Update app.py

app.py

@@ -5,7 +5,7 @@ import PyPDF2 as pdf
 import io
 import re
 import streamlit as st
-from transformers import pipeline, AutoModelForSequenceClassification, AutoTokenizer
+from transformers import pipeline, AutoTokenizer, AutoModelForSequenceClassification
 from sklearn.feature_extraction.text import TfidfVectorizer
 from sklearn.metrics.pairwise import cosine_similarity
 
@@ -15,18 +15,20 @@ if not api_key:
     st.error("API key not found. Please set GOOGLE_API_KEY in your environment variables.")
     st.stop()
 
-# Configure the generative AI client
+# Initialize the generative AI model
 genai.configure(api_key=api_key)
+gemini_model = genai.generate  # Use the correct method for text generation
 
-# Initialize pre-trained Hugging Face pipelines
-skill_extractor = pipeline("ner", model="dslim/bert-base-NER", aggregation_strategy="simple")
-education_extractor = pipeline("ner", model="dbmdz/bert-large-cased-finetuned-conll03-english", aggregation_strategy="simple")
+# Load updated pre-trained models
+skill_extractor = pipeline("ner", model="Jean-Baptiste/roberta-large-ner-english", aggregation_strategy="simple")
+education_extractor = pipeline("ner", model="Jean-Baptiste/roberta-large-ner-english", aggregation_strategy="simple")
 
-# Define the Hugging Face sentiment analysis model
+# Sentiment analysis model
 task = "sentiment-analysis"
-model_name = "roberta-base"
+model_name = "cardiffnlp/twitter-roberta-base-sentiment"
 tokenizer = AutoTokenizer.from_pretrained(model_name)
 model = AutoModelForSequenceClassification.from_pretrained(model_name)
+sentiment_analysis = pipeline(task, model=model, tokenizer=tokenizer)
 
 # Extract text from uploaded PDF file
 def input_pdf_text(uploaded_file):
@@ -58,21 +60,21 @@ def extract_contact_info(resume_text):
 def extract_skills(resume_text):
     try:
         ner_results = skill_extractor(resume_text)
-        skills = [entity['word'] for entity in ner_results if 'skill' in entity['entity_group'].lower()]
+        skills = [entity['word'] for entity in ner_results if entity['entity_group'].lower() == 'skill']
         return ", ".join(skills) if skills else "Not Available"
     except Exception:
         return "Not Available"
 
-# Extract education information using NER model
+# Extract education using NER model
 def extract_education(resume_text):
     try:
         ner_results = education_extractor(resume_text)
-        education_entities = [entity['word'] for entity in ner_results if 'education' in entity['entity_group'].lower()]
+        education_entities = [entity['word'] for entity in ner_results if entity['entity_group'].lower() == 'education']
         return ", ".join(education_entities) if education_entities else "Not Available"
     except Exception:
         return "Not Available"
 
-# Extract years of experience using regex
+# Extract experience years
 def extract_experience_years(text):
     patterns = [
         r"(\d{4})\s?[–\-to]+\s?(\d{4})",
@@ -84,10 +86,10 @@ def extract_experience_years(text):
     for pattern in patterns:
         matches = re.findall(pattern, text)
         for match in matches:
-            if len(match) >= 2 and match[0].isdigit() and match[1].isdigit():
-                start_year, end_year = int(match[0]), int(match[1])
+            if len(match) == 2 and match[1].isdigit():
+                start_year, end_year = map(int, match[:2])
                 years += max(0, end_year - start_year)
-            elif "years" in match or "yrs" in match:
+            elif len(match) == 1 and match[0].isdigit():
                 years += int(match[0])
     return years
 
@@ -117,8 +119,8 @@ def get_gemini_response(input_text, job_description):
         - Leadership experience (years)
         - Match percentage
         """
-        response = genai.generate_text(prompt=prompt)
-        return response.candidates[0]['content']  # Parse response based on the API's response structure
+        response = gemini_model(prompt=prompt, max_output_tokens=100)
+        return response['text'].strip()
     except Exception as e:
         st.error(f"Error generating response from Gemini model: {str(e)}")
         return ""