Update src/streamlit_app.py

src/streamlit_app.py

@@ -22,28 +22,63 @@ st.set_page_config(
 )
 
 # Enhanced data loading with error handling
+# Upload section in sidebar
+st.sidebar.header("📤 Upload Excel Files")
+uploaded_file1 = st.sidebar.file_uploader("Upload File 1 (e.g. BITS_INTERNS)", type=["xlsx"])
+uploaded_file2 = st.sidebar.file_uploader("Upload File 2 (e.g. ICFAI Responses)", type=["xlsx"])
+
 @st.cache_data
-def load_data():
+
+def load_uploaded_data(upload1, upload2):
     try:
-        # Use actual file paths
-        df1 = pd.read_excel("src/BITS_INTERNS.xlsx", sheet_name="Form Responses 1")
-        df2 = pd.read_excel("/mnt/data/Summer of AI - ICFAI  (Responses) (3).xlsx")
+        if not upload1 or not upload2:
+            return None, "Please upload both Excel files."
+
+        df1 = pd.read_excel(upload1)
+        df2 = pd.read_excel(upload2)
 
         df1.columns = df1.columns.str.strip()
         df2.columns = df2.columns.str.strip()
-        
-        # Find common columns and combine
+
+        # Use only common columns
         common_cols = list(set(df1.columns) & set(df2.columns))
         df1 = df1[common_cols]
         df2 = df2[common_cols]
 
         combined_df = pd.concat([df1, df2], ignore_index=True)
         return combined_df, None
-    except FileNotFoundError as e:
-        return None, f"File not found: {e}"
     except Exception as e:
-        return None, f"Error loading data: {str(e)}"
+        return None, str(e)
+
+def extract_numeric_ram(ram) -> Optional[int]:
+    if pd.isna(ram):
+        return None
+    ram_str = str(ram).lower().replace(" ", "")
+    gb_match = re.search(r"(\d+(?:\.\d+)?)(?:gb|g)", ram_str)
+    if gb_match:
+        return int(float(gb_match.group(1)))
+    mb_match = re.search(r"(\d+)(?:mb|m)", ram_str)
+    if mb_match:
+        return max(1, int(int(mb_match.group(1)) / 1024))
+    plain_match = re.search(r"(\d+)", ram_str)
+    if plain_match:
+        return int(plain_match.group(1))
+    return None
+
+st.title("🧠 LLM Compatibility Advisor")
+st.markdown("Get personalized recommendations from **150+ popular open source AI models** with download sizes!")
+
+df, error = load_uploaded_data(uploaded_file1, uploaded_file2)
+
+if error:
+    st.error(error)
+    st.stop()
+
+if df is None or df.empty:
+    st.warning("No data loaded. Please upload both files.")
+    st.stop()
 
+st.success(f"✅ Loaded {len(df)} student entries from both files.")
 
 # Enhanced RAM extraction with better parsing
 def extract_numeric_ram(ram) -> Optional[int]: