app.py

import gradio as gr
import pandas as pd
from transformers import AutoTokenizer, AutoModelForSequenceClassification, pipeline
import matplotlib.pyplot as plt

# Load model and tokenizer
model_name = "ac0hik/Sentiment_Analysis_French"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)

# Create sentiment analysis pipeline
classifier = pipeline('sentiment-analysis', model=model, tokenizer=tokenizer)

# Function to process a single text input
def analyze_text(text):
    result = classifier(text)[0]
    return result

# Function to process a CSV file
def analyze_csv(file):
    df = pd.read_csv(file)
    texts = df['text'].tolist()
    results = classifier(texts)
    
    sentiments = [result['label'] for result in results]
    df['sentiment'] = sentiments
    
    pos_count = sentiments.count('positive')
    neg_count = sentiments.count('negative')
    neu_count = sentiments.count('neutral')
    
    # Create a pie chart
    labels = 'Positive', 'Negative', 'Neutral'
    sizes = [pos_count, neg_count, neu_count]
    colors = ['#ff9999','#66b3ff','#99ff99']
    fig, ax = plt.subplots()
    ax.pie(sizes, labels=labels, colors=colors, autopct='%1.1f%%', startangle=90)
    ax.axis('equal')  # Equal aspect ratio ensures that pie is drawn as a circle.
    
    return df, fig

# Gradio interface
with gr.Blocks() as demo:
    gr.Markdown("## Sentiment Analysis")
    text_input = gr.Textbox(label="Enter Text")
    text_output = gr.JSON(label="Sentiment Analysis Result")
    text_button = gr.Button("Analyze Text")
    
    csv_input = gr.File(label="Upload CSV", file_types=['csv'])
    csv_output = gr.Dataframe(label="Sentiment Analysis Results")
    csv_chart = gr.Plot(label="Sentiment Distribution")
    csv_button = gr.Button("Analyze CSV")
    
    text_button.click(analyze_text, inputs=text_input, outputs=text_output)
    csv_button.click(analyze_csv, inputs=csv_input, outputs=[csv_output, csv_chart])

# Launch the Gradio app
demo.launch()