Delete pages/model1.py

pages/model1.py

@@ -1,108 +0,0 @@
-import os
-import zipfile
-import tempfile
-import fitz 
-import streamlit as st
-from langchain_huggingface import HuggingFaceEndpoint, ChatHuggingFace
-
-os.environ["HF_TOKEN"]=os.getenv('HF_Token')
-os.environ["HUGGINGFACEHUB_API_KEY"]=os.getenv('HF_Token')
-
-llm_base = HuggingFaceEndpoint(
-    repo_id="meta-llama/Llama-3.1-8B-Instruct",
-    provider="novita",
-    temperature=0.7,
-    max_new_tokens=150,
-    task="conversational"
-)
-
-llm_chat = ChatHuggingFace(
-    llm=llm_base,
-    repo_id="meta-llama/Llama-3.2-3B-Instruct",
-    provider="novita",
-    temperature=0.7,
-    max_new_tokens=150,
-    task="conversational"
-)
-
-
-def extract_pdf_text(file_bytes):
-    with tempfile.NamedTemporaryFile(delete=False, suffix=".pdf") as tmp:
-        tmp.write(file_bytes)
-        with fitz.open(tmp.name) as doc:
-            return "\n".join(page.get_text() for page in doc)
-            
-def resume_extraction_prompt(text):
-    return (
-        "You are a resume parser. Extract the following details:\n"
-        "- Full Name\n- Education\n- Work Experience\n- Skills\n- Project Titles and Results\n\n"
-        f"Resume Content:\n{text}"
-    )
-
-def jd_extraction_prompt(text):
-    return (
-        "You are an expert recruiter. Extract this info from the job description:\n"
-        "- Job ID\n- Company\n- Role\n- Required Experience\n- Required Skills\n- Education\n- Location\n\n"
-        f"JD Content:\n{text}"
-    )
-
-def resume_matching_prompt(jd, resumes):
-    return (
-        "You are an AI hiring assistant. Based on the job description and candidate summaries,"
-        " identify the best match considering skills, experience, and education.\n\n"
-        f"Job Description:\n{jd}\n\n"
-        f"Resumes:\n{resumes}\n\n"
-        "Return the name of the best candidate and a short justification."
-    )
-
-st.set_page_config(page_title="TalentMatch AI", layout="centered")
-st.title("🧠 TalentMatch AI: Resume & JD Matcher")
-
-with st.expander("📘 About this App"):
-    st.markdown("""
-This AI-powered app extracts key details from resumes and job descriptions to intelligently
-match the most suitable candidate. Upload a ZIP of PDF resumes and a job description (or paste text)
-to get started!
-""")
-
-zip_upload = st.file_uploader("📎 Upload ZIP of PDF Resumes", type=["zip"])
-jd_upload = st.file_uploader("📝 Upload Job Description (PDF/TXT)", type=["pdf", "txt"])
-jd_text = st.text_area("✍️ Or Paste Job Description Text")
-
-# Match Button
-if st.button("🚀 Match Best Candidate"):
-    if not zip_upload or not (jd_upload or jd_text.strip()):
-        st.warning("⚠ Please upload both resumes and a job description.")
-    else:
-        # Extract JD text
-        jd_content = jd_text.strip()
-        if jd_upload:
-            jd_content = (
-                extract_pdf_text(jd_upload.read()) if jd_upload.name.endswith(".pdf")
-                else jd_upload.read().decode("utf-8")
-            )
-
-        if len(jd_content.split()) < 30:
-            st.warning("📄 JD text seems too short. Please provide a more detailed description.")
-        else:
-            jd_info = llm_chat.invoke(jd_extraction_prompt(jd_content)).content
-            st.subheader("🔍 Extracted Job Details")
-            st.markdown(jd_info)
-
-        resume_summaries = ""
-        with tempfile.TemporaryDirectory() as tmpdir:
-            with zipfile.ZipFile(zip_upload, "r") as zip_ref:
-                pdfs = [f for f in zip_ref.namelist() if f.endswith(".pdf")]
-                if not pdfs:
-                    st.error("No PDF resumes found in the ZIP.")
-                    st.stop()
-                for pdf in pdfs:
-                    with zip_ref.open(pdf) as f:
-                        text = extract_pdf_text(f.read())
-                        summary = llm_chat.invoke(resume_extraction_prompt(text)).content
-                        resume_summaries += f"\n\n📄 {pdf}\n{summary}"
-
-        with st.spinner("🧮 Evaluating best match..."):
-            match = llm_chat.invoke(resume_matching_prompt(jd_content, resume_summaries)).content
-            st.subheader("✅ Top Candidate Match")
-            st.markdown(match)