def _draw_back_pane(self, pane):
mode = None
with self.state.lock:
back_enabled = self.state.back_enabled
back_mode = self.state.back_mode
back_filt_mode = self.state.back_filt_mode
state_layer = self.state.layer
selected_unit = self.state.selected_unit
back_what_to_disp = self.get_back_what_to_disp()
if back_what_to_disp == 'disabled':
pane.data[:] = to_255(self.settings.window_background)
elif back_what_to_disp == 'stale':
pane.data[:] = to_255(self.settings.stale_background)
else:
# One of the backprop modes is enabled and the back computation (gradient or deconv) is up to date
grad_blob = self.net.blobs['data'].diff
# Manually deprocess (skip mean subtraction and rescaling)
#grad_img = self.net.deprocess('data', diff_blob)
grad_blob = grad_blob[0] # bc01 -> c01
grad_blob = grad_blob.transpose((1, 2, 0)) # c01 -> 01c
grad_img = grad_blob[:, :,
self._net_channel_swap_inv] # e.g. BGR -> RGB
# Mode-specific processing
assert back_mode in ('grad', 'deconv')
assert back_filt_mode in ('raw', 'gray', 'norm', 'normblur')
if back_filt_mode == 'raw':
grad_img = norm01c(grad_img, 0)
elif back_filt_mode == 'gray':
grad_img = grad_img.mean(axis=2)
grad_img = norm01c(grad_img, 0)
elif back_filt_mode == 'norm':
grad_img = np.linalg.norm(grad_img, axis=2)
grad_img = norm01(grad_img)
else:
grad_img = np.linalg.norm(grad_img, axis=2)
cv2.GaussianBlur(grad_img, (0, 0),
self.settings.caffevis_grad_norm_blur_radius,
grad_img)
grad_img = norm01(grad_img)
# If necessary, re-promote from grayscale to color
if len(grad_img.shape) == 2:
grad_img = np.tile(grad_img[:, :, np.newaxis], 3)
grad_img_resize = ensure_uint255_and_resize_to_fit(
grad_img, pane.data.shape)
pane.data[0:grad_img_resize.shape[0],
0:grad_img_resize.shape[1], :] = grad_img_resize
def _draw_back_pane(self, pane):
mode = None
with self.state.lock:
back_enabled = self.state.back_enabled
back_mode = self.state.back_mode
back_filt_mode = self.state.back_filt_mode
state_layer = self.state.layer
selected_unit = self.state.selected_unit
back_what_to_disp = self.get_back_what_to_disp()
if back_what_to_disp == 'disabled':
pane.data[:] = to_255(self.settings.window_background)
elif back_what_to_disp == 'stale':
pane.data[:] = to_255(self.settings.stale_background)
else:
# One of the backprop modes is enabled and the back computation (gradient or deconv) is up to date
grad_blob = self.net.blobs['data'].diff
# Manually deprocess (skip mean subtraction and rescaling)
#grad_img = self.net.deprocess('data', diff_blob)
grad_blob = grad_blob[0] # bc01 -> c01
grad_blob = grad_blob.transpose((1,2,0)) # c01 -> 01c
grad_img = grad_blob[:, :, self._net_channel_swap_inv] # e.g. BGR -> RGB
# Mode-specific processing
assert back_mode in ('grad', 'deconv')
assert back_filt_mode in ('raw', 'gray', 'norm', 'normblur')
if back_filt_mode == 'raw':
grad_img = norm01c(grad_img, 0)
elif back_filt_mode == 'gray':
grad_img = grad_img.mean(axis=2)
grad_img = norm01c(grad_img, 0)
elif back_filt_mode == 'norm':
grad_img = np.linalg.norm(grad_img, axis=2)
grad_img = norm01(grad_img)
else:
grad_img = np.linalg.norm(grad_img, axis=2)
cv2.GaussianBlur(grad_img, (0,0), self.settings.caffevis_grad_norm_blur_radius, grad_img)
grad_img = norm01(grad_img)
# If necessary, re-promote from grayscale to color
if len(grad_img.shape) == 2:
grad_img = np.tile(grad_img[:,:,np.newaxis], 3)
grad_img_resize = ensure_uint255_and_resize_to_fit(grad_img, pane.data.shape)
pane.data[0:grad_img_resize.shape[0], 0:grad_img_resize.shape[1], :] = grad_img_resize
def _draw_back_pane(self, pane):
mode = None
with self.state.lock:
back_enabled = self.state.back_enabled
back_mode = self.state.back_mode
back_filt_mode = self.state.back_filt_mode
state_layer = self.state.layer
selected_unit = self.state.selected_unit
back_what_to_disp = self.get_back_what_to_disp()
if back_what_to_disp == 'disabled':
pane.data[:] = to_255(self.settings.window_background)
elif back_what_to_disp == 'stale':
pane.data[:] = to_255(self.settings.stale_background)
else:
# One of the backprop modes is enabled and the back computation (gradient or deconv) is up to date
grad_img = self.my_net.get_input_gradient_as_image()
# Mode-specific processing
assert back_mode in ('grad', 'deconv')
assert back_filt_mode in ('raw', 'gray', 'norm', 'normblur')
if back_filt_mode == 'raw':
grad_img = norm01c(grad_img, 0)
elif back_filt_mode == 'gray':
grad_img = grad_img.mean(axis=2)
grad_img = norm01c(grad_img, 0)
elif back_filt_mode == 'norm':
grad_img = np.linalg.norm(grad_img, axis=2)
grad_img = norm01(grad_img)
else:
grad_img = np.linalg.norm(grad_img, axis=2)
cv2.GaussianBlur(grad_img, (0, 0),
self.settings.caffevis_grad_norm_blur_radius,
grad_img)
grad_img = norm01(grad_img)
# If necessary, re-promote from grayscale to color
if len(grad_img.shape) == 2:
grad_img = np.tile(grad_img[:, :, np.newaxis], 3)
grad_img_resize = ensure_uint255_and_resize_to_fit(
grad_img, pane.data.shape)
pane.data[0:grad_img_resize.shape[0],
0:grad_img_resize.shape[1], :] = grad_img_resize