app.py

import gradio as gr
import requests
import os
import pandas as pd

# Hugging Face Inference API endpoint for DeepSeek
API_URL = "https://router.huggingface.co/models/deepseek-ai/DeepSeek-V2"
headers = {"Authorization": f"Bearer {os.environ['HF_API_TOKEN']}"}

def preprocess_csv(file_path):
    df = pd.read_csv(file_path)
    events = []
    for _, row in df.iterrows():
        events.append(f"On {row['Date']} at {row['Time']}, state was {row['State']} → {row['Message Text']}")
    return "\n".join(events)

def analyze_log(file_path, mode):
    text = preprocess_csv(file_path)

    # Build prompt depending on mode
    if mode == "Summarize":
        prompt = "Summarize the following AHU alarm log:\n\n" + text
    elif mode == "Highlight anomalies":
        prompt = "Identify unusual or repeated alarms in this AHU log and explain possible causes:\n\n" + text
    elif mode == "Suggest maintenance":
        prompt = "Based on this AHU alarm log, suggest maintenance actions:\n\n" + text
    else:
        prompt = text

    # Call Hugging Face Inference API
    response = requests.post(API_URL, headers=headers, json={"inputs": prompt})

    if response.status_code != 200:
        return f"Error {response.status_code}: {response.text}"

    try:
        result = response.json()
    except Exception:
        return f"Failed to decode JSON. Raw response: {response.text}"

    # Extract generated text
    if isinstance(result, list) and "generated_text" in result[0]:
        return result[0]["generated_text"]
    elif isinstance(result, dict) and "generated_text" in result:
        return result["generated_text"]
    else:
        return str(result)

iface = gr.Interface(
    fn=analyze_log,
    inputs=[
        gr.File(type="filepath", label="Upload Log File"),
        gr.Dropdown(choices=["Summarize", "Highlight anomalies", "Suggest maintenance"], label="Analysis Mode")
    ],
    outputs="text",
    title="AHU Log Analyzer (DeepSeek API)",
    description="Upload your log file (CSV) and choose how you want it analyzed using DeepSeek via Hugging Face API."
)

if __name__ == "__main__":
    iface.launch()