Updated UI, files and resources

src/config.py

@@ -7,8 +7,8 @@ BASE_DIR = os.path.dirname(os.path.abspath(__file__))
 DATA_DIR = os.path.join(BASE_DIR, 'data')
 MODELS_DIR = os.path.join(BASE_DIR, 'models')
 
-TRAIN_IMAGES_DIR = os.path.join(DATA_DIR, 'images')
-TEST_IMAGES_DIR = os.path.join(DATA_DIR, 'images')
+TRAIN_IMAGES_DIR = os.path.join(DATA_DIR, 'images','train')
+TEST_IMAGES_DIR = os.path.join(DATA_DIR, 'images','test')
 
 TRAIN_CSV_PATH = os.path.join(DATA_DIR, 'train.csv')
 TEST_CSV_PATH = os.path.join(DATA_DIR, 'test.csv')
@@ -35,7 +35,7 @@ print(f"Blank Symbol: '{BLANK_TOKEN_SYMBOL}' at index {BLANK_TOKEN}")
 
 # --- Image Preprocessing Parameters ---
 IMG_HEIGHT = 32 # Target height for all input images to the model
-MAX_IMG_WIDTH = 1024 # Adjust this value based on your typical image widths and available RAM
+MAX_IMG_WIDTH = 720 # Adjust this value based on your typical image widths and available RAM
 
 # --- Training Parameters ---
 BATCH_SIZE = 10

src/models/handwritten_name_ocr_model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8625e259aa257335a787b54bbc9325fdf4fe46d6b67debb4f50665fe2c47ad8
+size 21382549

src/streamlit_app.py

@@ -79,7 +79,7 @@ with tab1:
         st.markdown("""
             **[📃 Project Documentation ](https://drive.google.com/file/d/1HBrQT_UnzNLdEsouW9wMk4alAeCsQxZb/view?usp=sharing)**
                         
-            **[🎞️ Demo Presentation ](https://drive.google.com/drive/folders/1rOmwyTJkDCsU-Wuh-_CzvQ9sdb_ci_kX?usp=sharing)**
+            **[🎞️ Demo Presentation ](https://drive.google.com/file/d/1j_S8cijxy6zxIn3cWg6tuLPNWB_7nwdI/view?usp=sharing)**
                         
             **[📚 Dataset (from Kaggle)](https://www.kaggle.com/datasets/landlord/handwriting-recognition)**
             
@@ -88,43 +88,41 @@ with tab1:
 
 # --- Tab 2: Predict Name (Main Content: Prediction Section) ---
 with tab2:
-        st.header("Predict on a New Image")
-        st.markdown("Upload a clear image of a single handwritten name or word for recognition.")
+    st.markdown("Upload a clear image of a single handwritten name or word for recognition.")
 
-        if ocr_model is None:
-            st.warning("Model not loaded. Please train or load a model in the 'Train & Evaluate' tab before attempting prediction.")
+    uploaded_file = st.file_uploader("🖼️ Choose an image...", type=["png", "jpg", "jpeg", "jfif"])
+    if uploaded_file is not None:
+        try:
+            image_pil = Image.open(uploaded_file).convert('L')
+            st.image(image_pil, caption="Uploaded Image", use_container_width=True) 
+            st.write("---")
+            st.write("Processing and Recognizing...")
+
+            processed_image_tensor = preprocess_user_image_for_ocr(image_pil, IMG_HEIGHT).to(device)
+            
+            ocr_model.eval() # Ensure model is in eval mode for prediction
+            with torch.no_grad():
+                output = ocr_model(processed_image_tensor)
+            
+            predicted_texts = ctc_greedy_decode(output, char_indexer)
+            predicted_text = predicted_texts[0]
+
+            st.success(f"Recognized Text: **{predicted_text}**")
+
+        except Exception as e:
+            st.error(f"Error processing image or recognizing text: {e}")
+            st.info("💡 **Tips for best results:**\n"
+                    "- Ensure the handwritten text is clear and on a clean background.\n"
+                    "- Only include one name/word per image.\n"
+                    "- The model is trained on specific characters. Unusual symbols might not be recognized.")
+            st.exception(e)
+                
         else:
-            uploaded_file = st.file_uploader("🖼️ Choose an image...", type=["png", "jpg", "jpeg", "jfif"])
-
-            if uploaded_file is not None:
-                try:
-                    image_pil = Image.open(uploaded_file).convert('L')
-                    st.image(image_pil, caption="Uploaded Image", use_container_width=True) 
-                    st.write("---")
-                    st.write("Processing and Recognizing...")
-
-                    processed_image_tensor = preprocess_user_image_for_ocr(image_pil, IMG_HEIGHT).to(device)
-                    
-                    ocr_model.eval() # Ensure model is in eval mode for prediction
-                    with torch.no_grad():
-                        output = ocr_model(processed_image_tensor)
-                    
-                    predicted_texts = ctc_greedy_decode(output, char_indexer)
-                    predicted_text = predicted_texts[0]
-
-                    st.success(f"Recognized Text: **{predicted_text}**")
-
-                except Exception as e:
-                    st.error(f"Error processing image or recognizing text: {e}")
-                    st.info("💡 **Tips for best results:**\n"
-                            "- Ensure the handwritten text is clear and on a clean background.\n"
-                            "- Only include one name/word per image.\n"
-                            "- The model is trained on specific characters. Unusual symbols might not be recognized.")
-                    st.exception(e)
+            st.warning("Model not loaded. Please train or load a model in the 'Train & Evaluate' tab before attempting prediction.")
 
 # --- Tab 3: Train & Evaluate ---
 with tab3:
-    st.header("Model Training and Evaluation")
+    st.subheader("Model Training and Evaluation")
     st.markdown("Here you can train a new OCR model or load a pre-trained one.")
 
         # --- Model Loading / Initialization (Cached) ---
@@ -272,7 +270,7 @@ with tab3:
 
     # --- Training History Plots Section ---
     st.subheader("Training History Plots")
-    if st.session_state.training_history: # Check if history exists in session state
+    if st.session_state.training_history:
         history_df = pd.DataFrame({
             'Epoch': range(1, len(st.session_state.training_history['train_loss']) + 1),
             'Train Loss': st.session_state.training_history['train_loss'],