upload-v1.5.12

recommend.py

@@ -1,49 +1,21 @@
-# Streamlit app script
-import streamlit as st
-from recommend import recommend
-# A simple function to check login credentials (for demonstration purposes)
-def check_login(username, password):
-    # Hardcoding a simple example username and password
-    user = "admin"
-    pwd = "pass123"
-    return username == user and password == pwd
-
-# Main application code
-def main():
-    # Initialize session state for login status
-    if "logged_in" not in st.session_state:
-        st.session_state.logged_in = False
-
-    # If not logged in, display login form
-    if not st.session_state.logged_in:
-        st.title("Login Page")
-        username = st.text_input("Username")
-        password = st.text_input("Password", type="password")
-        if st.button("Login"):
-            if check_login(username, password):
-                # Update session state to indicate user is logged in
-                # st.session_state.username = username
-                st.session_state.logged_in = True
-                st.rerun()  # Rerun the script to reflect the new state
-            else:
-                st.error("Invalid credentials. Please try again.")
-
-    # If logged in, redirect to another page or show different content
-    else:
-        # This can be another Streamlit page, or a condition to render a different view
-        st.title(f"Welcome :)!")
-        cols = st.columns([3,1])
-        with cols[0]:
-            query = st.text_input('Search here', placeholder="Describe what you're looking for", label_visibility="collapsed")
-        with cols[1]:
-            btn = st.button('Search')
-        if btn and query:
-            with st.spinner('Searching...'):
-                st.write_stream(recommend(query))
-        # Example: Provide a logout button
-        if st.sidebar.button("Logout"):
-            st.session_state.logged_in = False
-            st.rerun()
-
-if __name__ == "__main__":
-    main()
+from sentence_transformers import SentenceTransformer
+from scipy.spatial.distance import cosine
+import numpy as np
+import pandas as pd
+from datasets import load_dataset
+import pickle as pkl
+def recommend(query, n=5):
+    # Load the model
+    model = SentenceTransformer('all-MiniLM-L6-v2', device='cpu')
+    # Load the data
+    # data = pd.read_csv('data/medium_articles.csv')
+    data = load_dataset('Mohamed-BC/Articles')['train'].to_pandas()
+    # get the embeddings
+    a_embeddings = pkl.load(open('data/articles_embeddings.pkl', 'rb'))
+    # Encode the query
+    q_embedding = model.encode(query)
+    # Calculate the cosine similarity
+    cos_sim = np.array([1 - cosine(q_embedding, emb) for emb in a_embeddings[:1000]])
+    # Get the top n recommendations
+    top_n = np.argsort(cos_sim)[-n:]
+    return data.iloc[top_n]['title']