Spaces:
Sleeping
Sleeping
| import os | |
| import streamlit as st | |
| import google.generativeai as genai | |
| from transformers import pipeline, AutoModelForSequenceClassification, AutoTokenizer | |
| from db import get_dataset_summary, get_entry_by_index | |
| # Configure Gemini API key | |
| GEMINI_API_KEY = os.getenv("gemini_api") | |
| if GEMINI_API_KEY: | |
| genai.configure(api_key=GEMINI_API_KEY) | |
| else: | |
| st.error("⚠️ Google API key is missing! Set it in Hugging Face Secrets.") | |
| # Load pre-trained sentiment analysis model | |
| MODEL_NAME = "cardiffnlp/twitter-roberta-base-sentiment" | |
| try: | |
| tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME) | |
| model = AutoModelForSequenceClassification.from_pretrained(MODEL_NAME) | |
| sentiment_pipeline = pipeline("sentiment-analysis", model=model, tokenizer=tokenizer) | |
| except Exception as e: | |
| st.error(f"❌ Error loading sentiment model: {e}") | |
| # Load Topic Extraction Model | |
| try: | |
| topic_pipeline = pipeline("zero-shot-classification", model="facebook/bart-large-mnli") | |
| except Exception as e: | |
| st.error(f"❌ Error loading topic extraction model: {e}") | |
| # Predefined topic labels for classification | |
| TOPIC_LABELS = [ | |
| "Technology", "Politics", "Business", "Sports", "Entertainment", | |
| "Health", "Science", "Education", "Finance", "Travel", "Food" | |
| ] | |
| def analyze_sentiment(text): | |
| try: | |
| sentiment_result = sentiment_pipeline(text)[0] | |
| label = sentiment_result['label'] | |
| score = sentiment_result['score'] | |
| sentiment_mapping = { | |
| "LABEL_0": "Negative", | |
| "LABEL_1": "Neutral", | |
| "LABEL_2": "Positive" | |
| } | |
| return sentiment_mapping.get(label, "Unknown"), score | |
| except Exception as e: | |
| return f"Error analyzing sentiment: {e}", None | |
| def extract_topic(text): | |
| try: | |
| topic_result = topic_pipeline(text, TOPIC_LABELS) | |
| top_topic = topic_result["labels"][0] | |
| confidence = topic_result["scores"][0] | |
| return top_topic, confidence | |
| except Exception as e: | |
| return f"Error extracting topic: {e}", None | |
| def is_dataset_query(prompt): | |
| keywords = ["dataset", "data", "csv", "mongodb", "historical"] | |
| return any(keyword in prompt.lower() for keyword in keywords) | |
| def extract_entry_index(prompt): | |
| # Map ordinal words to indices (0-indexed) | |
| ordinals = { | |
| "first": 0, | |
| "1st": 0, | |
| "second": 1, | |
| "2nd": 1, | |
| "third": 2, | |
| "3rd": 2, | |
| "fourth": 3, | |
| "4th": 3, | |
| "fifth": 4, | |
| "5th": 4, | |
| } | |
| for word, index in ordinals.items(): | |
| if word in prompt.lower(): | |
| return index | |
| return None | |
| def chatbot_response(user_prompt): | |
| if not user_prompt: | |
| return None, None, None, None, None | |
| # Check if the query is about a specific dataset entry. | |
| entry_index = extract_entry_index(user_prompt) | |
| if entry_index is not None: | |
| entry_text = get_entry_by_index(entry_index) | |
| if entry_text: | |
| # Create a combined prompt for Gemini to generate detailed insights. | |
| combined_prompt = ( | |
| f"Analyze the following dataset entry from MongoDB:\n\n{entry_text}\n\n" | |
| "Provide detailed insights, including sentiment analysis and category extraction." | |
| ) | |
| model_gen = genai.GenerativeModel("gemini-1.5-pro") | |
| ai_response = model_gen.generate_content(combined_prompt) | |
| # Analyze the entry text. | |
| sentiment_label, sentiment_confidence = analyze_sentiment(entry_text) | |
| topic_label, topic_confidence = extract_topic(entry_text) | |
| return ai_response.text, sentiment_label, sentiment_confidence, topic_label, topic_confidence | |
| else: | |
| return f"❌ No entry found for index {entry_index+1}.", None, None, None, None | |
| # Otherwise, if the query is about the dataset in general. | |
| if is_dataset_query(user_prompt): | |
| dataset_insights = get_dataset_summary() | |
| combined_prompt = ( | |
| f"{user_prompt}\n\nDataset Insights:\n{dataset_insights}\n\n" | |
| "Provide a detailed answer that incorporates these dataset insights." | |
| ) | |
| else: | |
| combined_prompt = user_prompt | |
| model_gen = genai.GenerativeModel("gemini-1.5-pro") | |
| ai_response = model_gen.generate_content(combined_prompt) | |
| # Run sentiment analysis and topic extraction on the original user prompt. | |
| sentiment_label, sentiment_confidence = analyze_sentiment(user_prompt) | |
| topic_label, topic_confidence = extract_topic(user_prompt) | |
| return ai_response.text, sentiment_label, sentiment_confidence, topic_label, topic_confidence | |