Skip to content

Add class_order_per_task param to the loader #265

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 6 commits into from
Jul 30, 2025
Merged

Add class_order_per_task param to the loader #265

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
a434c6d
feat: add class_order_per_task param to the loader
sokovninn Jul 20, 2025
d129f6d
docs: remove permanent overwriting
sokovninn Jul 21, 2025
30e731d
fix: class ordering changed in dataset
sokovninn Jul 25, 2025
111ad80
Merge branch 'feat/class-reordering-in-loader' of https://github.com/…
sokovninn Jul 25, 2025
70b58f2
fix: handle None value in class_order_per_task
sokovninn Jul 25, 2025
ffeccad
Merge branch 'main' into feat/class-reordering-in-loader
sokovninn Jul 30, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
23 changes: 12 additions & 11 deletions luxonis_train/loaders/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,17 +12,18 @@ The default loader used with `LuxonisTrain`. It can either load data from an alr

**Parameters:**

| Key | Type | Default value | Description |
| --------------------- | --------------------------------------------------------------------------------------------------------- | ------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `dataset_name` | `str` | `None` | Name of the dataset to load. If not provided, the `dataset_dir` must be provided instead |
| `dataset_dir` | `str` | `None` | Path to the directory containing the dataset. If not provided, the `dataset_name` must be provided instead. Can be a path to a local directory or a URL. The data can be in a zip archive. New `LuxonisDataset` will be created using data from this directory and saved under the provided `dataset_name` |
| `dataset_type` | `Literal["coco", "voc", "darknet", "yolov6", "yolov4", "createml", "tfcsv", "clsdir", "segmask"] \| None` | `None` | Type of the dataset. If not provided, the type will be inferred from the directory structure |
| `team_id` | `str \| None` | `None` | Optional unique team identifier for the cloud |
| `bucket_storage` | `Literal["local", "s3", "gcs"]` | `"local"` | Type of the bucket storage |
| `delete_existing` | `bool` | `False` | Whether to delete the existing dataset with the same name. Only relevant if `dataset_dir` is provided. Use if you want to reparse the directory in case the data changed |
| `update_mode` | `Literal["all", "missing"]` | `all` | Select whether to download all remote dataset media files or only those missing locally |
| `min_bbox_visibility` | `float` | `0.0` | Minimum fraction of the original bounding box that must remain visible after augmentation. |
| `bbox_area_threshold` | `float` | `0.0004` | Minimum area threshold for bounding boxes to be considered valid (relative units in [0,1]). Boxes (and their related keypoints) with area below this will be filtered out. |
| Key | Type | Default value | Description |
| ---------------------- | --------------------------------------------------------------------------------------------------------- | ------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `dataset_name` | `str` | `None` | Name of the dataset to load. If not provided, the `dataset_dir` must be provided instead |
| `dataset_dir` | `str` | `None` | Path to the directory containing the dataset. If not provided, the `dataset_name` must be provided instead. Can be a path to a local directory or a URL. The data can be in a zip archive. New `LuxonisDataset` will be created using data from this directory and saved under the provided `dataset_name` |
| `dataset_type` | `Literal["coco", "voc", "darknet", "yolov6", "yolov4", "createml", "tfcsv", "clsdir", "segmask"] \| None` | `None` | Type of the dataset. If not provided, the type will be inferred from the directory structure |
| `team_id` | `str \| None` | `None` | Optional unique team identifier for the cloud |
| `bucket_storage` | `Literal["local", "s3", "gcs"]` | `"local"` | Type of the bucket storage |
| `delete_existing` | `bool` | `False` | Whether to delete the existing dataset with the same name. Only relevant if `dataset_dir` is provided. Use if you want to reparse the directory in case the data changed |
| `update_mode` | `Literal["all", "missing"]` | `all` | Select whether to download all remote dataset media files or only those missing locally |
| `min_bbox_visibility` | `float` | `0.0` | Minimum fraction of the original bounding box that must remain visible after augmentation. |
| `bbox_area_threshold` | `float` | `0.0004` | Minimum area threshold for bounding boxes to be considered valid (relative units in [0,1]). Boxes (and their related keypoints) with area below this will be filtered out. |
| `class_order_per_task` | `dict[str, List[str]] \| None` | `None` | If provided, the classes for the specified tasks will be reordered in the dataset. |

**Data Shape Definitions:**

Expand Down
6 changes: 6 additions & 0 deletions luxonis_train/loaders/luxonis_loader_torch.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,6 +29,7 @@ def __init__(
filter_task_names: list[str] | None = None,
min_bbox_visibility: float = 0.0,
bbox_area_threshold: float = 0.0004,
class_order_per_task: dict[str, list[str]] | None = None,
seed: int | None = None,
**kwargs,
):
Expand Down Expand Up @@ -81,6 +82,9 @@ def __init__(
@type bbox_area_threshold: float
@param bbox_area_threshold: Minimum area threshold for bounding boxes to be considered valid. In the range [0, 1].
Default is 0.0004, which corresponds to a small area threshold to remove invalid bboxes and respective keypoints.
@type class_order_per_task: dict[str, list[str]] | None
@param class_order_per_task: Dictionary mapping task names to a list of class names.
If provided, the classes for the specified tasks will be reordered.
@type seed: Optional[int]
@param seed: The random seed to use for the augmentations.
"""
Expand All @@ -100,6 +104,8 @@ def __init__(
bucket_type=bucket_type,
bucket_storage=bucket_storage,
)
if class_order_per_task is not None:
self.dataset.set_class_order_per_task(class_order_per_task)
self.loader = LuxonisLoader(
dataset=self.dataset,
view=self.view,
Expand Down
Loading