Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
@@ -72,28 +72,28 @@ def classify_image(inp):
|
|
72 |
prediction = model(img_t.unsqueeze(0)).softmax(-1).flatten()
|
73 |
|
74 |
modulator = model.layers[0].blocks[11].modulation.modulator.norm(2, 1, keepdim=True)
|
75 |
-
modulator = nn.Upsample(size=img_t.shape[1:], mode='bilinear')(modulator)
|
76 |
modulator = modulator.squeeze(1).detach().permute(1, 2, 0).numpy()
|
77 |
modulator = (modulator - modulator.min()) / (modulator.max() - modulator.min())
|
78 |
-
cam0 = show_cam_on_image(img_d, modulator, use_rgb=True)
|
79 |
|
80 |
modulator = model.layers[0].blocks[8].modulation.modulator.norm(2, 1, keepdim=True)
|
81 |
-
modulator = nn.Upsample(size=img_t.shape[1:], mode='bilinear')(modulator)
|
82 |
modulator = modulator.squeeze(1).detach().permute(1, 2, 0).numpy()
|
83 |
modulator = (modulator - modulator.min()) / (modulator.max() - modulator.min())
|
84 |
-
cam1 = show_cam_on_image(img_d, modulator, use_rgb=True)
|
85 |
|
86 |
modulator = model.layers[0].blocks[5].modulation.modulator.norm(2, 1, keepdim=True)
|
87 |
-
modulator = nn.Upsample(size=img_t.shape[1:], mode='bilinear')(modulator)
|
88 |
modulator = modulator.squeeze(1).detach().permute(1, 2, 0).numpy()
|
89 |
modulator = (modulator - modulator.min()) / (modulator.max() - modulator.min())
|
90 |
-
cam2 = show_cam_on_image(img_d, modulator, use_rgb=True)
|
91 |
|
92 |
modulator = model.layers[0].blocks[2].modulation.modulator.norm(2, 1, keepdim=True)
|
93 |
-
modulator = nn.Upsample(size=img_t.shape[1:], mode='bilinear')(modulator)
|
94 |
modulator = modulator.squeeze(1).detach().permute(1, 2, 0).numpy()
|
95 |
modulator = (modulator - modulator.min()) / (modulator.max() - modulator.min())
|
96 |
-
cam3 = show_cam_on_image(img_d, modulator, use_rgb=True)
|
97 |
|
98 |
return {labels[i]: float(prediction[i]) for i in range(1000)}, Image.fromarray(cam0), Image.fromarray(cam1), Image.fromarray(cam2), Image.fromarray(cam3)
|
99 |
|
|
|
72 |
prediction = model(img_t.unsqueeze(0)).softmax(-1).flatten()
|
73 |
|
74 |
modulator = model.layers[0].blocks[11].modulation.modulator.norm(2, 1, keepdim=True)
|
75 |
+
# modulator = nn.Upsample(size=img_t.shape[1:], mode='bilinear')(modulator)
|
76 |
modulator = modulator.squeeze(1).detach().permute(1, 2, 0).numpy()
|
77 |
modulator = (modulator - modulator.min()) / (modulator.max() - modulator.min())
|
78 |
+
cam0 = np.uint8(255 * modulator) # show_cam_on_image(img_d, modulator, use_rgb=True)
|
79 |
|
80 |
modulator = model.layers[0].blocks[8].modulation.modulator.norm(2, 1, keepdim=True)
|
81 |
+
# modulator = nn.Upsample(size=img_t.shape[1:], mode='bilinear')(modulator)
|
82 |
modulator = modulator.squeeze(1).detach().permute(1, 2, 0).numpy()
|
83 |
modulator = (modulator - modulator.min()) / (modulator.max() - modulator.min())
|
84 |
+
cam1 = np.uint8(255 * modulator) # show_cam_on_image(img_d, modulator, use_rgb=True)
|
85 |
|
86 |
modulator = model.layers[0].blocks[5].modulation.modulator.norm(2, 1, keepdim=True)
|
87 |
+
# modulator = nn.Upsample(size=img_t.shape[1:], mode='bilinear')(modulator)
|
88 |
modulator = modulator.squeeze(1).detach().permute(1, 2, 0).numpy()
|
89 |
modulator = (modulator - modulator.min()) / (modulator.max() - modulator.min())
|
90 |
+
cam2 = np.uint8(255 * modulator) # show_cam_on_image(img_d, modulator, use_rgb=True)
|
91 |
|
92 |
modulator = model.layers[0].blocks[2].modulation.modulator.norm(2, 1, keepdim=True)
|
93 |
+
# modulator = nn.Upsample(size=img_t.shape[1:], mode='bilinear')(modulator)
|
94 |
modulator = modulator.squeeze(1).detach().permute(1, 2, 0).numpy()
|
95 |
modulator = (modulator - modulator.min()) / (modulator.max() - modulator.min())
|
96 |
+
cam3 = np.uint8(255 * modulator) # show_cam_on_image(img_d, modulator, use_rgb=True)
|
97 |
|
98 |
return {labels[i]: float(prediction[i]) for i in range(1000)}, Image.fromarray(cam0), Image.fromarray(cam1), Image.fromarray(cam2), Image.fromarray(cam3)
|
99 |
|