Update app.py

app.py

@@ -1,45 +1,20 @@
 import streamlit as st
-from transformers import pipeline
 from sentence_transformers import SentenceTransformer, util
-import docx2txt
-import PyPDF2
 
-@st.cache_resource
+@st.cache_resource  # Cache the model for faster loading
 def load_model():
-    return SentenceTransformer('all-mpnet-base-v2')
+    return SentenceTransformer('sentence-transformers/all-mpnet-base-v2')
 
 model = load_model()
 
-def extract_text_from_file(file):
-    if file.type == "application/pdf":
-        with open(file, "rb") as f:
-            pdf_reader = PyPDF2.PdfReader(f)
-            text = "".join([page.extract_text() for page in pdf_reader.pages])
-    elif file.type == "application/vnd.openxmlformats-officedocument.wordprocessingml.document":
-        text = docx2txt.process(file)
-    else:
-        return None
-    return text
+st.title("Text Similarity Checker")
 
-st.title("Resume/Job Description Similarity Checker")
+text1 = st.text_input("Enter the first text:")
+text2 = st.text_input("Enter the second text:")
 
-job_description = st.text_area("Paste the job description here:")
-resume_file = st.file_uploader("Upload your resume (PDF or DOCX)", type=["pdf", "docx"])
-
-if job_description and resume_file:
-    resume_text = extract_text_from_file(resume_file)
-    if resume_text:
-        job_description_embedding = model.encode(job_description)
-        resume_embedding = model.encode(resume_text)
-
-        similarity_score = util.cos_sim(job_description_embedding, resume_embedding).item() * 100
-        st.write(f"**Similarity Score:** {similarity_score:.2f}%")
-
-        if similarity_score > 80:
-            st.success("Your resume seems to be a strong match for the job description!")
-        elif similarity_score > 50:
-            st.warning("Your resume shows some alignment, but consider enhancing it further to better reflect the job requirements.")
-        else:
-            st.error("Your resume might not align well with the job description. Consider revising it based on the listed requirements.")
-    else:
-        st.error("Unsupported file type. Please upload a PDF or DOCX file.")
+if text1 and text2:
+    # Calculate embeddings and similarity
+    embedding1 = model.encode(text1)
+    embedding2 = model.encode(text2)
+    similarity = util.cos_sim(embedding1, embedding2).item() * 100
+    st.write(f"**Similarity Score:** {similarity:.2f}%")