Class CocoDataset

• Defined in File datasets.h

Inheritance Relationships

Base Type

• public mindspore::dataset::Dataset (Class Dataset)

Class Documentation

class CocoDataset : public mindspore::dataset::Dataset

A source dataset for reading and parsing Coco dataset.

Public Functions

CocoDataset(const std::vector<char> &dataset_dir, const std::vector<char> &annotation_file, const std::vector<char> &task, const bool &decode, const std::shared_ptr<Sampler> &sampler, const std::shared_ptr<DatasetCache> &cache, const bool &extra_metadata)

Constructor of CocoDataset.

Parameters

• dataset_dir – [in] Path to the root directory that contains the dataset.

• annotation_file – [in] Path to the annotation json.

• task – [in] Set the task type of reading coco data, now support ‘Detection’/’Stuff’/’Panoptic’/’Keypoint’.

• decode – [in] Decode the images after reading.

• sampler – [in] Shared pointer to a sampler object used to choose samples from the dataset. If sampler is not given, a RandomSampler will be used to randomly iterate the entire dataset (default = RandomSampler()).

• cache – [in] Tensor cache to use (default=nullptr which means no cache is used).

• extra_metadata – [in] Flag to add extra meta-data to row. (default=false).

CocoDataset(const std::vector<char> &dataset_dir, const std::vector<char> &annotation_file, const std::vector<char> &task, const bool &decode, const Sampler *sampler, const std::shared_ptr<DatasetCache> &cache, const bool &extra_metadata)

Constructor of CocoDataset.

Parameters

• dataset_dir – [in] Path to the root directory that contains the dataset.

• annotation_file – [in] Path to the annotation json.

• task – [in] Set the task type of reading coco data, now support ‘Detection’/’Stuff’/’Panoptic’/’Keypoint’.

• decode – [in] Decode the images after reading.

• sampler – [in] Raw pointer to a sampler object used to choose samples from the dataset.

• cache – [in] Tensor cache to use (default=nullptr which means no cache is used).

• extra_metadata – [in] Flag to add extra meta-data to row. (default=false).

CocoDataset(const std::vector<char> &dataset_dir, const std::vector<char> &annotation_file, const std::vector<char> &task, const bool &decode, const std::reference_wrapper<Sampler> sampler, const std::shared_ptr<DatasetCache> &cache, const bool &extra_metadata)

Constructor of CocoDataset.

Parameters

• dataset_dir – [in] Path to the root directory that contains the dataset.

• annotation_file – [in] Path to the annotation json.

• task – [in] Set the task type of reading coco data, now support ‘Detection’/’Stuff’/’Panoptic’/’Keypoint’.

• decode – [in] Decode the images after reading.

• sampler – [in] Sampler object used to choose samples from the dataset.

• cache – [in] Tensor cache to use (default=nullptr which means no cache is used).

• extra_metadata – [in] Flag to add extra meta-data to row. (default=false).

~CocoDataset() = default

Constructor of CocoDataset.