* add get_subset method, add comment, rename variable

* add unit test

* Please mypy

* Fix as comment, support negative index, and fix index access error

* add and refine docstring, handle indices=0

* handle indices=0

* add empty list indices test

* rename in_meta in docstring and comments to in_metainfo

* clean meta nameing

* Fix negative indices error

* test empty list of get_subset

* fix comments and docstring

* add unit test

* Fix as comment

* Fix as comment

* add docstring to mention wrapped dataset should not inherit from Basedataset

* Fix wrapped dataset docstring

* Fix wrapped dataset docstring

* Fix method name, docstring, and comments

* Fix comments

* Fix comments

* Fix comments

* [Enhancement] refactor base data elment

* fix comment

* fix comment

* fix pop not existing key without error

* [WIP] abstract-data-structure

* update docs

* update

* update BaseDataSample

* fix comment and coverage 100%

* update and add _set_field

* update

* split  into base_data_element and base_data_sample

* update

* update

* update

* fix typo

* basedataset first commit

* add base dataset

* add dataset

* add basedataset

* Fix test dataset

* Fix mypy and test

* Fix mypy and test

* remove unused code

* Update mmengine/dataset/base_dataset.py

Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>

* Update mmengine/dataset/base_dataset.py

Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>

* add more corner cases in unittest

* fix lint

* Fix as comment

* Fix lint

* update unitest

* Type hint Dick to dict

* rename max_refetch

* Fix as comment

* Fix typo

* Fix as comment

* BaseDataset is no more an abstrac Class, change UT and docs

* Fix as comment

* Fix as comment and refactor type error

* Add comment for full init

* Fix as comment and modify dataset_wrapper

* Fix as comment and modify dataset_wrapper

* Fix as comment

* Fix as comment

* Fix as comment

* Fix as comment

* Fix as comment

* Fix as comment

* Fix as comment

Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
Co-authored-by: Tao Gong <gongtao950513@gmail.com>

* Add DefaultSampler and InfiniteSampler

* Add unit test

* Add samplers to API reference

* Update docstring

* Improve according to comments

* Rename `num_replicas` to `world_size`

* Update docstring.

* Update mmengine/data/sampler.py

Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>

* Update mmengine/data/sampler.py

Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>

* Fix typo in unit test

Co-authored-by: Wenwei Zhang <40779233+ZwwWayne@users.noreply.github.com>
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>

* add unit tests of data abstract interface

* update

* update

* update docs of data element

* a draft of UT of datasample, to be finished

* update datasample test

* updata

* update

* fix comments

* fix comments

* fix comments

Co-authored-by: liukuikun <liukuikun@sensetime.com>

* move fileio directory from utils

* remove CephBackend from fileio and clear some deprecated docstring

* fix format

* fix a spellling error

* tmp commit

* add unittest for basedataset

* fix lint

* resolve all comments

* update

* update

* update

* update

* update

* update

* update

* resolve comments

* resolve comments

* update

* resolve comments

* resolve comments