Update app.py

app.py

@@ -1,14 +1,14 @@
+import os
+import google.generativeai as genai
 import spacy
 import nltk
 import gradio as gr
 import pandas as pd
 import re
 from fuzzywuzzy import fuzz
-from transformers import pipeline
 from nltk.corpus import stopwords
 from nltk.tokenize import word_tokenize
 from nltk.stem import WordNetLemmatizer
-import gemini
 import fitz  # PyMuPDF
 
 # Initialize NLTK resources
@@ -26,8 +26,15 @@ except:
     subprocess.run(["python3", "-m", "spacy", "download", "en_core_web_sm"])
     nlp = spacy.load("en_core_web_sm")
 
-# Initialize Gemini model for text analysis
-gemini_model = pipeline('text-classification', model='google/gemini-1.5-flash')
+# Fetch the Google API key from Hugging Face secrets
+google_api_key = os.getenv("GOOGLE_API_KEY")
+
+# Check if the key is being fetched correctly
+if google_api_key is None:
+    raise ValueError("Google API Key is missing from environment variables.")
+
+# Configure the Gemini API with the secret key
+genai.configure(api_key=google_api_key)
 
 # Extract text from PDF using PyMuPDF
 def extract_text_from_pdf(pdf_path):
@@ -72,11 +79,16 @@ def calculate_match_percentage(resume_text, job_desc_text):
     match = fuzz.partial_ratio(resume_text, job_desc_text)
     return match
 
-# Use Gemini 1.5 to analyze text and extract job-related insights
+# Use Gemini 1.5 Flash to analyze text and extract job-related insights
 def gemini_analysis(text):
-    """Use Gemini 1.5 to analyze text and extract insights like roles and skills."""
-    analysis = gemini_model(text)
-    return analysis[0]['label']
+    """Use Gemini 1.5 Flash model to analyze text and extract insights like roles and skills."""
+    response = genai.generate_text(
+        model="gemini-1.5-flash",  # Use the Flash model here
+        temperature=0.7,
+        max_output_tokens=512,
+        input_text=text
+    )
+    return response['text']
 
 # Process resumes and calculate match with job description
 def process_uploaded_resumes(resume_files: list, job_desc: str):
@@ -98,7 +110,7 @@ def process_uploaded_resumes(resume_files: list, job_desc: str):
         # Compare named entities between resume and job description
         entity_match = len(resume_entities.intersection(job_desc_entities)) / len(job_desc_entities) * 100
         
-        # Use Gemini model to analyze job-related insights (optional)
+        # Use Gemini 1.5 Flash model to analyze job-related insights (optional)
         gemini_match = gemini_analysis(resume_text)
         
         # Calculate match percentage based on fuzzy matching