COCOData#

class COCOData[source]#

Class for loading the COCO dataset, inherits from DetectionData.

Implement the necessary methods to load the dataset.

__init__(data_loader: DataLoader, num_classes: Optional[int] = None, label_map: Optional[Dict[int, str]] = None, transform_field: Optional[str] = 'transforms')[source]#

__new__(*args, **kwargs)#

Attributes

`COCOData.classes_indices`	Return dict of classes as keys, and list of corresponding indices (in Dataset) of samples that include this class (in the label).
`COCOData.data_dimension`	Return how many dimensions the image data have.
`COCOData.data_loader`	Return the data loader.
`COCOData.has_images`	Return True if the data loader has images.
`COCOData.has_labels`	Return True if the data loader has labels.
`COCOData.n_of_samples_per_class`	Return a dictionary containing the number of samples per class.
`COCOData.num_classes`	Return the number of classes in the dataset.
`COCOData.num_samples`	Return the number of samples in the dataset.
`COCOData.original_num_samples`	Return the number of samples in the original dataset.
`COCOData.task_type`	Return the task type (object_detection).
`COCOData.transform_field`	Return the data loader.

Methods

`COCOData.assert_images_valid`()	Assert the image formatter defined is valid.
`COCOData.assert_labels_valid`()	Assert the label formatter defined is valid.
`COCOData.batch_of_index`(*indices)	Return batch samples of the given batch indices.
`COCOData.batch_to_images`(batch)	Convert the batch to a list of images.
`COCOData.batch_to_labels`(batch)	Convert the batch to a list of labels.
`COCOData.copy`([n_samples, shuffle, random_state])	Create new copy of this object, with the data-loader and dataset also copied, and altered by the given parameters.
`COCOData.from_dataset`(data[, batch_size, ...])	Create VisionData instance from a Dataset instance.
`COCOData.get_augmented_dataset`(aug)	Return a copy of the vision data object with the augmentation in the start of it.
`COCOData.get_classes`(batch_labels)	Get a labels batch and return classes inside it.
`COCOData.get_transform_type`()	Return transforms handler created from the transform field.
`COCOData.infer_on_batch`(batch, model, device)	Infer on a batch of images.
`COCOData.init_cache`()	Initialize the cache of the classes' metadata info.
`COCOData.is_sampled`()	Return whether the vision data is running on sample of the data.
`COCOData.label_id_to_name`(class_id)	Return the name of the class with the given id.
`COCOData.to_batch`(*samples)	Use the defined collate_fn to transform a few data items to batch format.
`COCOData.to_dataset_index`(*batch_indices)	Return for the given batch_index the sample index in the dataset object.
`COCOData.update_cache`(batch)	Get labels and update the classes' metadata info.
`COCOData.validate_format`(model[, device])	Validate the correctness of the data class implementation according to the expected format.
`COCOData.validate_get_classes`(batch)	Validate that the get_classes function returns data in the correct format.
`COCOData.validate_image_data`(batch)	Validate that the data is in the required format.
`COCOData.validate_infered_batch_predictions`(...)	Validate the infered predictions from the batch.
`COCOData.validate_label`(batch)	Validate the label.
`COCOData.validate_prediction`(batch, model, ...)	Validate the prediction.
`COCOData.validate_shared_label`(other)	Verify presence of shared labels.

load_model

COCOData.classes_indices