REvert "Add Model Evaluation Notebook"

pages/Model_Evaluation.py

@@ -15,16 +15,14 @@ from sklearn.preprocessing import label_binarize
 import streamlit as st
 import matplotlib.pyplot as plt
 from fpdf import FPDF
-from datasets import load_dataset
-from huggingface_hub import hf_hub_download 
-import requests
-from io import BytesIO
 
 # ---- Streamlit State Initialization ----
 if 'stop_eval' not in st.session_state:
     st.session_state.stop_eval = False
+
 if 'evaluation_done' not in st.session_state:
     st.session_state.evaluation_done = False
+
 if 'trigger_eval' not in st.session_state:
     st.session_state.trigger_eval = False
 
@@ -35,7 +33,7 @@ st.markdown("<h2 style='color: #2E86C1;'>📈 Model Evaluation</h2>", unsafe_all
 class_names = ['No_DR', 'Mild', 'Moderate', 'Severe', 'Proliferative_DR']
 label_map = {label: idx for idx, label in enumerate(class_names)}
 
-# ---- Text Cleaning Function ----
+# ---- Text Cleaning Function for PDF ----
 def clean_text(text):
     return text.encode('utf-8', 'ignore').decode('utf-8')
 
@@ -60,104 +58,70 @@ def apply_gaussian_filter(image, kernel_size=(5, 5), sigma=1.0):
     return cv2.GaussianBlur(image, kernel_size, sigma)
 
 # ---- Custom Dataset ----
-def __getitem__(self, idx):
-    # Get the image URL and label
-    raw_url = self.image_paths[idx]
-    label = int(self.labels[idx])
-
-    # Fix the URL if it's in blob format
-    img_url = raw_url.replace("/blob/", "/resolve/")
-
-    try:
-        print(f"Downloading: {img_url}")
-        response = requests.get(img_url, stream=True)
-        if response.status_code != 200:
-            raise ValueError(f"Failed to download image from {img_url} - status code: {response.status_code}")
-
-        # Read and convert the image
-        image = Image.open(BytesIO(response.content)).convert("RGB")
-    except Exception as e:
-        raise RuntimeError(f"Error loading image from {img_url}: {e}")
-
-    image = np.array(image)
-
-    image = apply_median_filter(image)
-    image = apply_clahe(image)
-    image = apply_gamma_correction(image)
-    image = apply_gaussian_filter(image)
-
-    image = Image.fromarray(image)
-    if self.transform:
-        image = self.transform(image)
-
-    return image, torch.tensor(label, dtype=torch.long)
-    
+class DDRDataset(Dataset):
+    def __init__(self, csv_path, transform=None):
+        self.data = pd.read_csv(csv_path)
+        self.image_paths = self.data['new_path'].tolist()
+        self.labels = self.data['label'].tolist()
+        self.transform = transform
+
+    def __len__(self):
+        return len(self.image_paths)
+
+    def __getitem__(self, idx):
+        img_path = self.image_paths[idx]
+        label = int(self.labels[idx])
+
+        image = cv2.imread(img_path)
+        image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
+
+        # Apply preprocessing
+        image = apply_median_filter(image)
+        image = apply_clahe(image)
+        image = apply_gamma_correction(image)
+        image = apply_gaussian_filter(image)
+
+        image = Image.fromarray(image)
+        if self.transform:
+            image = self.transform(image)
+
+        return image, torch.tensor(label, dtype=torch.long)
+
 # ---- Image Transforms ----
 val_transform = transforms.Compose([
     transforms.Resize((224, 224)),
     transforms.ToTensor(),
     transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
 ])
-    
-# ---- DDRDataset Class Definition ----
-class DDRDataset(Dataset):
-        def __init__(self, csv_path, transform=None):
-            self.data = pd.read_csv(csv_path)
-            self.image_paths = self.data['new_path']
-            self.labels = self.data['label']
-            self.transform = transform
-    
-        def __len__(self):
-            return len(self.data)
-    
-        def __getitem__(self, idx):
-            img_path = self.image_paths[idx]
-            label = self.labels[idx]
-    
-            try:
-                image = Image.open(img_path).convert("RGB")
-            except Exception as e:
-                raise RuntimeError(f"Error loading image from {img_path}: {e}")
-    
-            if self.transform:
-                image = self.transform(image)
-    
-            return image, torch.tensor(label, dtype=torch.long)
-    
-# ---- Load Data from Hugging Face (cached) ----
+
+# ---- Load Data (with caching) ----
 @st.cache_resource
-def load_test_data_from_huggingface():
-    dataset = load_dataset(
-        "Ci-Dave/DDR_dataset_train_test",
-        data_files={"test": "splits/test_labels_newpath.csv"},
-        split="test"
-    )
-    df = dataset.to_pandas()
-    csv_path = "test_labels_temp.csv"
-    df.to_csv(csv_path, index=False)
+def load_test_data(csv_path):
     dataset = DDRDataset(csv_path=csv_path, transform=val_transform)
     return DataLoader(dataset, batch_size=32, shuffle=False)
 
-# ---- Load Model from Hugging Face (cached) ----
+# ---- Load Model (with caching) ----
 @st.cache_resource
 def load_model():
-    model_path = hf_hub_download(repo_id="Ci-Dave/Densenet121", filename="Pretrained_Densenet-121.pth")
     model = models.densenet121(pretrained=False)
     model.classifier = nn.Linear(model.classifier.in_features, len(class_names))
-    model.load_state_dict(torch.load(model_path, map_location=torch.device('cpu')))
+    model.load_state_dict(torch.load(r"D:\\DR_Classification\\training\\Pretrained_Densenet-121.pth", map_location=torch.device('cpu')))
     model.eval()
     return model
 
-# ---- UI Buttons ----
+# ---- Main UI Buttons ----
+csv_path = r"D:\\DR_Classification\\splits\\test_labels.csv"
 model = load_model()
-test_loader = load_test_data_from_huggingface()
+test_loader = load_test_data(csv_path)
 
 col1, col2 = st.columns([1, 1])
+
 with col1:
     if st.button("🚀 Start Evaluation"):
         st.session_state.stop_eval = False
         st.session_state.evaluation_done = False
         st.session_state.trigger_eval = True
+
 with col2:
     if st.button("🚩 Stop Evaluation"):
         st.session_state.stop_eval = True