Update app.py

app.py

@@ -41,153 +41,121 @@ with st.sidebar:
 
 
 
-st.subheader ("Job Description", divider = "orange")
+        
+ 
+  
+import streamlit as st
+import pandas as pd
+from pypdf import PdfReader
+from transformers import pipeline
+from sklearn.feature_extraction.text import TfidfVectorizer
+from sklearn.metrics.pairwise import cosine_similarity
+import plotly.express as px
 
-txt = st.text_area("Paste the job description and then press Ctrl + Enter", key = "text 1")
+st.subheader("Job Description", divider="orange")
+txt = st.text_area("Paste the job description and then press Ctrl + Enter", key="text_job_desc")
 job = pd.Series(txt, name="Text")
 
-st.subheader("Candidate Profile 1", divider = "red")
-
-
-if 'upload_count' not in st.session_state:
-    st.session_state['upload_count'] = 0
-
-max_attempts = 3
-
-if st.session_state['upload_count'] < max_attempts:
-    uploaded_files = st.file_uploader(
-        "Upload your resume", accept_multiple_files=True, type="pdf", key="candidate 1"
+st.subheader("Candidate Profile 1", divider="red")
+if 'upload_count_candidate1' not in st.session_state:
+    st.session_state['upload_count_candidate1'] = 0
+max_attempts_candidate1 = 3
+if st.session_state['upload_count_candidate1'] < max_attempts_candidate1:
+    uploaded_files_candidate1 = st.file_uploader(
+        "Upload your resume", accept_multiple_files=True, type="pdf", key="candidate_1_upload"
     )
-
-if uploaded_files:
-    st.session_state['upload_count'] += 1
-for uploaded_file in uploaded_files:
-    pdf_reader = PdfReader(uploaded_file)
-    text_data = ""
-    for page in pdf_reader.pages:
-        text_data += page.extract_text()
-        data = pd.Series(text_data, name = 'Text')
-        
-        
-        frames = [job, data]
-        result = pd.concat(frames)
-        
-
-        model = GLiNER.from_pretrained("urchade/gliner_base")
-        labels = ["person", "country","organization", "role", "skills", "year"]
-        entities = model.predict_entities(text_data, labels)
-        df = pd.DataFrame(entities)
-        
-
-        st.title("Profile of candidate 1")
-        fig = px.treemap(entities, path=[px.Constant("all"), 'text', 'label'],
-                values='score', color='label')
-        fig.update_layout(margin = dict(t=50, l=25, r=25, b=25))
-        st.plotly_chart(fig, key = "tv")
-        
-       
-            
-        
-        vectorizer = TfidfVectorizer()
-        tfidf_matrix = vectorizer.fit_transform(result)
-        tfidf_df = pd.DataFrame(tfidf_matrix.toarray(), columns=vectorizer.get_feature_names_out())
-    
-        cosine_sim_matrix = cosine_similarity(tfidf_matrix)
-        cosine_sim_df = pd.DataFrame(cosine_sim_matrix)
-        
-
-        
-        st.subheader("Measuring similarity between keywords of candidate profile 1 and job description")
-        fig = px.imshow(cosine_sim_df, text_auto=True, labels=dict(x="Keyword similarity", y="Resumes", color="Productivity"),
-                        x=['Resume 1', 'Jon Description'],
-                        y=['Resume 1', 'Job Description'])
-        st.plotly_chart(fig, key = "bar")
-        
-
-        
-        for i, similarity_score in enumerate(cosine_sim_matrix[0][1:]):
-            st.write(f"Similarity of job description with candidate profile 1. {i + 1}: {similarity_score:.4f}")
-            st.info("A score closer to 1 (0.80, 0.90) means higher similarity between candidate profile 1 and job description. A score closer to 0 (0.20, 0.30) means lower similarity between candidate profile 1 and job description.")
-
+    if uploaded_files_candidate1:
+        st.session_state['upload_count_candidate1'] += 1
+        for uploaded_file in uploaded_files_candidate1:
+            pdf_reader = PdfReader(uploaded_file)
+            text_data = ""
+            for page in pdf_reader.pages:
+                text_data += page.extract_text()
+            data_candidate1 = pd.Series(text_data, name='Text')
+            frames_candidate1 = [job, data_candidate1]
+            result_candidate1 = pd.concat(frames_candidate1)
+            model_candidate1 = GLiNER.from_pretrained("urchade/gliner_base")
+            labels_candidate1 = ["person", "country", "organization", "role", "skills", "year"]
+            entities_candidate1 = model_candidate1(text_data, labels=labels_candidate1)
+            df_candidate1 = pd.DataFrame(entities_candidate1)
+
+            st.title("Profile of candidate 1")
+            fig_entities_candidate1 = px.treemap(entities_candidate1, path=[px.Constant("all"), 'text', 'label'],
+                                                values='score', color='label')
+            fig_entities_candidate1.update_layout(margin=dict(t=50, l=25, r=25, b=25))
+            st.plotly_chart(fig_entities_candidate1, key="tv_candidate1")
+
+            vectorizer_candidate1 = TfidfVectorizer()
+            tfidf_matrix_candidate1 = vectorizer_candidate1.fit_transform(result_candidate1)
+            tfidf_df_candidate1 = pd.DataFrame(tfidf_matrix_candidate1.toarray(),
+                                               columns=vectorizer_candidate1.get_feature_names_out())
+            cosine_sim_matrix_candidate1 = cosine_similarity(tfidf_matrix_candidate1)
+            cosine_sim_df_candidate1 = pd.DataFrame(cosine_sim_matrix_candidate1)
+
+            st.subheader("Measuring similarity between keywords of candidate profile 1 and job description")
+            fig_similarity_candidate1 = px.imshow(cosine_sim_df_candidate1, text_auto=True,
+                                                 labels=dict(x="Keyword similarity", y="Resumes", color="Productivity"),
+                                                 x=['Resume 1', 'Jon Description'],
+                                                 y=['Resume 1', 'Job Description'])
+            st.plotly_chart(fig_similarity_candidate1, key="bar_candidate1")
+
+            for i, similarity_score in enumerate(cosine_sim_matrix_candidate1[0][1:]):
+                st.write(f"Similarity of job description with candidate profile 1. {i + 1}: {similarity_score:.4f}")
+                st.info("A score closer to 1 (0.80, 0.90) means higher similarity between candidate profile 1 and job description. A score closer to 0 (0.20, 0.30) means lower similarity between candidate profile 1 and job description.")
 else:
-    st.warning(f"You have reached the maximum URL attempts ({max_attempts}).")
-    if 'upload_count' in st.session_state and st.session_state['upload_count'] > 0:
-        st.info(f"Files uploaded {st.session_state['upload_count']} time(s).")
-
-
-
-
-
-st.subheader ("Candidate Profile 2", divider = "green")
-
-
-
-if 'upload_count' not in st.session_state:
-    st.session_state['upload_count'] = 0
-
-max_attempts = 3
-
-if st.session_state['upload_count'] < max_attempts:
-    uploaded_files = st.file_uploader(
-        "Upload your resume in .pdf format", accept_multiple_files=True, type="pdf", key="candidate 2"
+    st.warning(f"You have reached the maximum upload attempts ({max_attempts_candidate1}) for Candidate Profile 1.")
+    if 'upload_count_candidate1' in st.session_state and st.session_state['upload_count_candidate1'] > 0:
+        st.info(f"Files uploaded for Candidate Profile 1: {st.session_state['upload_count_candidate1']} time(s).")
+
+st.subheader("Candidate Profile 2", divider="green")
+if 'upload_count_candidate2' not in st.session_state:
+    st.session_state['upload_count_candidate2'] = 0
+max_attempts_candidate2 = 3
+if st.session_state['upload_count_candidate2'] < max_attempts_candidate2:
+    uploaded_files_candidate2 = st.file_uploader(
+        "Upload your resume in .pdf format", accept_multiple_files=True, type="pdf", key="candidate_2_upload"
     )
-
-if uploaded_files:
-    st.session_state['upload_count'] += 1
-for uploaded_file in uploaded_files:
-    pdf_reader = PdfReader(uploaded_file)
-    text_data = ""
-    for page in pdf_reader.pages:
-        text_data += page.extract_text()
-        data = pd.Series(text_data, name = 'Text')
-        
-        
-        frames = [job, data]
-        result = pd.concat(frames)
-        
-
-        model = GLiNER.from_pretrained("urchade/gliner_base")
-        labels = ["person", "country","organization", "date", "time", "role", "skills", "year"]
-        entities = model.predict_entities(text_data, labels)
-        df = pd.DataFrame(entities)
-        
-
-        
-        fig = px.treemap(entities, path=[px.Constant("all"), 'text', 'label'],
-                values='score', color='label')
-        fig.update_layout(margin = dict(t=50, l=25, r=25, b=25))
-        st.plotly_chart(fig, key = "iris")
-        
-       
-            
-        
-        vectorizer = TfidfVectorizer()
-        tfidf_matrix = vectorizer.fit_transform(result)
-        tfidf_df = pd.DataFrame(tfidf_matrix.toarray(), columns=vectorizer.get_feature_names_out())
-    
-        cosine_sim_matrix = cosine_similarity(tfidf_matrix)
-        cosine_sim_df = pd.DataFrame(cosine_sim_matrix)
-        
-
-        
-        
-        fig = px.imshow(cosine_sim_df, text_auto=True, labels=dict(x="Keyword similarity", y="Resumes", color="Productivity"),
-                        x=['Resume 2', 'Jon Description'],
-                        y=['Resume 2', 'Job Description'])
-        st.plotly_chart(fig, key = "radio")
-        
-
-        
-        for i, similarity_score in enumerate(cosine_sim_matrix[0][1:]):
-            st.write(f"Similarity with Candidate Profile. A score closer to 1 means higher similarity. {i + 1}: {similarity_score:.4f}")
-
+    if uploaded_files_candidate2:
+        st.session_state['upload_count_candidate2'] += 1
+        for uploaded_file in uploaded_files_candidate2:
+            pdf_reader = PdfReader(uploaded_file)
+            text_data = ""
+            for page in pdf_reader.pages:
+                text_data += page.extract_text()
+            data_candidate2 = pd.Series(text_data, name='Text')
+            frames_candidate2 = [job, data_candidate2]
+            result_candidate2 = pd.concat(frames_candidate2)
+            model_candidate2 = GLiNER.from_pretrained("urchade/gliner_base")
+            labels_candidate2 = ["person", "country", "organization", "date", "time", "role", "skills", "year"]
+            entities_candidate2 = model_candidate2(text_data, labels=labels_candidate2)
+            df_candidate2 = pd.DataFrame(entities_candidate2)
+
+            st.title("Profile of candidate 2")
+            fig_entities_candidate2 = px.treemap(entities_candidate2, path=[px.Constant("all"), 'text', 'label'],
+                                                values='score', color='label')
+            fig_entities_candidate2.update_layout(margin=dict(t=50, l=25, r=25, b=25))
+            st.plotly_chart(fig_entities_candidate2, key="iris_candidate2")
+
+            vectorizer_candidate2 = TfidfVectorizer()
+            tfidf_matrix_candidate2 = vectorizer_candidate2.fit_transform(result_candidate2)
+            tfidf_df_candidate2 = pd.DataFrame(tfidf_matrix_candidate2.toarray(),
+                                               columns=vectorizer_candidate2.get_feature_names_out())
+            cosine_sim_matrix_candidate2 = cosine_similarity(tfidf_matrix_candidate2)
+            cosine_sim_df_candidate2 = pd.DataFrame(cosine_sim_matrix_candidate2)
+
+            st.subheader("Measuring similarity between keywords of candidate profile 2 and job description")
+            fig_similarity_candidate2 = px.imshow(cosine_sim_df_candidate2, text_auto=True,
+                                                 labels=dict(x="Keyword similarity", y="Resumes", color="Productivity"),
+                                                 x=['Resume 2', 'Jon Description'],
+                                                 y=['Resume 2', 'Job Description'])
+            st.plotly_chart(fig_similarity_candidate2, key="radio_candidate2")
+
+            for i, similarity_score in enumerate(cosine_sim_matrix_candidate2[0][1:]):
+                st.write(f"Similarity with Candidate Profile 2. A score closer to 1 means higher similarity. {i + 1}: {similarity_score:.4f}")
 else:
-    st.warning(f"You have reached the maximum URL attempts ({max_attempts}).")
-    if 'upload_count' in st.session_state and st.session_state['upload_count'] > 0:
-        st.info(f"Files uploaded {st.session_state['upload_count']} time(s).")
-
-
-
+    st.warning(f"You have reached the maximum upload attempts ({max_attempts_candidate2}) for Candidate Profile 2.")
+    if 'upload_count_candidate2' in st.session_state and st.session_state['upload_count_candidate2'] > 0:
+        st.info(f"Files uploaded for Candidate Profile 2: {st.session_state['upload_count_candidate2']} time(s).")