updated

pages/1_📊_About.py

@@ -10,7 +10,7 @@ with tab1:
     st.header("Product Discovery")
     with st.expander("Details"):
         st.markdown("**Application:** Gofynd.com")
-        st.markdown("**Total Products:** 40,000")
+        st.markdown("**Total Products:** 14,000")
         st.markdown("**Data Used:**")
         st.markdown("- Text fields - Name, Brand, Category, Gender")
         st.markdown("- Images - Single Images (First image of the product)")
@@ -21,14 +21,25 @@ with tab1:
         st.markdown("Supported Input data format:")
         st.markdown("Similarity Metric: Inner Product")
         st.markdown("Type of Index: FLAT_L2")
-        st.subheader("Input Sample Data Format:")
-        # Insert your input sample data here
-        1. Pull the Catalog data from Fynd Platform
-        2. Parse Textual and Image data from Catalog data
-        3. Generate Embeddings for Textual fields
-        4. Generate Embeddings for Image 
-        5. Aggregate Embeddings with help of defined weightage
-        6. Store it to Vector DB - Milvus
+        
+        st.subheader("Input Sample Data:")
+        st.markdown("Sample data can be displayed here.")
+        
+        st.subheader("STEPS")
+        st.markdown("**TRAINING:**")
+        st.markdown("1. Pull the Catalog data from Fynd Platform")
+        st.markdown("2. Parse Textual and Image data from Catalog data")
+        st.markdown("3. Generate Embeddings for Textual fields")
+        st.markdown("4. Generate Embeddings for Image")
+        st.markdown("5. Aggregate Embeddings with help of defined weightage")
+        st.markdown("6. Store it to Vector DB - Milvus")
+        
+        st.markdown("**SERVING:**")
+        st.markdown("1. Convert the Users’s query as Input text string to Embeddings with CLIP model")
+        st.markdown("2. Pass the user’s search query to LLM - OpenAI’S GPT3 to detect the search entities - Query, Brand, Category, Price, Sorting type")
+        st.markdown("3. Prepare and Apply Search Query on Collection of Milvus DB on which the data is indexed")
+        st.markdown("4. Fetch top N results")
+        st.markdown("5. Apply filters and return the results")
 
 with tab2:
     st.header("Question And Answering")