Update src/streamlit_app.py

src/streamlit_app.py

@@ -1,7 +1,7 @@
 """
 ================================================================================
   SimCLR ResNet-50 Visual Search Engine GUI
-  -----------------------------------------
+--------------------------------------------------------------------------------
   A premium, interactive web-based graphical user interface (GUI) built with
   Streamlit, powered by ONNX Runtime and FAISS for real-time visual similarity
   retrieval on CIFAR-10.
@@ -15,10 +15,9 @@
     6. Dedicated interactive "Ablation Study Dashboard" showing the Exp 38-42 findings
 
   Usage:
-    streamlit run app.py
+    streamlit run src/streamlit_app.py
 ================================================================================
 """
-
 import os
 import sys
 import time
@@ -26,7 +25,6 @@ import json
 import random
 import numpy as np
 from PIL import Image
-
 import streamlit as st
 
 # Windows console encoding fix
@@ -34,14 +32,16 @@ sys.stdout.reconfigure(encoding='utf-8')
 sys.stderr.reconfigure(encoding='utf-8')
 
 # Ensure we can import from src/
-sys.path.append(os.path.dirname(os.path.abspath(__file__)))
-sys.path.append(os.path.join(os.path.dirname(os.path.abspath(__file__)), "src"))
+SRC_DIR = os.path.dirname(os.path.abspath(__file__))
+sys.path.append(SRC_DIR)
 
 # ==========================================================================
 # Caching Assets for Instant Performance
 # ==========================================================================
-BASE_DIR = os.path.dirname(os.path.abspath(__file__))
-DEPLOY_DIR = os.path.join(BASE_DIR, "deployment")
+# Navigate up one level from 'src' to the root directory to access 'deployment'
+ROOT_DIR = os.path.dirname(SRC_DIR)
+DEPLOY_DIR = os.path.join(ROOT_DIR, "deployment")
+
 ONNX_PATH = os.path.join(DEPLOY_DIR, "simclr_encoder_exp41.onnx")
 INDEX_PATH = os.path.join(DEPLOY_DIR, "cifar10_index.faiss")
 META_PATH = os.path.join(DEPLOY_DIR, "metadata.json")
@@ -49,7 +49,6 @@ META_PATH = os.path.join(DEPLOY_DIR, "metadata.json")
 CIFAR_MEAN = np.array([0.4914, 0.4822, 0.4465], dtype=np.float32)
 CIFAR_STD = np.array([0.2023, 0.1994, 0.2010], dtype=np.float32)
 
-
 @st.cache_resource
 def load_onnx_model():
     """Load ONNX inference session and cache it."""
@@ -58,21 +57,18 @@ def load_onnx_model():
     providers = ['CPUExecutionProvider']
     return ort.InferenceSession(ONNX_PATH, providers=providers)
 
-
 @st.cache_resource
 def load_faiss_index():
     """Load FAISS index and cache it."""
     import faiss
     return faiss.read_index(INDEX_PATH)
 
-
 @st.cache_data
 def load_metadata():
     """Load metadata dictionary and cache it."""
     with open(META_PATH, "r", encoding="utf-8") as f:
         return json.load(f)
 
-
 # -- Check that deployment files exist ---------------------------------------
 missing_files = []
 for p, name in [(ONNX_PATH, "ONNX Model"), (INDEX_PATH, "FAISS Index"), (META_PATH, "Metadata JSON")]:
@@ -81,7 +77,7 @@ for p, name in [(ONNX_PATH, "ONNX Model"), (INDEX_PATH, "FAISS Index"), (META_PA
 
 if missing_files:
     st.error(f" Missing deployment assets: {', '.join(missing_files)}")
-    st.warning(" Please run `export_onnx.py` then `build_faiss.py` to generate these files before running the GUI!")
+    st.warning(" Please ensure the 'deployment' folder contains the required ONNX, FAISS, and JSON files before running the GUI!")
     st.stop()
 
 # Load cached assets
@@ -118,7 +114,6 @@ def preprocess_image(pil_img):
     
     return img_batch
 
-
 def perform_search(features_batch, top_k=5):
     """
     Normalizes features, queries FAISS, and returns matched metadata.
@@ -149,7 +144,6 @@ def perform_search(features_batch, top_k=5):
         
     return results
 
-
 # ==========================================================================
 # Streamlit UI Configuration
 # ==========================================================================
@@ -230,7 +224,6 @@ st.sidebar.markdown("""
 * **Similarity Metric**: Exact Cosine Similarity
 * **Database Size**: 10,000 Images
 """)
-
 st.sidebar.caption("Group 20, CISC 867, Queen's University, Spring 2026")
 
 # ==========================================================================
@@ -240,7 +233,7 @@ st.title("Real-time SimCLR Image Retrieval Engine")
 st.markdown("##### Self-Supervised Representation Learning with ResNet-50 & FAISS Indexing")
 
 # Define Tabs
-tab1, tab2 = st.tabs([" Real-Time Search", " Ablation Study Dashboard"])
+tab1, tab2 = st.tabs(["🔍 Real-Time Search", "📊 Ablation Study Dashboard"])
 
 # ==========================================================================
 # TAB 1: Visual Search Engine
@@ -256,7 +249,7 @@ with tab1:
     query_info = None
 
     with col_left:
-        st.markdown("###  Select Query Image")
+        st.markdown("### 📤 Select Query Image")
         
         # Method 1: Upload a file
         uploaded_file = st.file_uploader("Upload an image...", type=["jpg", "png", "jpeg"])
@@ -289,7 +282,7 @@ with tab1:
                 st.error("Reference image file not found.")
 
     with col_right:
-        st.markdown("###Visual Similarity Search Results")
+        st.markdown("### 🔎 Visual Similarity Search Results")
         
         if query_image is not None:
             t_start = time.time()
@@ -324,7 +317,7 @@ with tab1:
                     st.markdown(f'<div class="glass-card"><div class="metric-value">N/A</div><div class="metric-label">Upload class unknown</div></div>', unsafe_allow_html=True)
             
             # Display Results Grid
-            st.markdown("#### Retreived Nearest Neighbors")
+            st.markdown("#### Retrieved Nearest Neighbors")
             
             # Create dynamic grid columns
             grid_cols = st.columns(3)
@@ -371,13 +364,13 @@ with tab1:
                     col.markdown(card_content, unsafe_allow_html=True)
                     
         else:
-            st.info(" Please upload a custom image or click the button to select a random test image to query the visual search engine!")
+            st.info("💡 Please upload a custom image or click the button to select a random test image to query the visual search engine!")
 
 # ==========================================================================
 # TAB 2: Ablation Study Dashboard
 # ==========================================================================
 with tab2:
-    st.markdown("###  Experiment Ablation Study & Color Jitter Findings")
+    st.markdown("### 🧪 Experiment Ablation Study & Color Jitter Findings")
     st.markdown("##### Evaluating Natalie's Midterm Pipelines against Mahmoud's Final Jittered Re-runs")
     
     st.markdown("""
@@ -412,7 +405,7 @@ with tab2:
             st.markdown(card_html, unsafe_allow_html=True)
             
     st.markdown("---")
-    st.markdown("####  Key Project Takeaways")
+    st.markdown("#### 💡 Key Project Takeaways")
     col1, col2 = st.columns(2)
     
     with col1:
@@ -427,10 +420,10 @@ with tab2:
     with col2:
         st.markdown("""
         ##### 3. Model Architecture & Stem Tuning ⚙️
-        Modifying the standard ResNet-50 conv1 stem from $7\\times7$ (stride 2) to a custom $3\\times3$ (stride 1) and removing the initial MaxPool was crucial to preserve the resolution of $32\\times32$ CIFAR-10 images. 
+        Modifying the standard ResNet-50 conv1 stem from 7x7 (stride 2) to a custom 3x3 (stride 1) and removing the initial MaxPool was crucial to preserve the resolution of 32x32 CIFAR-10 images. 
         
         ##### 4. Near Foundation-Model Upper Bound 🏆
         Our zero-shot CLIP ViT-B/32 foundation model evaluation sets the academic upper bound at **88.80%**. Our custom-trained SimCLR ResNet-50 achieves **95% of this performance** (**84.30%**) while using **8,000x less training data**!
         """)
 
-    st.success("🎉 Weights & Biases Dashboard has archived all 5 experiments complete with training curves, checkpoints, and t-SNE files.")
+    st.success("🎉 Weights & Biases Dashboard has archived all 5 experiments complete with training curves, checkpoints, and t-SNE files.")