def show_seg_gt(self, num_samples=5):
imgs = []
seg = []
for i in range(num_samples):
img, lbl = self.ds[i]
imgs.append(img)
seg.append(lbl)
show([(self.iv.image(imgs[i]), self.iv.segmentation(seg[i]),
self.iv.segment_key_with_lbls(seg[i], self.seglabels))
for i in range(len(seg))])
def show_idx(idxlist, val_dataset, source='imagenet'):
"""
Show images corresponding to the inputted index list
"""
if len(idxlist) == 1:
print("image", idxlist[0])
show(
renormalize.as_image(split_img(val_dataset[idxlist[0]][0]),
source=source))
else:
print("idx list", idxlist)
show([[
renormalize.as_image(split_img(val_dataset[idx][0]), source=source)
] for idx in idxlist])
def show_top_activating_imgs_per_units_with_seg(self, units, top_num=1):
if self.topk is None:
self.compute_topk_imgs()
top_indexes = self.topk.result()[1]
show([[
'unit %d' % u,
'img %d' % i,
'pred: %s' % self._get_pred(i),
[
self.iv.masked_image(
self.ds[i.item()][0],
self.model.retained_layer(self.layername)[0], u)
],
[self.iv.heatmap(self.model.retained_layer(self.layername)[0], u)],
[self.iv.segmentation(self.ds[i.item()][1])]
] for u in units for i in top_indexes[u, :top_num]])
def show_seg_results(self):
if self.unit_images is None:
self.compute_top_unit_imgs()
level_at_99 = self.rq.quantiles(0.99).cuda()[None, :, None, None]
sample_size = 20
def compute_selected_segments(batch, *args):
img, seg = batch
# show(iv.segmentation(seg))
image_batch = img.cuda()
seg_batch = seg.cuda()
_ = self.model(image_batch)
acts = self.model.retained_layer(self.layername)
hacts = self.upfn(acts)
iacts = (hacts >
level_at_99).float() # indicator where > 0.99 percentile.
return tally.conditional_samples(iacts, seg_batch)
condi99 = tally.tally_conditional_mean(compute_selected_segments,
dataset=self.ds,
sample_size=sample_size,
loader=self.ds_loader,
pass_with_lbl=True)
self.iou99 = tally.iou_from_conditional_indicator_mean(condi99)
bolded_string = "\033[1m" + self.layername + "\033[0m"
print(bolded_string)
iou_unit_label_99 = sorted([
(unit, concept.item(), self.seglabels[int(concept)],
bestiou.item())
for unit, (bestiou, concept) in enumerate(zip(*self.iou99.max(0)))
],
key=lambda x: -x[-1])
for unit, concept, label, score in iou_unit_label_99[:20]:
show([
'unit %d; iou %g; label "%s"' % (unit, score, label),
[self.unit_images[unit]]
])