app.py

from flask import Flask, request, jsonify
from transformers import pipeline, AutoTokenizer, AutoModelForSeq2SeqLM
import os, time

app = Flask(__name__)

# Load from local 'model/' folder
tokenizer = AutoTokenizer.from_pretrained("./model")
model = AutoModelForSeq2SeqLM.from_pretrained("./model")
summarizer = pipeline("summarization", model=model, tokenizer=tokenizer)

# Warm up
summarizer("Warm-up input.", max_length=30, min_length=5, do_sample=False)

@app.route("/summarize", methods=["POST"])
def summarize():
    try:
        data = request.get_json()
        text = data["text"]
        max_len = int(data.get("max_length", 150))  # Default to 150 if not provided
        min_len = int(data.get("min_length", 30))   # Default to 30 if not provided

        summary_output = summarizer(
            text,
            max_length=max_len,
            min_length=min_len,
            do_sample=False
        )

        summary = summary_output[0]["summary_text"]

        # Token count of the summary
        tokenized = tokenizer(summary, return_tensors="pt")
        token_count = len(tokenized["input_ids"][0])

        return jsonify({
            "summary": summary,
            "token_count": token_count
        })

    except Exception as e:
        return jsonify({"error": str(e)}), 400
        
if __name__ == "__main__":
    app.run(host="0.0.0.0", port=7860)