app.py

import gradio as gr
import os
import torch
import numpy as np
import random
from huggingface_hub import login, HfFolder
from transformers import AutoTokenizer, AutoModelForSequenceClassification, AutoModelForCausalLM, TextIteratorStreamer
from scipy.special import softmax
import logging
import spaces
from threading import Thread
from collections.abc import Iterator
import csv

# Increase CSV field size limit
csv.field_size_limit(1000000)

# Setup logging
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(message)s')

# Set a seed for reproducibility
seed = 42
np.random.seed(seed)
random.seed(seed)
torch.manual_seed(seed)
if torch.cuda.is_available():
    torch.cuda.manual_seed_all(seed)

# Login to Hugging Face
token = os.getenv("hf_token")
HfFolder.save_token(token)
login(token)

model_paths = [
    'karths/binary_classification_train_port',
    'karths/binary_classification_train_perf',
    "karths/binary_classification_train_main",
    "karths/binary_classification_train_secu",
    "karths/binary_classification_train_reli",
    "karths/binary_classification_train_usab",
    "karths/binary_classification_train_comp"
]

quality_mapping = {
    'binary_classification_train_port': 'Portability',  
    'binary_classification_train_main': 'Maintainability',
    'binary_classification_train_secu': 'Security',
    'binary_classification_train_reli': 'Reliability',
    'binary_classification_train_usab': 'Usability',
    'binary_classification_train_perf': 'Performance',
    'binary_classification_train_comp': 'Compatibility'
}

# Pre-load models and tokenizer for quality prediction
tokenizer = AutoTokenizer.from_pretrained("distilroberta-base")
models = {path: AutoModelForSequenceClassification.from_pretrained(path) for path in model_paths}  # Load to CPU initially

def get_quality_name(model_name):
    return quality_mapping.get(model_name.split('/')[-1], "Unknown Quality")


def model_prediction(model, text, device):
    model.to(device) # Move the *specific* model to the GPU
    model.eval()
    inputs = tokenizer(text, return_tensors="pt", padding=True, truncation=True, max_length=512)
    inputs = {k: v.to(device) for k, v in inputs.items()}
    with torch.no_grad():
        outputs = model(**inputs)
        logits = outputs.logits
        probs = softmax(logits.cpu().numpy(), axis=1)
    avg_prob = np.mean(probs[:, 1])
    model.to("cpu")  # Move the model *back* to the CPU
    return avg_prob

# --- Llama 3.2 3B Model Setup ---
LLAMA_MAX_MAX_NEW_TOKENS = 512  # Max tokens for Explanation
LLAMA_DEFAULT_MAX_NEW_TOKENS = 512  # Max tokens for explantion
LLAMA_MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "2048")) # Reduced
llama_device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")  # Explicit device
llama_model_id = "meta-llama/Llama-3.2-3B-Instruct"
llama_tokenizer = AutoTokenizer.from_pretrained(llama_model_id)
llama_model = AutoModelForCausalLM.from_pretrained(
    llama_model_id,
    device_map="auto",  # Let Transformers handle optimal device placement
    torch_dtype=torch.bfloat16,
)
llama_model.eval()

if llama_tokenizer.pad_token is None:
    llama_tokenizer.pad_token = llama_tokenizer.eos_token

def llama_generate(
    message: str,
    max_new_tokens: int = LLAMA_DEFAULT_MAX_NEW_TOKENS,
    temperature: float = 0.6,
    top_p: float = 0.9,
    top_k: int = 50,
    repetition_penalty: float = 1.2,
) -> str: # Return string, not iterator

    inputs = llama_tokenizer(message, return_tensors="pt", padding=True, truncation=True, max_length=LLAMA_MAX_INPUT_TOKEN_LENGTH).to(llama_model.device)

    if inputs.input_ids.shape[1] > LLAMA_MAX_INPUT_TOKEN_LENGTH:
        inputs.input_ids = inputs.input_ids[:, -LLAMA_MAX_INPUT_TOKEN_LENGTH:]
        gr.Warning(f"Trimmed input from conversation as it was longer than {LLAMA_MAX_INPUT_TOKEN_LENGTH} tokens.")

    # Generate *without* streaming
    with torch.no_grad():  # Ensure no gradient calculation
        generate_ids = llama_model.generate(
            **inputs,
            max_new_tokens=max_new_tokens,
            do_sample=True,
            top_p=top_p,
            top_k=top_k,
            temperature=temperature,
            num_beams=1,
            repetition_penalty=repetition_penalty,
            pad_token_id=llama_tokenizer.pad_token_id,  # Pass pad_token_id here
            eos_token_id=llama_tokenizer.eos_token_id,  # Pass eos_token_id here

        )
    output_text = llama_tokenizer.decode(generate_ids[0], skip_special_tokens=True)
    torch.cuda.empty_cache()  # Clear cache after each generation
    return output_text


def generate_explanation(issue_text, top_qualities):
    if not top_qualities:
        return "<div style='color: red;'>No explanation available as no quality tags were predicted.</div>"

    prompt = f"""
Given the following issue description:
---
{issue_text}
---
Explain why this issue might be classified under the following quality categories. Provide a concise explanation for each category, relating it back to the issue description:
"""
    for quality, _ in top_qualities:
        prompt += f"- {quality}\n"


    try:
        explanation = llama_generate(prompt)
        # Format the explanation for better readability
        formatted_explanation = ""
        for quality, _ in top_qualities:
            formatted_explanation += f"<p><b>{quality}:</b></p>"  # Bold the quality name
            # Find the explanation for this specific quality.  This is a simple
            # approach that works if Llama follows the prompt structure.
            # A more robust approach might use regex or sentence embeddings.
            start = explanation.find(quality)
            if start != -1:
                start += len(quality) + 2 # Move past "Quality:"
                end = explanation.find("\n", start)  # Find next newline
                if end == -1:
                    end = len(explanation)
                formatted_explanation += f"<p>{explanation[start:end].strip()}</p>" # Add the explanation text
            else:
                formatted_explanation += f"<p>Explanation for {quality} not found.</p>"

        return f"<div style='overflow-y: scroll; max-height: 400px;'>{formatted_explanation}</div>" #Added scroll
    except Exception as e:
        logging.error(f"Error during Llama generation: {e}")
        return "<div style='color: red;'>An error occurred while generating the explanation.</div>"


 # @spaces.GPU(duration=60)  # Apply the GPU decorator *only* to the main interface
def main_interface(text):
    if not text.strip():
        return "<div style='color: red;'>No text provided. Please enter a valid issue description.</div>", "", ""

    if len(text) < 30:
        return "<div style='color: red;'>Text is less than 30 characters.</div>", "", ""

    device = "cuda" if torch.cuda.is_available() else "cpu" # Keep this for model_prediction
    results = []
    for model_path, model in models.items():
        quality_name = get_quality_name(model_path)
        avg_prob = model_prediction(model, text, device)  # Pass the device
        if avg_prob >= 0.95:
            results.append((quality_name, avg_prob))
        logging.info(f"Model: {model_path}, Quality: {quality_name}, Average Probability: {avg_prob:.3f}")


    if not results:
        return "<div style='color: red;'>No recommendation. Prediction probability is below the threshold. </div>", "", ""

    top_qualities = sorted(results, key=lambda x: x[1], reverse=True)[:3]
    output_html = render_html_output(top_qualities)
    explanation = generate_explanation(text, top_qualities)

    return output_html, "", explanation

def render_html_output(top_qualities):
    styles = """
    <style>
        .quality-container {
            font-family: Arial, sans-serif;
            text-align: center;
            margin-top: 20px;
        }
        .quality-label, .ranking {
            display: inline-block;
            padding: 0.5em 1em;
            font-size: 18px;
            font-weight: bold;
            color: white;
            background-color: #007bff;
            border-radius: 0.5rem;
            margin-right: 10px;
            box-shadow: 0 2px 4px rgba(0, 0, 0, 0.2);
        }
        .probability {
            display: block;
            margin-top: 10px;
            font-size: 16px;
            color: #007bff;
        }
    </style>
    """
    html_content = ""
    ranking_labels = ['Top 1 Prediction', 'Top 2 Prediction', 'Top 3 Prediction']
    top_n = min(len(top_qualities), len(ranking_labels))
    for i in range(top_n):
        quality, prob = top_qualities[i]
        html_content += f"""
        <div class="quality-container">
            <span class="ranking">{ranking_labels[i]}</span>
            <span class="quality-label">{quality}</span>
        </div>
        """
    return styles + html_content

example_texts = [
    ["The algorithm does not accurately distinguish between the positive and negative classes during edge cases.\n\nEnvironment: Production\nReproduction: Run the classifier on the test dataset with known edge cases."],
    ["The regression tests do not cover scenarios involving concurrent user sessions.\n\nEnvironment: Test automation suite\nReproduction: Update the test scripts to include tests for concurrent sessions."],
    ["There is frequent miscommunication between the development and QA teams regarding feature specifications.\n\nEnvironment: Inter-team meetings\nReproduction: Audit recent communication logs and meeting notes between the teams."],
    ["The service-oriented architecture does not effectively isolate failures, leading to cascading failures across services.\n\nEnvironment: Microservices architecture\nReproduction: Simulate a service failure and observe the impact on other services."]
]

interface = gr.Interface(
    fn=main_interface,
    inputs=gr.Textbox(lines=7, label="Issue Description", placeholder="Enter your issue text here"),
    outputs=[
        gr.HTML(label="Prediction Output"),
        gr.Textbox(label="Predictions", visible=False),
        gr.HTML(label="Explanation") # Change to gr.HTML
    ],
    title="QualityTagger",
    description="This tool classifies text into different quality domains such as Security, Usability, etc., and provides explanations.",
    examples=example_texts
)
interface.launch(share=True)