one more fix
Browse files
dronescapes_reader/dronescapes_representations.py
CHANGED
@@ -16,7 +16,7 @@ class DepthRepresentation(NpzRepresentation):
|
|
16 |
|
17 |
@overrides
|
18 |
def plot_fn(self, x: tr.Tensor) -> np.ndarray:
|
19 |
-
x = x.numpy()
|
20 |
x = np.clip(x, self.min_depth, self.max_depth)
|
21 |
x = np.nan_to_num((x - x.min()) / (x.max() - x.min()), 0)
|
22 |
y = hot(x.squeeze())[..., 0:3]
|
@@ -27,7 +27,7 @@ class OpticalFlowRepresentation(NpzRepresentation):
|
|
27 |
"""OpticalFlowRepresentation. Implements depth task-specific stuff, like using flow_vis."""
|
28 |
@overrides
|
29 |
def plot_fn(self, x: tr.Tensor) -> np.ndarray:
|
30 |
-
return flow_vis.flow_to_color(x.squeeze().nan_to_num(0).numpy())
|
31 |
|
32 |
class SemanticRepresentation(NpzRepresentation):
|
33 |
"""SemanticRepresentation. Implements depth task-specific stuff, like using flow_vis."""
|
@@ -46,7 +46,7 @@ class SemanticRepresentation(NpzRepresentation):
|
|
46 |
|
47 |
@overrides
|
48 |
def plot_fn(self, x: tr.Tensor) -> np.ndarray:
|
49 |
-
x = x.squeeze().nan_to_num(0).numpy()
|
50 |
new_images = np.zeros((*x.shape, 3), dtype=np.uint8)
|
51 |
for i in range(self.n_classes):
|
52 |
new_images[x == i] = self.color_map[i]
|
|
|
16 |
|
17 |
@overrides
|
18 |
def plot_fn(self, x: tr.Tensor) -> np.ndarray:
|
19 |
+
x = x.detach().cpu().numpy()
|
20 |
x = np.clip(x, self.min_depth, self.max_depth)
|
21 |
x = np.nan_to_num((x - x.min()) / (x.max() - x.min()), 0)
|
22 |
y = hot(x.squeeze())[..., 0:3]
|
|
|
27 |
"""OpticalFlowRepresentation. Implements depth task-specific stuff, like using flow_vis."""
|
28 |
@overrides
|
29 |
def plot_fn(self, x: tr.Tensor) -> np.ndarray:
|
30 |
+
return flow_vis.flow_to_color(x.squeeze().nan_to_num(0).detach().cpu().numpy())
|
31 |
|
32 |
class SemanticRepresentation(NpzRepresentation):
|
33 |
"""SemanticRepresentation. Implements depth task-specific stuff, like using flow_vis."""
|
|
|
46 |
|
47 |
@overrides
|
48 |
def plot_fn(self, x: tr.Tensor) -> np.ndarray:
|
49 |
+
x = x.squeeze().nan_to_num(0).detach().cpu().numpy()
|
50 |
new_images = np.zeros((*x.shape, 3), dtype=np.uint8)
|
51 |
for i in range(self.n_classes):
|
52 |
new_images[x == i] = self.color_map[i]
|
dronescapes_reader/multitask_dataset.py
CHANGED
@@ -28,7 +28,7 @@ class NpzRepresentation:
|
|
28 |
|
29 |
def save_to_disk(self, data: tr.Tensor, path: Path):
|
30 |
"""stores this item to the disk which can then be loaded via `load_from_disk`"""
|
31 |
-
np.save(path, data.cpu().numpy(), allow_pickle=False)
|
32 |
|
33 |
def plot_fn(self, x: tr.Tensor) -> np.ndarray:
|
34 |
"""very basic implementation of converting this representation to a viewable image. You should overwrite this"""
|
@@ -37,7 +37,7 @@ class NpzRepresentation:
|
|
37 |
assert len(x.shape) == 3, x.shape # guaranteed to be (H, W, C) at this point
|
38 |
if x.shape[-1] != 3: x = x[..., 0:1]
|
39 |
if x.shape[-1] == 1: x = x.repeat(1, 1, 3)
|
40 |
-
x = x.nan_to_num(0).cpu().numpy() # guaranteed to be (H, W, 3) at this point hopefully
|
41 |
_min, _max = x.min((0, 1), keepdims=True), x.max((0, 1), keepdims=True)
|
42 |
if x.dtype != np.uint8: x = np.nan_to_num((x - _min) / (_max - _min) * 255, 0).astype(np.uint8)
|
43 |
return x
|
@@ -119,7 +119,7 @@ class MultiTaskDataset(Dataset):
|
|
119 |
for task_name in self.task_names:
|
120 |
t = self.task_types[task_name]
|
121 |
try:
|
122 |
-
t = t(task_name)
|
123 |
except Exception:
|
124 |
pass
|
125 |
self._tasks.append(t)
|
|
|
28 |
|
29 |
def save_to_disk(self, data: tr.Tensor, path: Path):
|
30 |
"""stores this item to the disk which can then be loaded via `load_from_disk`"""
|
31 |
+
np.save(path, data.cpu().detach().numpy(), allow_pickle=False)
|
32 |
|
33 |
def plot_fn(self, x: tr.Tensor) -> np.ndarray:
|
34 |
"""very basic implementation of converting this representation to a viewable image. You should overwrite this"""
|
|
|
37 |
assert len(x.shape) == 3, x.shape # guaranteed to be (H, W, C) at this point
|
38 |
if x.shape[-1] != 3: x = x[..., 0:1]
|
39 |
if x.shape[-1] == 1: x = x.repeat(1, 1, 3)
|
40 |
+
x = x.nan_to_num(0).cpu().detach().numpy() # guaranteed to be (H, W, 3) at this point hopefully
|
41 |
_min, _max = x.min((0, 1), keepdims=True), x.max((0, 1), keepdims=True)
|
42 |
if x.dtype != np.uint8: x = np.nan_to_num((x - _min) / (_max - _min) * 255, 0).astype(np.uint8)
|
43 |
return x
|
|
|
119 |
for task_name in self.task_names:
|
120 |
t = self.task_types[task_name]
|
121 |
try:
|
122 |
+
t = t(task_name) # hack for not isinstance(self.task_types, NpzRepresentation) but callable
|
123 |
except Exception:
|
124 |
pass
|
125 |
self._tasks.append(t)
|