequence = tokenizer.texts_to_sequences([text])
    padded_sequence = pad_sequences(sequence, maxlen=maxlen)
    
    prediction = model.predict(padded_sequence)[0][0]
    sentiment = "Positive" if prediction >= 0.5 else "Negative"
    confidence = round(prediction, 4)
    
    return sentiment, confidence

# Gradio Interface
def gradio_predict(text):
    sentiment, confidence = predict_sentiment(text)
    return f"Sentiment: {sentiment}, Confidence: {confidence:.4f}"

# Create Gradio Interface
interface = gr.Interface(fn=gradio_predict, 
                         inputs=gr.Textbox(lines=2, placeholder="Enter your text here..."), 
                         outputs="text",
                         title="Sentiment Analysis",
                         description="Enter a movie review or any text to analyze its sentiment.")

# Launch the Gradio Interface
interface.launch()