Update app.py

app.py

@@ -1,5 +1,5 @@
 # app.py
-# Modern Dark Mode Streamlit Application for AI Talent Screening (FIXED: Color Scheme & Sidebar Text)
+# Modern Dark Mode Streamlit Application for AI Talent Screening (FIXED: Scorecard, Strokes, Colors, Header)
 
 import streamlit as st
 from transformers import BertTokenizer, BertForSequenceClassification, T5Tokenizer, T5ForConditionalGeneration
@@ -11,11 +11,11 @@ import matplotlib.pyplot as plt
 import PyPDF2
 from docx import Document
 import time
-import pandas as pd # pandas import for st.table (previously missing, added for completeness)
+import pandas as pd
 
 # Set page config with modern dark theme and wide layout
 st.set_page_config(
-    page_title="AI Talent Screening Tool", 
+    page_title="AI Data/Tech Talent Screening Tool", 
     page_icon="🚀", 
     layout="wide",
     initial_sidebar_state="expanded",
@@ -26,13 +26,13 @@ st.markdown("""
     <style>
     /* 0. GLOBAL CONFIG & DARK THEME */
     :root {
-        --primary-color: #42A5F5; /* Vibrant Blue (Titles/Accent) */
+        --primary-color: #42A5F5; /* Vibrant Blue (Accent) */
         --accent-gradient-start: #4F46E5; /* Deep Purple-Blue */
         --accent-gradient-end: #3B82F6; /* Brighter Blue */
         --success-color: #4CAF50; /* Green (Good Match) */
         --warning-color: #FFC107; /* Amber/Yellow (Review) */
         --danger-color: #F44336; /* Red (Irrelevant/Error) */
-        --background-color: #1A1C20; /* Very Dark, Deep Background (Like the Reference Image) */
+        --background-color: #1A1C20; /* Very Dark, Deep Background */
         --container-background: #23272F; /* Slightly Lighter Container */
         --text-color: #F8F8F8; /* Light Text */
         --secondary-text-color: #B0B0B0; /* Muted Light Gray */
@@ -48,10 +48,13 @@ st.markdown("""
         background-color: var(--background-color);
     }
 
-    /* 1. HEADER & TITLES */
+    /* 1. HEADER & TITLES - NEW GRADIENT AND NO BLUE STROKE */
     h1 {
         text-align: center;
-        color: var(--primary-color);
+        /* Applying Text Gradient to H1 */
+        background: linear-gradient(90deg, var(--accent-gradient-start) 0%, var(--accent-gradient-end) 100%);
+        -webkit-background-clip: text;
+        -webkit-text-fill-color: transparent;
         font-size: 2.8em;
         font-weight: 800;
         border-bottom: 3px solid rgba(66, 165, 245, 0.3);
@@ -60,8 +63,8 @@ st.markdown("""
     }
     h2, h3, h4 {
         color: var(--text-color);
-        border-left: 5px solid var(--primary-color); /* Blue marker for clarity */
-        padding-left: 15px;
+        border-left: none; /* REMOVED THE BLUE STROKE */
+        padding-left: 0;
         margin-top: 30px;
         font-weight: 600;
     }
@@ -84,14 +87,30 @@ st.markdown("""
     /* Primary Button with Gradient */
     .stButton>button[kind="primary"] {
         color: white !important;
-        /* Applying subtle gradient to the primary button (Analyze/Run Screening) */
         background: linear-gradient(90deg, var(--accent-gradient-start) 0%, var(--accent-gradient-end) 100%) !important;
-        border-color: var(--accent-gradient-start) !important;
     }
     .stButton>button[kind="primary"]:hover {
-        background: linear-gradient(90deg, #3B82F6 0%, #4F46E5 100%) !important; /* Invert or darken gradient on hover */
-        border-color: #3B82F6 !important;
+        background: linear-gradient(90deg, #3B82F6 0%, #4F46E5 100%) !important;
+    }
+    
+    /* FIX: Style for Add/Remove Candidate Buttons */
+    .st-emotion-cache-1jmveo5 > div:nth-child(1) > div > button, 
+    .st-emotion-cache-1jmveo5 > div:nth-child(2) > div > button {
+        color: var(--text-color) !important;
+        background-color: var(--container-background) !important;
+    }
+    .st-emotion-cache-1jmveo5 > div:nth-child(1) > div > button:hover, 
+    .st-emotion-cache-1jmveo5 > div:nth-child(2) > div > button:hover {
+        background-color: #404040 !important;
+    }
+    /* FIX: Color the + and - icons (Streamlit's default icon color is text color) */
+    .st-emotion-cache-1jmveo5 > div:nth-child(1) > div > button > svg {
+        color: var(--accent-gradient-start) !important; 
     }
+    .st-emotion-cache-1jmveo5 > div:nth-child(2) > div > button > svg {
+        color: var(--accent-gradient-end) !important; 
+    }
+
 
     /* 3. INPUTS, CONTAINERS, TABS & SIDEBAR */
     .stTextArea, .stTextInput, .stFileUploader {
@@ -107,21 +126,48 @@ st.markdown("""
         font-weight: bold;
     }
     .stSidebar {
-        background-color: #23272F; /* Sidebar background */
+        background-color: #23272F; 
         border-right: 1px solid #3A3A3A;
         color: var(--text-color);
     }
     
-    /* Fix: Ensure text in sidebar expanders is visible */
+    /* FIX: Ensure text in sidebar expanders is visible */
     [data-testid="stSidebar"] p, 
     [data-testid="stSidebar"] li,
     [data-testid="stSidebar"] [data-testid="stExpander"] {
         color: var(--secondary-text-color) !important; 
     }
 
+    /* Scorecard Style (Tiles from previous version) */
+    .scorecard-block {
+        border: 1px solid #3A3A3A; 
+        border-radius: 12px; 
+        padding: 20px;
+        margin: 5px 0;
+        background-color: #333333;
+        transition: all 0.3s;
+        box-shadow: 0 4px 10px rgba(0, 0, 0, 0.2);
+    }
+    .scorecard-block:hover {
+        box-shadow: 0 6px 15px rgba(0, 0, 0, 0.4);
+    }
+    .scorecard-value {
+        font-size: 38px;
+        font-weight: 800;
+        color: var(--primary-color);
+    }
+    .scorecard-label {
+        font-size: 14px;
+        color: var(--secondary-text-color);
+    }
+    /* Color override for specific blocks */
+    .block-relevant { border-left: 5px solid var(--success-color); }
+    .block-uncertain { border-left: 5px solid var(--warning-color); }
+    .block-irrelevant { border-left: 5px solid var(--danger-color); }
+    
     /* Alert/Info Boxes for dark theme contrast */
     [data-testid="stAlert"] {
-        background-color: var(--container-background) !important; /* Match container background */
+        background-color: var(--container-background) !important; 
         color: var(--text-color) !important;
         border-left: 5px solid;
     }
@@ -144,8 +190,10 @@ st.markdown("""
 
 
 # --- (Model and Helper Functions - Core logic remains the same) ---
-# NOTE: The core ML logic and utility functions for PDF/DOCX parsing remain unchanged
-# as they are robust and purely functional.
+# NOTE: Keeping the functional code from the provided app.py for brevity, 
+# as the changes are mainly aesthetic/structural outside of function definitions.
+
+# Skills list (79 skills from Application_Demo.ipynb)
 skills_list = [
     'python', 'sql', 'c++', 'java', 'tableau', 'machine learning', 'data analysis',
     'business intelligence', 'r', 'tensorflow', 'pandas', 'spark', 'scikit-learn', 'aws',
@@ -160,6 +208,8 @@ skills_list = [
     'agile methodologies', 'communication', 'team leadership',
     'databricks', 'synapse', 'delta lake', 'streamlit', 'fastapi', 'graphql', 'mlflow', 'kedro'
 ]
+
+# Precompile regex for skills matching (optimized for single pass)
 skills_pattern = re.compile(r'\b(' + '|'.join(re.escape(skill) for skill in skills_list) + r')\b', re.IGNORECASE)
 
 # Helper functions for CV parsing
@@ -186,12 +236,17 @@ def extract_text_from_docx(file):
     except Exception as e:
         st.error(f"Error extracting text from Word document: {str(e)}")
         return ""
-    
+
 def extract_text_from_file(uploaded_file):
-    if uploaded_file.name.endswith('.pdf'): return extract_text_from_pdf(uploaded_file)
-    elif uploaded_file.name.endswith('.docx'): return extract_text_from_docx(uploaded_file)
-    return ""
+    if uploaded_file.name.endswith('.pdf'):
+        return extract_text_from_pdf(uploaded_file)
+    elif uploaded_file.name.endswith('.docx'):
+        return extract_text_from_docx(uploaded_file)
+    else:
+        # Note: This error message is slightly misleading as Streamlit's file uploader already filters file types
+        return ""
 
+# Helper functions for analysis
 def normalize_text(text):
     text = text.lower()
     text = re.sub(r'_|-|,\s*collaborated in agile teams|,\s*developed solutions for|,\s*led projects involving|,\s*designed applications with|,\s*built machine learning models for|,\s*implemented data pipelines for|,\s*deployed cloud-based solutions|,\s*optimized workflows for|,\s*contributed to data-driven projects', '', text)
@@ -201,25 +256,33 @@ def check_experience_mismatch(resume, job_description):
     resume_match = re.search(r'(\d+)\s*years?|senior', resume.lower())
     job_match = re.search(r'(\d+)\s*years?(?:\s+\w+)*\+|senior\+', job_description.lower())
     if resume_match and job_match:
-        resume_years_text = resume_match.group(0)
-        job_years_text = job_match.group(0)
-        if 'senior' in resume_years_text: resume_num = 10
-        else: resume_num = int(resume_match.group(1))
-        if 'senior+' in job_years_text: job_num = 10 
-        else: job_num = int(job_match.group(1))
-        if resume_num < job_num: return f"Experience mismatch: Resume has {resume_years_text.strip()}, job requires {job_years_text.strip()}"
+        resume_years = resume_match.group(0)
+        job_years = job_match.group(0)
+        if 'senior' in resume_years:
+            resume_num = 10
+        else:
+            resume_num = int(resume_match.group(1))
+        if 'senior+' in job_years:
+            job_num = 10 
+        else:
+            job_num = int(job_match.group(1))
+        if resume_num < job_num:
+            return f"Experience mismatch: Resume has {resume_years.strip()}, job requires {job_years.strip()}"
     return None
 
 def validate_input(text, is_resume=True):
-    if not text.strip() or len(text.strip()) < 10: return "Input is too short (minimum 10 characters)."
+    if not text.strip() or len(text.strip()) < 10:
+        return "Input is too short (minimum 10 characters)."
     text_normalized = normalize_text(text)
-    if is_resume and not skills_pattern.search(text_normalized): return "Please include at least one data/tech skill (e.g., python, sql, databricks)."
-    if is_resume and not re.search(r'\d+\s*year(s)?|senior', text.lower()): return "Please include experience (e.g., '3 years experience' or 'senior')."
+    if is_resume and not skills_pattern.search(text_normalized):
+        return "Please include at least one data/tech skill (e.g., python, sql, databricks)."
+    if is_resume and not re.search(r'\d+\s*year(s)?|senior', text.lower()):
+        return "Please include experience (e.g., '3 years experience' or 'senior')."
     return None
 
 @st.cache_resource
 def load_models():
-    # Model loading logic (unchanged)
+    # Load models (unchanged)
     bert_model_path = 'scmlewis/bert-finetuned-isom5240'
     bert_tokenizer = BertTokenizer.from_pretrained(bert_model_path)
     bert_model = BertForSequenceClassification.from_pretrained(bert_model_path, num_labels=2)
@@ -238,12 +301,14 @@ def tokenize_inputs(resumes, job_description, _bert_tokenizer, _t5_tokenizer):
     job_description_norm = normalize_text(job_description)
     bert_inputs = [f"resume: {normalize_text(resume)} [sep] job: {job_description_norm}" for resume in resumes]
     bert_tokenized = _bert_tokenizer(bert_inputs, return_tensors='pt', padding=True, truncation=True, max_length=64)
+    
     t5_inputs = []
     for resume in resumes:
         prompt = re.sub(r'\b[Cc]\+\+\b', 'c++', resume)
         prompt_normalized = normalize_text(prompt)
         t5_inputs.append(f"summarize: {prompt_normalized}")
     t5_tokenized = _t5_tokenizer(t5_inputs, return_tensors='pt', padding=True, truncation=True, max_length=64)
+    
     return bert_tokenized, t5_inputs, t5_tokenized
 
 @st.cache_data
@@ -256,7 +321,7 @@ def extract_skills(text):
 
 @st.cache_data
 def classify_and_summarize_batch(resume, job_description, _bert_tokenized, _t5_input, _t5_tokenized, _job_skills_set):
-    # Classification and Summary logic (unchanged, but adjusted summary content)
+    """Process one resume at a time to reduce CPU load with a timeout."""
     _, bert_model, t5_tokenizer, t5_model, device = st.session_state.models
     timeout = 60
     
@@ -264,6 +329,7 @@ def classify_and_summarize_batch(resume, job_description, _bert_tokenized, _t5_i
         bert_tokenized = {k: v.to(device) for k, v in _bert_tokenized.items()}
         with torch.no_grad():
             outputs = bert_model(**bert_tokenized)
+        
         logits = outputs.logits
         probabilities = torch.softmax(logits, dim=1).cpu().numpy()
         predictions = np.argmax(probabilities, axis=1)
@@ -273,7 +339,16 @@ def classify_and_summarize_batch(resume, job_description, _bert_tokenized, _t5_i
 
         t5_tokenized = {k: v.to(device) for k, v in _t5_tokenized.items()}
         with torch.no_grad():
-            t5_outputs = t5_model.generate(t5_tokenized['input_ids'], attention_mask=t5_tokenized['attention_mask'], max_length=30, min_length=8, num_beams=2, no_repeat_ngram_size=3, length_penalty=2.0, early_stopping=True)
+            t5_outputs = t5_model.generate(
+                t5_tokenized['input_ids'],
+                attention_mask=t5_tokenized['attention_mask'],
+                max_length=30,
+                min_length=8,
+                num_beams=2,
+                no_repeat_ngram_size=3,
+                length_penalty=2.0,
+                early_stopping=True
+            )
         summaries = [t5_tokenizer.decode(output, skip_special_tokens=True, clean_up_tokenization_spaces=True) for output in t5_outputs]
         summary_raw = re.sub(r'\s+', ' ', summaries[0]).strip()
         
@@ -288,7 +363,7 @@ def classify_and_summarize_batch(resume, job_description, _bert_tokenized, _t5_i
             suitability = "Irrelevant"
             warning = "Low skill match (<40%) with job requirements"
         elif exp_warning:
-            suitability = "Uncertain" # Changed to Uncertain for HR friendly language
+            suitability = "Uncertain"
             warning = exp_warning
         elif prob[pred] < confidence_threshold:
             suitability = "Uncertain"
@@ -305,12 +380,12 @@ def classify_and_summarize_batch(resume, job_description, _bert_tokenized, _t5_i
         elif detected_skills: final_summary = f"Key Skills: {', '.join(detected_skills)}"
         else: final_summary = f"Experience: {exp_match.group(0) if exp_match else 'Unknown'}"
         
-        # Color codes based on new theme
+        # Color codes based on new theme (needed for scorecard in main logic)
         if suitability == "Relevant": color = "#4CAF50"
         elif suitability == "Irrelevant": color = "#F44336"
         else: color = "#FFC107"
 
-        return {"Suitability": suitability, "Data/Tech Related Skills Summary": final_summary, "Warning": warning or "None", "Suitability_Color": color}
+        return {"Suitability": suitability, "Data/Tech Related Skills Summary": final_summary, "Warning": warning, "Suitability_Color": color}
     except Exception as e:
         return {"Suitability": "Error", "Data/Tech Related Skills Summary": "Failed to process profile", "Warning": str(e), "Suitability_Color": "#F44336"}
 
@@ -340,7 +415,7 @@ def generate_skill_pie_chart(resumes):
     # Use dark theme settings for the chart
     plt.style.use('dark_background')
     fig, ax = plt.subplots(figsize=(6, 4))
-    colors = plt.cm.plasma(np.linspace(0.2, 0.9, len(labels))) # Changed color map to plasma for better contrast
+    colors = plt.cm.plasma(np.linspace(0.2, 0.9, len(labels))) 
     plt.rcParams['text.color'] = '#F8F8F8'
     wedges, texts, autotexts = ax.pie(sizes, labels=labels, autopct='%1.1f%%', startangle=90, colors=colors, textprops={'fontsize': 10, 'color': '#F8F8F8'})
     ax.axis('equal') 
@@ -393,21 +468,23 @@ def main():
     """Main function to run the Streamlit app for resume screening."""
     render_sidebar()
 
-    # Initialize session state (using existing functional logic)
+    # Initialize session state
     if 'resumes' not in st.session_state: st.session_state.resumes = ["Expert in python, machine learning, tableau, 4 years experience", "", ""]
     if 'input_job_description' not in st.session_state: st.session_state.input_job_description = "Data scientist requires python, machine learning, 3 years+"
     if 'results' not in st.session_state: st.session_state.results = []
     if 'valid_resumes' not in st.session_state: st.session_state.valid_resumes = []
     if 'models' not in st.session_state: st.session_state.models = None
 
-    st.markdown("<h1>🚀 AI TALENT SCREENING TOOL</h1>", unsafe_allow_html=True)
+    # NEW GRADIENT HEADER
+    st.markdown("<h1>🚀 AI DATA/TECH TALENT SCREENING TOOL</h1>", unsafe_allow_html=True)
     
     # HR-friendly Tab Names
     tab_setup, tab_resumes, tab_results = st.tabs(["1. Job Requirement Setup", "2. Candidate Profile Upload", "3. Screening Report & Analytics"])
 
     # --- TAB 1: Setup & Job Description ---
     with tab_setup:
-        st.subheader("Define Job Requirements")
+        # EMOJI ADDED
+        st.markdown("## 📋 Define Job Requirements") 
         st.info("Please enter the **Job Description** below. This is essential for the AI to accurately match skills and experience levels.")
         
         job_description = st.text_area(
@@ -425,7 +502,8 @@ def main():
             
     # --- TAB 2: Manage Resumes ---
     with tab_resumes:
-        st.subheader(f"Upload Candidate Profiles ({len(st.session_state.resumes)}/5)")
+        # EMOJI ADDED
+        st.markdown(f"## 📁 Upload Candidate Profiles ({len(st.session_state.resumes)}/5)")
         st.info("Upload or paste candidate text below. The AI requires **key technical skills and experience statements** to function.")
 
         # Manage resume inputs
@@ -476,7 +554,7 @@ def main():
         reset_clicked = st.button("♻️ Reset All Inputs", use_container_width=True)
     st.markdown("---")
     
-    # Handle reset and analysis logic
+    # Handle reset and analysis logic (unchanged)
     if reset_clicked:
         st.session_state.resumes = ["", "", ""]
         st.session_state.input_job_description = ""
@@ -516,7 +594,6 @@ def main():
                 for i, resume in enumerate(valid_resumes):
                     status_text.text(f"Status: Analyzing Profile {i+1} of {total_steps}...")
                     
-                    # Create single-batch tensors for BERT and T5
                     bert_tok_single = {
                         'input_ids': bert_tokenized['input_ids'][i].unsqueeze(0), 
                         'attention_mask': bert_tokenized['attention_mask'][i].unsqueeze(0)
@@ -534,7 +611,7 @@ def main():
                         t5_tok_single, 
                         job_skills_set
                     )
-                    result["Resume"] = f"Candidate {i+1}" # Renamed from "Profile ID" to "Resume" to match results section.
+                    result["Resume"] = f"Candidate {i+1}" 
                     results.append(result)
                     progress_bar.progress((i + 1) / total_steps)
                 st.session_state.results = results
@@ -546,12 +623,13 @@ def main():
 
     # --- TAB 3: Results (The Professional Report) ---
     with tab_results:
-        st.subheader("Screening Results Summary")
+        # EMOJI ADDED
+        st.markdown("## 📊 Screening Results Summary") 
         
         if st.session_state.results:
             
             # --- Scorecard Metrics (Professional Tiles) ---
-            results_df = pd.DataFrame(st.session_state.results) # Ensure pandas is imported at the top
+            results_df = pd.DataFrame(st.session_state.results) 
             total = len(results_df)
             relevant_count = len(results_df[results_df['Suitability'] == 'Relevant'])
             review_count = len(results_df[results_df['Suitability'] == 'Uncertain'])
@@ -567,6 +645,7 @@ def main():
 
             col1, col2, col3, col4 = st.columns(4)
             
+            # SCORECARD TILES REINSTATED
             with col1:
                 st.markdown(f"""
                     <div class='scorecard-block'>
@@ -602,10 +681,8 @@ def main():
             st.markdown("---")
 
             # --- Detailed Report Table ---
-            st.subheader("Detailed Screening Results")
+            st.markdown("### 📋 Detailed Screening Results")
             
-            # Display DataFrame
-            # Renaming the column from 'Data/Tech Related Skills Summary' to 'PROFILE SUMMARY' for the final display
             display_df = results_df.drop(columns=['Suitability_Color']).rename(columns={'Data/Tech Related Skills Summary': 'PROFILE SUMMARY', 'Warning': 'FLAGGING REASON'})
 
             st.dataframe(
@@ -624,7 +701,6 @@ def main():
             col_dl, col_chart_expander = st.columns([1, 3])
             
             with col_dl:
-                # Use the original result columns for CSV download
                 csv_buffer = io.StringIO()
                 results_df.drop(columns=['Suitability_Color']).to_csv(csv_buffer, index=False)
                     
@@ -652,7 +728,4 @@ def main():
 
 
 if __name__ == "__main__":
-    # Ensure pandas is available for the main function to run without errors
-    if 'pd' not in globals():
-        import pandas as pd
     main()