ss

app.py

@@ -19,7 +19,6 @@ from transformers import AutoModelForCausalLM, AutoTokenizer
 import time
 import faiss
 import re
-import openai
 
 # Download NLTK resources
 try:
@@ -47,7 +46,7 @@ with st.sidebar:
     
     # Advanced options
     st.subheader("Advanced Options")
-    top_k = st.selectbox("Number of results to display", options=[1, 5, 10, 20, 50], index=2)
+    top_k = st.selectbox("Number of results to display", [1,2,3,4,5], index=4)
     
     # LLM Settings
     st.subheader("LLM Settings")
@@ -66,7 +65,7 @@ with st.sidebar:
     st.markdown("### 📊 Models Used")
     st.markdown("- **Embedding**: BAAI/bge-large-en-v1.5")
     st.markdown("- **Cross-Encoder**: ms-marco-MiniLM-L6-v2")
-    st.markdown("- **LLM**: Qwen/Qwen3-14B")
+    st.markdown("- **LLM**: Qwen/Qwen3-1.7B")
     st.markdown("### 📈 Scoring Formula")
     st.markdown("**Final Score = Cross-Encoder (0-1) + BM25 (0.1-0.2) + Intent (0-0.3)**")
 
@@ -81,22 +80,28 @@ if 'resume_texts' not in st.session_state:
     st.session_state.resume_texts = []
 if 'file_names' not in st.session_state:
     st.session_state.file_names = []
-if 'explanations_generated' not in st.session_state:
-    st.session_state.explanations_generated = False
 if 'current_job_description' not in st.session_state:
     st.session_state.current_job_description = ""
-if 'vllm_4b_endpoint' not in st.session_state:
-    st.session_state.vllm_4b_endpoint = "http://localhost:8001/v1"  # Qwen3-4B vLLM endpoint
-if 'vllm_14b_endpoint' not in st.session_state:
-    st.session_state.vllm_14b_endpoint = "http://localhost:8002/v1"  # Qwen3-14B vLLM endpoint
+if 'qwen3_1_7b_tokenizer' not in st.session_state:
+    print("[Init] Loading Qwen3-1.7B Tokenizer...")
+    st.session_state.qwen3_1_7b_tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-1.7B")
+    print("[Init] Qwen3-1.7B Tokenizer Loaded.")
+if 'qwen3_1_7b_model' not in st.session_state:
+    print("[Init] Loading Qwen3-1.7B Model...")
+    st.session_state.qwen3_1_7b_model = AutoModelForCausalLM.from_pretrained(
+        "Qwen/Qwen3-1.7B", torch_dtype="auto", device_map="auto"
+    )
+    print("[Init] Qwen3-1.7B Model Loaded.")
 
 @st.cache_resource
 def load_embedding_model():
     """Load and cache the BGE embedding model"""
+    print("[Cache] Attempting to load Embedding Model (BAAI/bge-large-en-v1.5)...")
     try:
         with st.spinner("🔄 Loading BAAI/bge-large-en-v1.5 model..."):
             model = SentenceTransformer('BAAI/bge-large-en-v1.5')
             st.success("✅ Embedding model loaded successfully!")
+            print("[Cache] Embedding Model (BAAI/bge-large-en-v1.5) LOADED.")
             return model
     except Exception as e:
         st.error(f"❌ Error loading embedding model: {str(e)}")
@@ -105,33 +110,44 @@ def load_embedding_model():
 @st.cache_resource
 def load_cross_encoder():
     """Load and cache the Cross-Encoder model"""
+    print("[Cache] Attempting to load Cross-Encoder Model (ms-marco-MiniLM-L6-v2)...")
     try:
         with st.spinner("🔄 Loading Cross-Encoder ms-marco-MiniLM-L6-v2..."):
             from sentence_transformers import CrossEncoder
             model = CrossEncoder('cross-encoder/ms-marco-MiniLM-L6-v2')
             st.success("✅ Cross-Encoder model loaded successfully!")
+            print("[Cache] Cross-Encoder Model (ms-marco-MiniLM-L6-v2) LOADED.")
             return model
     except Exception as e:
         st.error(f"❌ Error loading Cross-Encoder model: {str(e)}")
         return None
 
-def vllm_chat_completion(prompt, endpoint, max_tokens=200, temperature=0.7):
-    openai.api_base = endpoint
-    openai.api_key = "EMPTY"  # vLLM does not require a real key
-    response = openai.ChatCompletion.create(
-        model="Qwen/Qwen3-4B" if "4b" in endpoint else "Qwen/Qwen3-14B",
-        messages=[{"role": "user", "content": prompt}],
-        max_tokens=max_tokens,
-        temperature=temperature,
-        stream=False
+def generate_qwen3_response(prompt, tokenizer, model, max_new_tokens=200):
+    messages = [{"role": "user", "content": prompt}]
+    text = tokenizer.apply_chat_template(
+        messages,
+        tokenize=False,
+        add_generation_prompt=True,
+        enable_thinking=True
+    )
+    model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
+    generated_ids = model.generate(
+        **model_inputs,
+        max_new_tokens=max_new_tokens
     )
-    return response.choices[0].message.content.strip()
+    output_ids = generated_ids[0][len(model_inputs.input_ids[0]):].tolist()
+    response = tokenizer.decode(output_ids, skip_special_tokens=True).strip("\n")
+    return response
 
 class ResumeScreener:
     def __init__(self):
-        # Load models
+        print("[ResumeScreener] Initializing...")
+        st.text("Initializing Screener: Loading embedding model...")
         self.embedding_model = load_embedding_model()
+        st.text("Initializing Screener: Loading cross-encoder model...")
         self.cross_encoder = load_cross_encoder()
+        print("[ResumeScreener] Initialized.")
+        st.text("Screener Ready.")
     
     def extract_text_from_file(self, file_path, file_type):
         """Extract text from various file types"""
@@ -213,72 +229,75 @@ class ResumeScreener:
     
     def advanced_pipeline_ranking(self, resume_texts, job_description):
         """Advanced pipeline: FAISS recall -> Cross-encoder -> BM25 -> LLM intent -> Final ranking"""
+        print("[Pipeline] Advanced Pipeline Ranking started.")
         if not resume_texts:
             return []
-            
-        # Stage 1: FAISS Recall (Top 50)
-        st.write("🔍 **Stage 1**: FAISS Recall - Finding top 50 candidates...")
+        st.info("🔍 Stage 1: FAISS Recall - Finding top candidates...")
+        print("[Pipeline] Calling faiss_recall.")
         top_50_indices = self.faiss_recall(resume_texts, job_description, top_k=50)
-        
-        # Stage 2: Cross-Encoder Re-ranking (Top 20)
-        st.write("🎯 **Stage 2**: Cross-Encoder Re-ranking - Selecting top 20...")
+        print(f"[Pipeline] faiss_recall returned {len(top_50_indices)} indices.")
+
+        st.info("🎯 Stage 2: Cross-Encoder Re-ranking - Selecting top candidates...")
+        print("[Pipeline] Calling cross_encoder_rerank.")
         top_20_results = self.cross_encoder_rerank(resume_texts, job_description, top_50_indices, top_k=20)
-        
-        # Stage 3: BM25 Keyword Matching
-        st.write("🔤 **Stage 3**: BM25 Keyword Matching...")
+        print(f"[Pipeline] cross_encoder_rerank returned {len(top_20_results)} results.")
+
+        st.info("🔤 Stage 3: BM25 Keyword Matching...")
+        print("[Pipeline] Calling add_bm25_scores.")
         top_20_with_bm25 = self.add_bm25_scores(resume_texts, job_description, top_20_results)
-        
-        # Stage 4: LLM Intent Analysis
-        st.write("🤖 **Stage 4**: LLM Intent Analysis...")
+        print(f"[Pipeline] add_bm25_scores processed.")
+
+        st.info("🤖 Stage 4: LLM Intent Analysis (Qwen3-1.7B)...")
+        print("[Pipeline] Calling add_intent_scores.")
         top_20_with_intent = self.add_intent_scores(resume_texts, job_description, top_20_with_bm25)
-        
-        # Stage 5: Final Combined Ranking (Top 5)
-        st.write("🏆 **Stage 5**: Final Combined Ranking...")
+        print(f"[Pipeline] add_intent_scores processed.")
+
+        st.info("🏆 Stage 5: Final Combined Ranking...")
+        print("[Pipeline] Calling calculate_final_scores.")
         final_results = self.calculate_final_scores(top_20_with_intent)
-        
+        print(f"[Pipeline] calculate_final_scores returned {len(final_results)} results.")
+        print("[Pipeline] Advanced Pipeline Ranking finished.")
         return final_results[:5]  # Return top 5
     
     def faiss_recall(self, resume_texts, job_description, top_k=50):
         """Stage 1: Use FAISS for initial recall to find top 50 resumes"""
-        try:
-            # Get job embedding
-            job_embedding = self.get_embedding(job_description)
-            
-            # Get resume embeddings
-            resume_embeddings = []
-            progress_bar = st.progress(0)
-            
-            for i, text in enumerate(resume_texts):
-                if text:
-                    embedding = self.embedding_model.encode(text[:8192], 
-                                                          convert_to_numpy=True,
-                                                          normalize_embeddings=True)
-                    resume_embeddings.append(embedding)
-                else:
-                    resume_embeddings.append(np.zeros(1024))
-                progress_bar.progress((i + 1) / len(resume_texts))
-            
-            progress_bar.empty()
-            
-            # Create FAISS index
-            resume_embeddings = np.array(resume_embeddings).astype('float32')
-            dimension = resume_embeddings.shape[1]
-            index = faiss.IndexFlatIP(dimension)  # Inner product for cosine similarity
-            index.add(resume_embeddings)
-            
-            # Search for top K
-            job_embedding = job_embedding.reshape(1, -1).astype('float32')
-            scores, indices = index.search(job_embedding, min(top_k, len(resume_texts)))
-            
-            return indices[0].tolist()
-            
-        except Exception as e:
-            st.error(f"Error in FAISS recall: {str(e)}")
-            # Fallback: return all indices
-            return list(range(min(top_k, len(resume_texts))))
+        print("[faiss_recall] Method started.")
+        st.text("FAISS Recall: Embedding job description...")
+        job_embedding = self.get_embedding(job_description)
+        print("[faiss_recall] Job description embedded.")
+        st.text(f"FAISS Recall: Embedding {len(resume_texts)} resumes...")
+        resume_embeddings = []
+        progress_bar = st.progress(0)
+        
+        for i, text in enumerate(resume_texts):
+            if text:
+                embedding = self.embedding_model.encode(text[:8192], 
+                                                      convert_to_numpy=True,
+                                                      normalize_embeddings=True)
+                resume_embeddings.append(embedding)
+            else:
+                resume_embeddings.append(np.zeros(1024))
+            progress_bar.progress((i + 1) / len(resume_texts))
+            if i % 10 == 0: # Print progress every 10 resumes
+                print(f"[faiss_recall] Embedded resume {i+1}/{len(resume_texts)}")
+        
+        progress_bar.empty()
+        print("[faiss_recall] All resumes embedded.")
+        st.text("FAISS Recall: Building FAISS index...")
+        resume_embeddings = np.array(resume_embeddings).astype('float32')
+        dimension = resume_embeddings.shape[1]
+        index = faiss.IndexFlatIP(dimension)  # Inner product for cosine similarity
+        index.add(resume_embeddings)
+        print("[faiss_recall] FAISS index built.")
+        st.text("FAISS Recall: Searching index...")
+        job_embedding = job_embedding.reshape(1, -1).astype('float32')
+        scores, indices = index.search(job_embedding, min(top_k, len(resume_texts)))
+        print("[faiss_recall] FAISS search complete.")
+        return indices[0].tolist()
     
     def cross_encoder_rerank(self, resume_texts, job_description, top_50_indices, top_k=20):
         """Stage 2: Use Cross-Encoder to re-rank top 50 and select top 20"""
+        print("[cross_encoder_rerank] Method started.")
         try:
             if not self.cross_encoder:
                 st.error("Cross-encoder not loaded!")
@@ -299,6 +318,8 @@ class ResumeScreener:
             if not pairs:
                 return [(idx, 0.0) for idx in top_50_indices[:top_k]]
             
+            st.text(f"Cross-Encoder: Preparing {len(pairs)} pairs for re-ranking...")
+            print(f"[cross_encoder_rerank] Prepared {len(pairs)} pairs.")
             # Get cross-encoder scores
             progress_bar = st.progress(0)
             scores = []
@@ -310,8 +331,11 @@ class ResumeScreener:
                 batch_scores = self.cross_encoder.predict(batch)
                 scores.extend(batch_scores)
                 progress_bar.progress(min(1.0, (i + batch_size) / len(pairs)))
+                print(f"[cross_encoder_rerank] Processed batch {i//batch_size + 1}")
             
             progress_bar.empty()
+            print("[cross_encoder_rerank] All pairs scored.")
+            st.text("Cross-Encoder: Re-ranking complete.")
             
             # Combine indices with scores and sort
             indexed_scores = list(zip(valid_indices, scores))
@@ -325,6 +349,8 @@ class ResumeScreener:
     
     def add_bm25_scores(self, resume_texts, job_description, top_20_results):
         """Stage 3: Add BM25 scores to top 20 resumes"""
+        print("[add_bm25_scores] Method started.")
+        st.text("BM25: Calculating keyword scores...")
         try:
             # Get texts for top 20
             top_20_texts = [resume_texts[idx] for idx, _ in top_20_results]
@@ -352,6 +378,8 @@ class ResumeScreener:
                 bm25_score = normalized_bm25[i] if i < len(normalized_bm25) else 0.15
                 results_with_bm25.append((idx, cross_score, bm25_score))
             
+            print("[add_bm25_scores] BM25 scores calculated and normalized.")
+            st.text("BM25: Keyword scores added.")
             return results_with_bm25
             
         except Exception as e:
@@ -360,63 +388,75 @@ class ResumeScreener:
     
     def add_intent_scores(self, resume_texts, job_description, top_20_with_bm25):
         """Stage 4: Add LLM intent analysis scores"""
-        try:
-            results_with_intent = []
-            progress_bar = st.progress(0)
-            
-            for i, (idx, cross_score, bm25_score) in enumerate(top_20_with_bm25):
-                intent_score = self.analyze_intent(resume_texts[idx], job_description)
-                results_with_intent.append((idx, cross_score, bm25_score, intent_score))
-                progress_bar.progress((i + 1) / len(top_20_with_bm25))
-            
-            progress_bar.empty()
-            return results_with_intent
-            
-        except Exception as e:
-            st.error(f"Error adding intent scores: {str(e)}")
-            return [(idx, cross_score, bm25_score, 0.1) for idx, cross_score, bm25_score in top_20_with_bm25]
+        print("[add_intent_scores] Method started.")
+        st.text(f"LLM Intent: Analyzing intent for {len(top_20_with_bm25)} candidates (Qwen3-1.7B)...")
+        results_with_intent = []
+        progress_bar = st.progress(0)
+        
+        for i, (idx, cross_score, bm25_score) in enumerate(top_20_with_bm25):
+            intent_score = self.analyze_intent(resume_texts[idx], job_description)
+            results_with_intent.append((idx, cross_score, bm25_score, intent_score))
+            progress_bar.progress((i + 1) / len(top_20_with_bm25))
+            print(f"[add_intent_scores] Intent analyzed for candidate {i+1}")
+        
+        progress_bar.empty()
+        print("[add_intent_scores] All intents analyzed.")
+        st.text("LLM Intent: Analysis complete.")
+        return results_with_intent
     
     def analyze_intent(self, resume_text, job_description):
-        """Analyze candidate's intent using LLM"""
+        """Analyze candidate's intent using Qwen3-1.7B LLM with thinking enabled."""
+        print(f"[analyze_intent] Analyzing intent for one resume (Qwen3-1.7B)...")
+        st.text("LLM Intent: Analyzing intent (Qwen3-1.7B)...")
         try:
-            # Truncate texts
-            resume_snippet = resume_text[:1500] if len(resume_text) > 1500 else resume_text
-            job_snippet = job_description[:800] if len(job_description) > 800 else job_description
+            resume_snippet = resume_text[:15000]
+            job_snippet = job_description[:5000]
             
-            prompt = f"""You are given a job description and a candidate's resume.
-Clearly answer: "Is the candidate likely seeking this job? Respond with 'Yes', 'Maybe', or 'No' and give a brief justification."
-
-Job Description:
-{job_snippet}
-
-Candidate Resume:
-{resume_snippet}
-
-Response format:
-Intent: [Yes/Maybe/No]
-Reason: [Brief justification]"""
+            prompt = f"""You are given a job description and a candidate's resume.\nAnalyze the candidate's resume in detail against the job description to determine if they are genuinely seeking this specific job, or if their profile is a more general fit or perhaps a mismatch.\nProvide a step-by-step thought process for your decision.\nFinally, clearly answer: \"Is the candidate likely seeking THIS SPECIFIC job? Respond with 'Yes', 'Maybe', or 'No' and give a brief justification based on your thought process.\"\n\nJob Description:\n{job_snippet}\n\nCandidate Resume:\n{resume_snippet}\n\nResponse format:\n<think>\n[Your detailed step-by-step thought process comparing resume to JD, noting specific alignments or mismatches that indicate intent. Be thorough.]\n</think>\nIntent: [Yes/Maybe/No]\nReason: [Brief justification based on your thought process]"""
             
-            response = vllm_chat_completion(
+            response_text = generate_qwen3_response(
                 prompt,
-                st.session_state.vllm_4b_endpoint,
-                max_tokens=100
+                st.session_state.qwen3_1_7b_tokenizer,
+                st.session_state.qwen3_1_7b_model,
+                max_new_tokens=20000
             )
+            print(f"[analyze_intent] Qwen3-1.7B full response (first 100 chars): {response_text[:100]}...")
+
+            thinking_content = "No detailed thought process extracted."
+            intent_decision_part = response_text
+
+            think_start_tag = "<think>"
+            think_end_tag = "</think>"
             
-            # Parse response
-            response_lower = response.lower()
-            if 'intent: yes' in response_lower or 'intent:yes' in response_lower:
-                return 0.3
-            elif 'intent: maybe' in response_lower or 'intent:maybe' in response_lower:
-                return 0.1
+            start_index = response_text.find(think_start_tag)
+            end_index = response_text.rfind(think_end_tag)
+
+            if start_index != -1 and end_index != -1 and start_index < end_index:
+                thinking_content = response_text[start_index + len(think_start_tag):end_index].strip()
+                intent_decision_part = response_text[end_index + len(think_end_tag):].strip()
+                print(f"[analyze_intent] Thinking content extracted (first 50 chars): {thinking_content[:50]}...")
             else:
-                return 0.0
-                
+                print("[analyze_intent] <think> block not found or malformed in response.")
+
+            response_lower = intent_decision_part.lower()
+            intent_score = 0.1
+            if 'intent: yes' in response_lower or 'intent:yes' in response_lower:
+                intent_score = 0.3
+            elif 'intent: no' in response_lower or 'intent:no' in response_lower:
+                intent_score = 0.0
+            
+            print(f"[analyze_intent] Parsed Intent: {intent_score}, Decision part: {intent_decision_part[:100]}...")
+            return intent_score
+
         except Exception as e:
-            st.warning(f"Error analyzing intent: {str(e)}")
-            return 0.1  # Default to "Maybe"
+            st.warning(f"Error analyzing intent with Qwen3-1.7B: {str(e)}")
+            print(f"[analyze_intent] EXCEPTION: {str(e)}")
+            return 0.1
     
     def calculate_final_scores(self, results_with_all_scores):
         """Stage 5: Calculate final combined scores"""
+        print("[calculate_final_scores] Method started.")
+        st.text("Final Ranking: Calculating combined scores...")
         try:
             final_results = []
             
@@ -438,6 +478,8 @@ Reason: [Brief justification]"""
             # Sort by final score
             final_results.sort(key=lambda x: x['final_score'], reverse=True)
             
+            print("[calculate_final_scores] Final scores calculated and sorted.")
+            st.text("Final Ranking: Complete.")
             return final_results
             
         except Exception as e:
@@ -482,99 +524,6 @@ Reason: [Brief justification]"""
                     found_skills.append(word)
         
         return list(set(found_skills))[:15]  # Return top 15 unique skills
-    
-    def generate_simple_explanation(self, score, semantic_score, bm25_score, skills):
-        """Generate simple explanation for the match (fallback)"""
-        if score > 0.8:
-            quality = "excellent"
-        elif score > 0.6:
-            quality = "strong"
-        elif score > 0.4:
-            quality = "moderate"
-        else:
-            quality = "limited"
-        
-        explanation = f"This candidate shows {quality} alignment with the position (score: {score:.2f}). "
-        
-        if semantic_score > bm25_score:
-            explanation += f"The resume demonstrates strong conceptual relevance ({semantic_score:.2f}) suggesting good experience fit. "
-        else:
-            explanation += f"The resume has high keyword match ({bm25_score:.2f}) indicating direct skill alignment. "
-        
-        if skills:
-            explanation += f"Key matching competencies include: {', '.join(skills[:5])}."
-        
-        return explanation
-    
-    def generate_llm_explanation(self, resume_text, job_description, score, skills, max_retries=3):
-        """Generate detailed explanation using Qwen3-14B"""
-        if not st.session_state.vllm_14b_endpoint:
-            return self.generate_simple_explanation(score, score, score, skills)
-        
-        # Truncate texts to manage token limits
-        resume_snippet = resume_text[:2000] if len(resume_text) > 2000 else resume_text
-        job_snippet = job_description[:1000] if len(job_description) > 1000 else job_description
-        
-        prompt = f"""You are an expert HR analyst. Analyze this individual candidate's resume against the job requirements and write EXACTLY 150 words explaining why this specific candidate is suitable for the position.
-
-Structure your 150-word analysis as follows:
-1. Experience alignment (40-50 words)
-2. Key strengths and skills match (40-50 words) 
-3. Unique value proposition (40-50 words)
-4. Overall recommendation (10-20 words)
-
-Job Requirements:
-{job_snippet}
-
-Candidate's Resume:
-{resume_snippet}
-
-Identified Matching Skills: {', '.join(skills[:10])}
-Compatibility Score: {score:.1%}
-
-Write a professional, detailed 150-word analysis for THIS INDIVIDUAL CANDIDATE:"""
-        
-        for attempt in range(max_retries):
-            try:
-                response = vllm_chat_completion(
-                    prompt,
-                    st.session_state.vllm_14b_endpoint,
-                    max_tokens=200
-                )
-                
-                # Extract the response and ensure it's about 150 words
-                explanation = response.strip()
-                word_count = len(explanation.split())
-                
-                # If response is close to 150 words (130-170), accept it
-                if 130 <= word_count <= 170:
-                    return explanation
-                
-                # If response is too short or too long, try again with adjusted prompt
-                if word_count < 130:
-                    # Response too short, try again
-                    continue
-                elif word_count > 170:
-                    # Response too long, truncate to approximately 150 words
-                    words = explanation.split()
-                    truncated = ' '.join(words[:150])
-                    # Add proper ending if truncated
-                    if not truncated.endswith('.'):
-                        truncated += '.'
-                    return truncated
-                
-                return explanation
-                
-            except Exception as e:
-                if attempt < max_retries - 1:
-                    time.sleep(2)  # Wait before retry
-                    continue
-                else:
-                    # Fallback to simple explanation
-                    return self.generate_simple_explanation(score, score, score, skills)
-        
-        # If all retries failed, use simple explanation
-        return self.generate_simple_explanation(score, score, score, skills)
 
 def create_download_link(df, filename="resume_screening_results.csv"):
     """Create download link for results"""
@@ -584,7 +533,7 @@ def create_download_link(df, filename="resume_screening_results.csv"):
 
 # Main App Interface
 st.title("🎯 AI-Powered Resume Screener")
-st.markdown("*Find the perfect candidates using BAAI/bge-large-en-v1.5 embeddings and Qwen3-14B explanations*")
+st.markdown("*Find the perfect candidates using BAAI/bge-large-en-v1.5 embeddings and Qwen3-1.7B for intent analysis*")
 st.markdown("---")
 
 # Initialize screener
@@ -611,7 +560,6 @@ if st.session_state.resume_texts:
             st.session_state.resume_texts = []
             st.session_state.file_names = []
             st.session_state.results = []
-            st.session_state.explanations_generated = False
             st.session_state.current_job_description = ""
             st.rerun()
 
@@ -782,12 +730,15 @@ with col1:
                  disabled=not (job_description and st.session_state.resume_texts),
                  type="primary",
                  help="Run the complete 5-stage advanced pipeline"):
+        print("--- Advanced Pipeline Analysis Button Clicked ---")
         if len(st.session_state.resume_texts) == 0:
             st.error("❌ Please upload resumes first!")
         elif not job_description.strip():
             st.error("❌ Please enter a job description!")
         else:
+            print("[UI Button] Pre-checks passed. Starting spinner and pipeline.")
             with st.spinner("🚀 Running Advanced Pipeline Analysis..."):
+                st.text("Pipeline Initiated: Starting advanced analysis...")
                 try:
                     # Run the advanced pipeline
                     pipeline_results = screener.advanced_pipeline_ranking(
@@ -814,81 +765,19 @@ with col1:
                             'intent_score': result_data['intent_score'],
                             'skills': skills,
                             'text': text,
-                            'text_preview': text[:500] + "..." if len(text) > 500 else text,
-                            'explanation': None  # No detailed explanation yet
+                            'text_preview': text[:500] + "..." if len(text) > 500 else text
                         })
                     
-                    # Add simple explanations for now
-                    for result in results:
-                        result['explanation'] = screener.generate_simple_explanation(
-                            result['final_score'], 
-                            result['cross_encoder_score'], 
-                            result['bm25_score'], 
-                            result['skills']
-                        )
-                    
                     # Store in session state
                     st.session_state.results = results
-                    st.session_state.explanations_generated = False
                     st.session_state.current_job_description = job_description
                     
                     st.success(f"🚀 Advanced pipeline complete! Found top {len(st.session_state.results)} candidates.")
+                    st.text("Displaying Top Candidates...")
                     
                 except Exception as e:
                     st.error(f"❌ Error during analysis: {str(e)}")
 
-# Second button: Generate AI explanations (slower, optional)
-with col2:
-    # Show this button only if we have results and LLM is enabled
-    show_explanation_button = (
-        st.session_state.results and 
-        use_llm_explanations and 
-        st.session_state.vllm_14b_endpoint and
-        not st.session_state.explanations_generated
-    )
-    
-    if show_explanation_button:
-        if st.button("🤖 Generate AI Explanations", 
-                     type="secondary",
-                     help="Generate detailed 150-word explanations using Qwen3-14B (takes longer)"):
-            with st.spinner("🤖 Generating detailed AI explanations..."):
-                try:
-                    explanation_progress = st.progress(0)
-                    explanation_text = st.empty()
-                    
-                    for i, result in enumerate(st.session_state.results):
-                        explanation_text.text(f"🤖 Generating AI explanation for candidate {i+1}/{len(st.session_state.results)}...")
-                        
-                        llm_explanation = screener.generate_llm_explanation(
-                            result['text'],
-                            st.session_state.current_job_description,
-                            result['final_score'],
-                            result['skills']
-                        )
-                        result['explanation'] = llm_explanation
-                        
-                        explanation_progress.progress((i + 1) / len(st.session_state.results))
-                    
-                    explanation_progress.empty()
-                    explanation_text.empty()
-                    
-                    # Mark explanations as generated
-                    st.session_state.explanations_generated = True
-                    
-                    st.success(f"🤖 AI explanations generated for all {len(st.session_state.results)} candidates!")
-                    
-                except Exception as e:
-                    st.error(f"❌ Error generating explanations: {str(e)}")
-    
-    elif st.session_state.results and st.session_state.explanations_generated:
-        st.info("✅ AI explanations already generated!")
-    
-    elif st.session_state.results and not use_llm_explanations:
-        st.info("💡 Enable 'Generate AI Explanations' in sidebar to use this feature")
-    
-    elif st.session_state.results and not st.session_state.vllm_14b_endpoint:
-        st.warning("⚠️ LLM model not available. Check your Hugging Face token.")
-
 # Display Results
 if st.session_state.results:
     st.header("🏆 Top Candidates")
@@ -956,7 +845,6 @@ if st.session_state.results:
                 "Intent_Score": result['intent_score'],
                 "Intent_Analysis": intent_text,
                 "Skills": "; ".join(result['skills']),
-                "AI_Explanation": result['explanation'],
                 "Resume_Preview": result['text_preview']
             })
         
@@ -987,9 +875,6 @@ if st.session_state.results:
                             st.write(f"• {skill}")
                 
                 with col2:
-                    st.write("**💡 AI-Generated Match Analysis:**")
-                    st.info(result['explanation'])
-                    
                     st.write("**📄 Resume Preview:**")
                     st.text_area("", result['text_preview'], height=200, disabled=True, key=f"preview_{result['rank']}")
     
@@ -1049,7 +934,6 @@ with col1:
         st.session_state.resume_texts = []
         st.session_state.file_names = []
         st.session_state.results = []
-        st.session_state.explanations_generated = False
         st.session_state.current_job_description = ""
         st.success("✅ Resumes cleared!")
         st.rerun()
@@ -1059,7 +943,6 @@ with col2:
         st.session_state.resume_texts = []
         st.session_state.file_names = []
         st.session_state.results = []
-        st.session_state.explanations_generated = False
         st.session_state.current_job_description = ""
         
         if torch.cuda.is_available():
@@ -1073,7 +956,7 @@ st.markdown("---")
 st.markdown(
     """
     <div style='text-align: center; color: #666;'>
-        🚀 Powered by BAAI/bge-large-en-v1.5 & Qwen3-14B | Built with Streamlit
+        🚀 Powered by BAAI/bge-large-en-v1.5 & Qwen3-1.7B | Built with Streamlit
     </div>
     """, 
     unsafe_allow_html=True

requirements.txt

@@ -1,6 +1,6 @@
 streamlit==1.31.0
-transformers==4.52.0
-torch==2.6.0
+transformers>=4.51.0
+torch==2.1.2
 pdfplumber==0.10.1
 PyPDF2==3.0.1
 python-docx==1.0.1
@@ -8,15 +8,12 @@ nltk==3.8.1
 faiss-cpu==1.7.4
 rank-bm25==0.2.2
 pandas==2.1.3
-numpy>=1.24,<2.0
+numpy==1.24.3
 tqdm==4.66.1
 huggingface-hub==0.30.0
 bitsandbytes==0.44.1
 accelerate==0.27.2
 datasets==2.18.0
 sentence-transformers==2.7.0
-tokenizers==0.21.1
 plotly==5.18.0
-einops
-vllm>=0.8.5
-openai>=1.0.0
+einops