Problems with the use of lib

[cgafool]

What would you like to be documented: 1、It is recommended that the community add the lib usage mode 2、In particular, you need to ensure that the lib code is used with minimal dependencies. 3、Basic lib operators need to be enriched and to be decoupled support various development modes of developers. such as dataset transformation、sample diversity partitioning and so on.

Why is this needed: Services such as inference, training, and evaluation are provided by developers. Therefore, the lib usage of the multi-container worker solution differs greatly from that of the current community.

[cgafool]

Labels:
1->kind/documentation 2->kind/documentation、kind/features 3->kind/features

[kubeedge-bot]

@JimmyYang20: GitHub didn't allow me to assign the following users: JoeyHwong.

Note that only kubeedge members, repo collaborators and people who have commented on this issue/PR can be assigned. Additionally, issues/PRs can only have 10 assignees at the same time. For more information please see the contributor guide

In response to [this](https://github.com/kubeedge/sedna/issues/194#issuecomment-918871604): >/assign @jaypume @JoeyHwong Instructions for interacting with me using PR comments are available [here](https://git.k8s.io/community/contributors/guide/pull-requests.md). If you have questions or suggestions related to my behavior, please file an issue against the [kubernetes/test-infra](https://github.com/kubernetes/test-infra/issues/new?title=Prow%20issue:) repository.

[JimmyYang20]

/assign @jaypume

[jaypume]

2、In particular, you need to ensure that the lib code is used with minimal dependencies.

Sure, optimization of Sedna's dependency needs to be done.

1、It is recommended that the community add the lib usage mode

If it is still not satisfy the requirement after optimization of Sedna, maybe we can consider that release a package like Sedna-mini to apply to different scenarios.

3、Basic lib operators need to be enriched and to be decoupled support various development modes of developers. such as dataset transformation、sample diversity partitioning and so on.

Some work of enhanced features about datasets is doing currently in Sedna, maybe we can have more discussion about what should Sedna provide about datasets.

[JoeyHwong-gk]

Please follow #184 , the basic dependencies on each feature should removed from requirement.txt. You can run pip install sedna[all] to install all the dependencies or just install the dependencies of target feature (e.g lifelong_learning) by pip install sedna[ll]. By default, the dependencies related to each single feature won't installed.

@jaypume