Jie Pu
6753b229eb
update generated crd code and fix ci
Signed-off-by: Jie Pu <i@jaypu.com>
3 years ago
Jie Pu
0e43075c44
Support showing knowledge base in crd status field
Signed-off-by: Jie Pu <i@jaypu.com>
3 years ago
Jie Pu
e1c920ac00
Add knowledge base types definition
Signed-off-by: Jie Pu <i@jaypu.com>
3 years ago
JimmyYang20
6f4b1f6896
LL: initial container can download label file
Signed-off-by: JimmyYang20 <yangjin39@huawei.com>
4 years ago
KubeEdge Bot
75e66ca576
Merge pull request #315 from JimmyYang20/il-compute
IL supports multipile device soc versions
4 years ago
JimmyYang20
f58213fcd5
IL supports multipile device soc versions
Signed-off-by: JimmyYang20 <yangjin39@huawei.com>
4 years ago
Iggie Wang
e5e80ccf15
Merge branch 'kubeedge:main' into main
4 years ago
JimmyYang20
104943df3b
LL: support schedule workers on multi-nodes
worker's spec supports nodeName and nodeSelector
Signed-off-by: JimmyYang20 <yangjin39@huawei.com>
4 years ago
hey-kong
5a39c7ed1b
update DirectoryorCreate to DirectoryOrCreate
Signed-off-by: hey-kong <iggiewang@gmail.com>
4 years ago
JimmyYang20
7310b7b253
Fixbug: db path has missed the mount volume prefix in LC
Signed-off-by: JimmyYang20 <yangjin39@huawei.com>
4 years ago
JimmyYang20
5bf70e60ef
IL: decouple eval task and deploy task
Signed-off-by: JimmyYang20 <yangjin39@huawei.com>
4 years ago
KubeEdge Bot
3d7c041749
Merge pull request #237 from JimmyYang20/fixbug
Fix IL bug: job misses first data when reads data.
4 years ago
JimmyYang20
9da7acbd2a
Fix IL bug: job misses first data when reads data.
Signed-off-by: JimmyYang20 <yangjin39@huawei.com>
4 years ago
JimmyYang20
8e7122db56
Fix IL bug: TrainTriggerStatus used in eval task
need to chenge TrainTriggerStatus to EvalTriggerStatus
Signed-off-by: JimmyYang20 <yangjin39@huawei.com>
4 years ago
JimmyYang20
0223441530
GM&LC: IL supports model hot updates
Signed-off-by: JimmyYang20 <yangjin39@huawei.com>
4 years ago
JimmyYang20
bf39a7c1bd
IL: LC supports to recover job when restart
Signed-off-by: JimmyYang20 <yangjin39@huawei.com>
5 years ago
JimmyYang20
44609f0090
lc: decouple all features into independent package
Signed-off-by: JimmyYang20 <yangjin39@huawei.com>
5 years ago
KubeEdge Bot
0c0e7cd337
Merge pull request #134 from llhuii/refactor-gm
gm: decouple all features into independent package
5 years ago
KubeEdge Bot
dc3e954a1c
Merge pull request #137 from JimmyYang20/main
LC: send dataset update to GM only when changed
5 years ago
JimmyYang20
4259fd3a18
LC: send dataset update to GM only when changed
number of samples has been sent to GM only when adding new data.
Signed-off-by: JimmyYang20 <yangjin39@huawei.com>
5 years ago
JimmyYang20
3582ec9876
LC: fix nil pointer dereference bug
It happened in evalTask of incremental job when deploy model hasn't
been synced to LC. evalTask should return error instead of logging
error. And it doesn't need job id info into error, same as trainTask.
Signed-off-by: JimmyYang20 <yangjin39@huawei.com>
5 years ago
llhuii
e52ac06ed1
gm: split all upstream logic into separate file
Signed-off-by: llhuii <liulinghui@huawei.com>
5 years ago
llhuii
48135b2393
gm: refactor all features into independent dir
All controllers are placed into globalmanager/controllers:
1. each feature has the independent subdirectory
2. upstream/downstream are kept as top level.
Commom types/utils/worker.go are placed into globalmanager/runtime.
Signed-off-by: llhuii <liulinghui@huawei.com>
5 years ago
JimmyYang20
f0c2a97084
IL: add support of multi-node train/eval
Signed-off-by: JimmyYang20 <yangjin39@huawei.com>
5 years ago
KubeEdge Bot
e92e1ddaab
Merge pull request #78 from llhuii/update-owners
Update owners
5 years ago
llhuii
b4e7c08961
Remove khalid-davis from OWNERS
Signed-off-by: llhuii <liulinghui@huawei.com>
5 years ago
JimmyYang20
babb8f9bde
Fix the review comments
Signed-off-by: JimmyYang20 <yangjin39@huawei.com>
5 years ago
JimmyYang20
ff0a5ee6c3
LC: add lifelonglearningjob
Signed-off-by: JimmyYang20 <yangjin39@huawei.com>
5 years ago
JimmyYang20
656520c093
Fix shared storage code in LC according the comments
Signed-off-by: JimmyYang20 <yangjin39@huawei.com>
5 years ago
JimmyYang20
f88f6b1d50
Add shared storage code for LC
Signed-off-by: JimmyYang20 <yangjin39@huawei.com>
5 years ago
llhuii
940ec7049c
Add pod template for incremental learning
Signed-off-by: llhuii <liulinghui@huawei.com>
5 years ago
llhuii
974d38f948
Update OWNERS
From https://git.k8s.io/community/contributors/guide/owners.md ,
the approvers should be not in the reviewers section.
Signed-off-by: llhuii <liulinghui@huawei.com>
5 years ago
llhuii
847e81ae8f
Add OWNERS files
Signed-off-by: llhuii <liulinghui@huawei.com>
5 years ago
llhuii
016e649931
Add copyright
also add a python script used to generate the copyright
Signed-off-by: llhuii <liulinghui@huawei.com>
5 years ago
llhuii
32ef5c1e90
Rename neptune to sedna
Signed-off-by: llhuii <liulinghui@huawei.com>
5 years ago
JimmyYang
37efbffed1
Add incremental learning job manager in LC
Signed-off-by: JimmyYang <yangjin39@huawei.com>
5 years ago
JimmyYang
c4cfc1bed7
Optimize localcontroller
1. reuse CR definition of pkg/apis in all managers.
2. cleanup server.
3. fix db getClient bug, and optimize it.
4. rename wsclient to gmclient.
5. fix GetUniqueIdentifier bug.
Signed-off-by: JimmyYang <yangjin39@huawei.com>
5 years ago
JimmyYang
8b0eb11921
Add federated learning job manager in localcontroller
Signed-off-by: JimmyYang <yangjin39@huawei.com>
5 years ago
llhuii
ea2d6891bc
Fix golint issues
Signed-off-by: llhuii <liulinghui@huawei.com>
5 years ago
JimmyYang
ab15aca37e
Add localcontroller for joint inference
Signed-off-by: JimmyYang <yangjin39@huawei.com>
5 years ago