Fixed normalization issue for EUROSAT_CUSTOM_MODEL

app.py

@@ -38,11 +38,9 @@ def Pn(m, x):
 def L(a,b,m,x):
     return np.sqrt((2*m+1)/(b-a))*Pn(m, 2*(x-b)/(b-a)+1)
 
-eurosat_transform = v2.Compose([
-    v2.Resize((64, 64)),  # Resize images to 64x64
-    #transforms.ToTensor(),  # Convert images to PyTorch tensors,
-    v2.ToDtype(torch.float32),
-    v2.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])  # Normalize images
+eurosat_transform = transforms.Compose([
+    transforms.Normalize(mean=[0.0, 0.0, 0.0], std=[255.0, 255.0, 255.0]),  # normalize to [0,1] first
+    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])  # Normalize images
 ])
 
 class CNN(nn.Module):
@@ -81,7 +79,7 @@ def run_lime(input_image,
 
     model, weights, preprocess, names = fetch_model(model_name)
 
-    input_image_processed = preprocess(torch.from_numpy(input_image.transpose(2,0,1))).unsqueeze(0)
+    input_image_processed = preprocess(torch.from_numpy(input_image.astype(np.float32).transpose(2,0,1))).unsqueeze(0)
     logits = model(input_image_processed)
     probs = F.softmax(logits, dim=1)
 
@@ -94,12 +92,12 @@ def run_lime(input_image,
     def classifier_fn(images):
         print('classifier_fn', type(images), images.shape)
         
-        zz = preprocess(torch.from_numpy(images[0].transpose(2,0,1)))
+        zz = preprocess(torch.from_numpy(images[0].transpose(2,0,1).astype(np.float32)))
         c, w, h = zz.shape 
         batch = torch.zeros(batch_size, c, w, h)
         print('len(images)', len(images))
         for i in range(batch_size):
-            batch[i] = preprocess(torch.from_numpy(images[i].transpose(2,0,1)))
+            batch[i] = preprocess(torch.from_numpy(images[i].transpose(2,0,1).astype(np.float32)))
         print('batch', type(batch), batch.shape)
 
         logits = model(batch)
@@ -190,7 +188,7 @@ def run_hdmr(input_image,
 
     sam_segmented_image = segmented_image(input_image, masks, alpha=0.9)
 
-    batch = preprocess(torch.from_numpy(input_image.transpose(2,0,1))).unsqueeze(0)
+    batch = preprocess(torch.from_numpy(input_image.astype(np.float32).transpose(2,0,1))).unsqueeze(0)
     # Unit normalize the logits
     logits = model(batch)
     logits = logits[0].detach().numpy()
@@ -211,7 +209,7 @@ def run_hdmr(input_image,
             x_seg = mask['segmentation']
             x_input[x_seg == 1] =  x_input[x_seg == 1] * np.power(x[sample,i],2)
 
-        batch = preprocess(torch.from_numpy(x_input.transpose(2,0,1))).unsqueeze(0)
+        batch = preprocess(torch.from_numpy(x_input.astype(np.float32).transpose(2,0,1))).unsqueeze(0)
         # Unit normalize the logits
         logits_sample = model(batch)
         probs = logits_sample.squeeze(0).softmax(0)