app.py

import gradio as gr
import pandas as pd
import re
from datasets import load_dataset
from sentence_transformers import SentenceTransformer
from sklearn.metrics.pairwise import cosine_similarity
from transformers import pipeline, set_seed
import numpy as np

# -------------------------------
# 1. Load and clean dataset
# -------------------------------
@st.cache_resource
def load_and_prepare_data():
    dataset = load_dataset("sentiment140")
    df = dataset["train"].to_pandas()
    df.dropna(subset=["text", "sentiment"], inplace=True)
    df["text_length"] = df["text"].apply(len)
    df = df[(df["text_length"] >= 5) & (df["text_length"] <= 280)]
    df["clean_text"] = df["text"].apply(clean_text)
    return df

def clean_text(text):
    text = text.lower()
    text = re.sub(r"http\S+", "", text)
    text = re.sub(r"@\w+", "", text)
    text = re.sub(r"#\w+", "", text)
    text = re.sub(r"[^\w\s]", "", text)
    text = re.sub(r"\s+", " ", text).strip()
    return text

# Load data once
df = load_and_prepare_data()
sample_df = df.sample(5000, random_state=42).reset_index(drop=True)
texts = sample_df["clean_text"].tolist()

# -------------------------------
# 2. Load models
# -------------------------------
embedding_model = SentenceTransformer("sentence-transformers/all-MiniLM-L6-v2")
generator = pipeline("text-generation", model="distilgpt2")
set_seed(42)

# -------------------------------
# 3. Helper functions
# -------------------------------
def get_top3_similarities(text_input):
    text_embeddings = embedding_model.encode(texts, show_progress_bar=False)
    input_embedding = embedding_model.encode([text_input])
    similarities = cosine_similarity(input_embedding, text_embeddings)[0]
    top_indices = similarities.argsort()[-3:][::-1]
    return [texts[i] for i in top_indices]

def generate_best_tweet(text_input):
    synthetic_outputs = generator(
        text_input,
        max_length=50,
        num_return_sequences=10,
        do_samp_