The streetscapes module
ModelBase
Bases: ABC
__init__(device=None)
A model serving as the base for all segmentation models.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
device
|
str
|
Device to use for processign. Defaults to None. |
None
|
load_image(path)
A list of images or paths to image files.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
path
|
Path | str
|
A path to an image. |
required |
Returns:
| Type | Description |
|---|---|
tuple[list[Path], list[ndarray]]
|
A tuple containing: 1. The ID of the image. 2. The images as a NumPy array. |
load_images(images)
A list of images or paths to image files.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
images
|
PathLike
|
A path or a list of paths to image files. |
required |
Returns:
| Type | Description |
|---|---|
tuple[list[Path], list[ndarray]]
|
A tuple containing: 1. The paths to the images. 2. The images as NumPy arrays. |
segment(paths, labels, batch_size=10)
Retrieve the paths of local images from a dataset.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
paths
|
PathLike
|
Path(s) to image file(s). |
required |
labels
|
dict
|
A flattened set of labels to look for,
with optional subsets of labels that should be
checked in order to eliminate overlaps.
Cf. |
required |
batch_size
|
int
|
Process the images in batches of this size. Defaults to 10. |
10
|
Returns:
| Type | Description |
|---|---|
Table
|
A table of information about the segmentations. |