Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

feat: Auto image build #67

Merged
merged 242 commits into from
Oct 18, 2023
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
242 commits
Select commit Hold shift + click to select a range
67263d9
fix: new dockerfile for llama image building
ishaansehgal99 Oct 3, 2023
06a6355
feat: added github action for build and deploy image
ishaansehgal99 Oct 4, 2023
999518b
fix: fix az login
ishaansehgal99 Oct 4, 2023
d8ae6fc
fix: fixed issues with git action
ishaansehgal99 Oct 4, 2023
10a28c8
feat: add workflow dispatch release
ishaansehgal99 Oct 4, 2023
2e50d93
nit: add for testing action
ishaansehgal99 Oct 4, 2023
5d59197
nit: testing git action
ishaansehgal99 Oct 4, 2023
b1e6c66
fix: add token permissions
ishaansehgal99 Oct 4, 2023
31bdb04
nit: trigger falcon
ishaansehgal99 Oct 4, 2023
6c984bb
nit: retrigger falcon
ishaansehgal99 Oct 4, 2023
5ae5b61
fix: set subscription
ishaansehgal99 Oct 4, 2023
260d5b3
nit: retrigger falcon
ishaansehgal99 Oct 4, 2023
98b5308
nit: fix secret
ishaansehgal99 Oct 4, 2023
55204d7
nit: change falcon
ishaansehgal99 Oct 4, 2023
17b5897
nit: remove set sub
ishaansehgal99 Oct 4, 2023
a770064
nit: retrigger falcon
ishaansehgal99 Oct 4, 2023
e306c72
nit: upgrade version
ishaansehgal99 Oct 4, 2023
17af4a7
nit: retrigger falcon
ishaansehgal99 Oct 4, 2023
2a79864
nit: fix
ishaansehgal99 Oct 4, 2023
9eb5e62
nit: retrigger falconm
ishaansehgal99 Oct 4, 2023
c030aeb
nit: cli
ishaansehgal99 Oct 4, 2023
23d6946
nit: retrigger falcon
ishaansehgal99 Oct 4, 2023
8278152
nit: cli
ishaansehgal99 Oct 4, 2023
da074aa
nit: run
ishaansehgal99 Oct 4, 2023
66c4283
nit: reorder
ishaansehgal99 Oct 4, 2023
295b006
nit: try login
ishaansehgal99 Oct 4, 2023
9ff4808
nit: try login
ishaansehgal99 Oct 4, 2023
bb56dbe
nit: retry falcon
ishaansehgal99 Oct 4, 2023
e37ff4b
nit: try pr
ishaansehgal99 Oct 5, 2023
5982e7a
nit: try client id
ishaansehgal99 Oct 5, 2023
6059ee3
nit: try sub id
ishaansehgal99 Oct 5, 2023
aaaa0f8
nit: trigger falcon
ishaansehgal99 Oct 5, 2023
b200381
nit: try sub
ishaansehgal99 Oct 5, 2023
f877bcf
nit: trigger falcon
ishaansehgal99 Oct 5, 2023
c2c7fa6
nit: sub
ishaansehgal99 Oct 5, 2023
10b27d0
nit: retry falcon
ishaansehgal99 Oct 5, 2023
b8360d6
nit: paths added
ishaansehgal99 Oct 5, 2023
4153601
nit: falcon retrigger
ishaansehgal99 Oct 5, 2023
7218c04
nit: update preset
ishaansehgal99 Oct 5, 2023
efcd3d0
nit: falcon changer
ishaansehgal99 Oct 5, 2023
62efdba
nit: maybe now
ishaansehgal99 Oct 5, 2023
2f90249
nit: try now
ishaansehgal99 Oct 5, 2023
432951f
nit: maybe now
ishaansehgal99 Oct 5, 2023
b5619e1
nit: maybe now
ishaansehgal99 Oct 5, 2023
42d91c3
nit: maybe now
ishaansehgal99 Oct 5, 2023
847d64a
nit: maybe now
ishaansehgal99 Oct 5, 2023
f4920f8
nit: maybe now
ishaansehgal99 Oct 5, 2023
c980136
nit: maybe now
ishaansehgal99 Oct 5, 2023
de28fe4
nit: maybe now
ishaansehgal99 Oct 5, 2023
608f9ea
nit: maybe now
ishaansehgal99 Oct 5, 2023
f58ce15
nit: maybe now
ishaansehgal99 Oct 5, 2023
f477cf0
nit: maybe now
ishaansehgal99 Oct 5, 2023
0e8865a
nit: maybe now
ishaansehgal99 Oct 5, 2023
097ef85
nit: maybe now
ishaansehgal99 Oct 5, 2023
d2ffe39
nit: maybe now
ishaansehgal99 Oct 5, 2023
84a8a17
nit: maybe now
ishaansehgal99 Oct 5, 2023
4824375
nit: maybe now
ishaansehgal99 Oct 5, 2023
f6956ec
nit: maybe now
ishaansehgal99 Oct 5, 2023
d97100a
nit: maybe now
ishaansehgal99 Oct 5, 2023
d0866c9
nit: maybe now
ishaansehgal99 Oct 5, 2023
621f305
nit: maybe now
ishaansehgal99 Oct 5, 2023
6d0378d
nit: maybe now
ishaansehgal99 Oct 5, 2023
6268f2a
nit: maybe now
ishaansehgal99 Oct 5, 2023
008ac29
nit: maybe now
ishaansehgal99 Oct 5, 2023
e9b8063
nit: maybe now
ishaansehgal99 Oct 5, 2023
958364a
nit: maybe now
ishaansehgal99 Oct 5, 2023
3685336
nit: maybe now
ishaansehgal99 Oct 5, 2023
b5ffc0c
nit: maybe now
ishaansehgal99 Oct 5, 2023
8dad3f0
nit: maybe now
ishaansehgal99 Oct 5, 2023
12d8afc
nit: maybe now
ishaansehgal99 Oct 5, 2023
9abc380
nit: maybe now
ishaansehgal99 Oct 5, 2023
0294d9c
nit: maybe now
ishaansehgal99 Oct 5, 2023
ff3ee48
nit: maybe now
ishaansehgal99 Oct 5, 2023
54946f1
nit: maybe now
ishaansehgal99 Oct 5, 2023
6528d19
nit: maybe now
ishaansehgal99 Oct 5, 2023
20b9bcd
fix: fixed relative paths in dockerfile
ishaansehgal99 Oct 5, 2023
3dce7f8
fix: fix relative path for docker file
ishaansehgal99 Oct 5, 2023
2adce12
nit: test image tag
ishaansehgal99 Oct 5, 2023
22f608e
nit: maybe now
ishaansehgal99 Oct 5, 2023
f8a85dd
nit: maybe now
ishaansehgal99 Oct 5, 2023
b900217
nit: trigger all rebuilds but falcon
ishaansehgal99 Oct 5, 2023
a266c7c
nit: maybe now
ishaansehgal99 Oct 5, 2023
576eb2c
nit: maybe now
ishaansehgal99 Oct 5, 2023
1eb5d97
feat: parallelized wget
ishaansehgal99 Oct 5, 2023
91263d1
feat: fix dockerfile for golang download script
ishaansehgal99 Oct 5, 2023
b38a01d
fix: fix file path docker file
ishaansehgal99 Oct 5, 2023
d6dca6b
fix: print filepath in download
ishaansehgal99 Oct 5, 2023
74e4fc7
fix: print each percent
ishaansehgal99 Oct 5, 2023
b9dc72c
fix: remove pull request trigger
ishaansehgal99 Oct 5, 2023
5280fd8
nit: rename falcon
ishaansehgal99 Oct 5, 2023
2c5d1b3
nit: remove file uncessary
ishaansehgal99 Oct 5, 2023
a6043ef
feat: added secrets for URL
ishaansehgal99 Oct 5, 2023
0731ef2
nit: retrigger
ishaansehgal99 Oct 5, 2023
d8418c1
nit: retrigger
ishaansehgal99 Oct 5, 2023
e46bafc
fix: updated dockerfiles
ishaansehgal99 Oct 6, 2023
eb55bac
feat: all new falcon images
ishaansehgal99 Oct 6, 2023
bccf8d7
feat: added token
ishaansehgal99 Oct 6, 2023
2ad7e67
fix filepath
ishaansehgal99 Oct 6, 2023
6273298
fix: add auth token
ishaansehgal99 Oct 6, 2023
144c930
fix: added download code for falcon
ishaansehgal99 Oct 8, 2023
41ad6eb
fix: fixed bugs with docker build and download weights
ishaansehgal99 Oct 8, 2023
00a5db7
fix: falcon naming and cmd line params for download script
ishaansehgal99 Oct 8, 2023
1e89f4b
nit: arglen fix
ishaansehgal99 Oct 9, 2023
c7ea9bb
feat: added e2e tests
ishaansehgal99 Oct 10, 2023
914d11b
fix: add delete nodepool
ishaansehgal99 Oct 10, 2023
5b60d15
fix: command
ishaansehgal99 Oct 10, 2023
075e4c4
fix
ishaansehgal99 Oct 10, 2023
05196d0
fix
ishaansehgal99 Oct 10, 2023
6f79beb
debug: e2e
ishaansehgal99 Oct 10, 2023
a09cab8
fix: nodepool name
ishaansehgal99 Oct 10, 2023
7c9c6dd
fix: nodepool name fix
ishaansehgal99 Oct 10, 2023
def4680
fix: llama name
ishaansehgal99 Oct 10, 2023
1612e94
fix: add node select
ishaansehgal99 Oct 10, 2023
1f53ffe
fix: add checks to nodepool
ishaansehgal99 Oct 10, 2023
c786492
fix: add checks to nodepool
ishaansehgal99 Oct 10, 2023
c92d835
nit: fixed tag
ishaansehgal99 Oct 10, 2023
b144854
nit: fix deployment name and nodepool check
ishaansehgal99 Oct 10, 2023
7a0d5ab
fix: naming of deployment and service
ishaansehgal99 Oct 10, 2023
302a89d
fix: fix tags
ishaansehgal99 Oct 10, 2023
2e1fd2b
fix: fix weights folder dir
ishaansehgal99 Oct 10, 2023
cb6c2c4
fix: build images
ishaansehgal99 Oct 10, 2023
c304ff3
run e2e on llama
ishaansehgal99 Oct 10, 2023
984a4cb
fix commit
ishaansehgal99 Oct 10, 2023
3328950
fix: yaml
ishaansehgal99 Oct 10, 2023
74e1a50
fix: include params.json
ishaansehgal99 Oct 10, 2023
0d88381
fix: new image
ishaansehgal99 Oct 10, 2023
637228e
fix: readded models
ishaansehgal99 Oct 10, 2023
3f8cf5c
fix workflows
ishaansehgal99 Oct 10, 2023
9175371
fix: print build args
ishaansehgal99 Oct 10, 2023
19899e3
fix: preset
ishaansehgal99 Oct 10, 2023
d689755
fix: add dockerfile to matrix
ishaansehgal99 Oct 10, 2023
56a6ddd
fix: retrigger latest
ishaansehgal99 Oct 10, 2023
687aa77
feat: add e2e yamls and additional tests
ishaansehgal99 Oct 10, 2023
5c54aa1
fix: use statefulsets
ishaansehgal99 Oct 11, 2023
15a0637
fix: params torchrun
ishaansehgal99 Oct 11, 2023
da5a52d
testing risk
ishaansehgal99 Oct 11, 2023
24ec36b
retrigger
ishaansehgal99 Oct 11, 2023
7edf1b4
fix: big fixes on svc and deployment
ishaansehgal99 Oct 11, 2023
04c50bb
fix: parallel pods and add log for master addr
ishaansehgal99 Oct 11, 2023
4bdffda
fix: added two ports
ishaansehgal99 Oct 11, 2023
e3e50bd
fix:retry
ishaansehgal99 Oct 11, 2023
6a28dde
name service
ishaansehgal99 Oct 11, 2023
80be520
fix: src_dir
ishaansehgal99 Oct 11, 2023
97ee181
update latest
ishaansehgal99 Oct 11, 2023
885f94b
fix: added falcon
ishaansehgal99 Oct 11, 2023
1bedcb7
retrigger e2e
ishaansehgal99 Oct 11, 2023
9cee0ad
feat: rebuild falcon with deepspeed
ishaansehgal99 Oct 11, 2023
0557fe9
fix: retrigger e2e
ishaansehgal99 Oct 11, 2023
430a380
fix
ishaansehgal99 Oct 11, 2023
97673c9
fix: revert changes retrigger e2e
ishaansehgal99 Oct 11, 2023
3b756d1
fix filepath
ishaansehgal99 Oct 11, 2023
5af6d72
fix falcon retry
ishaansehgal99 Oct 11, 2023
4e28dad
cleanup
ishaansehgal99 Oct 11, 2023
19f6434
fix: add cleanup fix
ishaansehgal99 Oct 11, 2023
9f00853
fix: added artifact for image tag
ishaansehgal99 Oct 11, 2023
6246400
run preset
ishaansehgal99 Oct 11, 2023
8cef8d4
added test acr
ishaansehgal99 Oct 11, 2023
f616ccd
self runner
ishaansehgal99 Oct 11, 2023
9a251e9
change all to self hosted
ishaansehgal99 Oct 11, 2023
260dae9
fix: add preset build
ishaansehgal99 Oct 11, 2023
8b2c573
fix: docker build
ishaansehgal99 Oct 12, 2023
a857803
retrigger
ishaansehgal99 Oct 12, 2023
a9bd838
retrigger
ishaansehgal99 Oct 12, 2023
beb5e78
restart
ishaansehgal99 Oct 12, 2023
4708100
fix
ishaansehgal99 Oct 12, 2023
f80a681
fix: add login
ishaansehgal99 Oct 12, 2023
8c37667
preset add
ishaansehgal99 Oct 12, 2023
08a209c
retry
ishaansehgal99 Oct 12, 2023
4700c36
fix preset
ishaansehgal99 Oct 12, 2023
7fde839
try fix verion
ishaansehgal99 Oct 12, 2023
6c0d4be
lets try this again
ishaansehgal99 Oct 12, 2023
25462bb
run e2e
ishaansehgal99 Oct 12, 2023
a86ea73
run e2e
ishaansehgal99 Oct 12, 2023
8b3f753
added federated identity
ishaansehgal99 Oct 12, 2023
b7a3c4d
retrigger with the correct parameters
ishaansehgal99 Oct 12, 2023
a205c23
fix: attach acr
ishaansehgal99 Oct 12, 2023
4e6d87b
pre
ishaansehgal99 Oct 12, 2023
2ff363e
run
ishaansehgal99 Oct 13, 2023
9b10ee8
login new client
ishaansehgal99 Oct 13, 2023
977c1c1
skip step need be
ishaansehgal99 Oct 13, 2023
87c47cc
retrigger
ishaansehgal99 Oct 13, 2023
1056e2d
retrigger
ishaansehgal99 Oct 13, 2023
c25ae03
try to retrigger
ishaansehgal99 Oct 13, 2023
583a4f6
folder restructing
ishaansehgal99 Oct 13, 2023
74600a2
Merge remote-tracking branch 'origin' into auto-image-build
ishaansehgal99 Oct 13, 2023
f6cd225
line
ishaansehgal99 Oct 13, 2023
b2d948a
fix: linter error
ishaansehgal99 Oct 13, 2023
307da12
fixed
ishaansehgal99 Oct 13, 2023
3dea57c
change file path
ishaansehgal99 Oct 13, 2023
404e6e2
fix: docker path
ishaansehgal99 Oct 13, 2023
018887b
fix:set context
ishaansehgal99 Oct 16, 2023
468ffc3
fix: preset
ishaansehgal99 Oct 16, 2023
716a519
fix: create path on pod
ishaansehgal99 Oct 16, 2023
75d30b2
fix: hostpath
ishaansehgal99 Oct 16, 2023
71f2758
fix: fix docker
ishaansehgal99 Oct 16, 2023
fb14672
fix: add docker
ishaansehgal99 Oct 16, 2023
34ac063
fix: filepath
ishaansehgal99 Oct 16, 2023
edae468
fix home filepaths
ishaansehgal99 Oct 16, 2023
325b9fe
Add docker push
ishaansehgal99 Oct 16, 2023
4017ead
Add
ishaansehgal99 Oct 16, 2023
413d1ed
fix: all docker filepaths weights folders, use local weights
ishaansehgal99 Oct 16, 2023
13c7a01
fix: falcon and llama paths
ishaansehgal99 Oct 16, 2023
d27942e
fix: remove uncessary go download and move common code to setup
ishaansehgal99 Oct 17, 2023
a731bdb
fix: check if ACR needs update
ishaansehgal99 Oct 17, 2023
dd48cd4
Merge branch 'main' into auto-image-build
ishaansehgal99 Oct 17, 2023
2eabf26
fix:retrigger
ishaansehgal99 Oct 17, 2023
6d827e3
fix: nodeselector
ishaansehgal99 Oct 17, 2023
cf3074a
fix: provide context for each runner
ishaansehgal99 Oct 17, 2023
8f96251
fix: provide context for each runner
ishaansehgal99 Oct 17, 2023
84f581f
move back
ishaansehgal99 Oct 17, 2023
fdd0bf9
fix: step name
ishaansehgal99 Oct 17, 2023
e06daff
fix: removed comments added the exclusion policies during build time
ishaansehgal99 Oct 17, 2023
a32c1a6
renaming
ishaansehgal99 Oct 17, 2023
71abf08
retry
ishaansehgal99 Oct 17, 2023
7bb3816
check
ishaansehgal99 Oct 17, 2023
47461a6
test
ishaansehgal99 Oct 17, 2023
d342311
test
ishaansehgal99 Oct 17, 2023
ea2ec07
jq trial
ishaansehgal99 Oct 17, 2023
2e31fcb
fix: json
ishaansehgal99 Oct 17, 2023
8809098
fix: include all the configs
ishaansehgal99 Oct 17, 2023
2418179
Merge branch 'main' into auto-image-build
ishaansehgal99 Oct 18, 2023
c692652
fix: artifact upload and download
ishaansehgal99 Oct 18, 2023
00fba8b
Merge branch 'auto-image-build' of https://github.com/Azure/kdm into …
ishaansehgal99 Oct 18, 2023
c6f0dcd
fix client id and image tag process for workflow dispatch
ishaansehgal99 Oct 18, 2023
6c81303
sudo
ishaansehgal99 Oct 18, 2023
826e205
sudo
ishaansehgal99 Oct 18, 2023
37c3737
fix
ishaansehgal99 Oct 18, 2023
5b5be3a
fix: resolve permissions issue and artifact name
ishaansehgal99 Oct 18, 2023
8572a12
fix: permissions fix
ishaansehgal99 Oct 18, 2023
fb40685
fix: permissions
ishaansehgal99 Oct 18, 2023
b44d9b2
test run trigger
ishaansehgal99 Oct 18, 2023
2134a71
Merge branch 'main' into auto-image-build
ishaansehgal99 Oct 18, 2023
4b7663e
retrigger larger images for running on e2e
ishaansehgal99 Oct 18, 2023
6799bae
Merge branch 'auto-image-build' of https://github.com/Azure/kdm into …
ishaansehgal99 Oct 18, 2023
d79e017
fix: uncomment e2e
ishaansehgal99 Oct 18, 2023
4b29c27
feat: added falcon 40b for e2e
ishaansehgal99 Oct 18, 2023
9fab5e5
fix file naming
ishaansehgal99 Oct 18, 2023
ac2366a
uncomment additional e2e
ishaansehgal99 Oct 18, 2023
9861c6e
check if image exists for running
ishaansehgal99 Oct 18, 2023
4783450
nit
ishaansehgal99 Oct 18, 2023
8cb38c7
fix: add safeguards to cleanup
ishaansehgal99 Oct 18, 2023
bbf0aa2
fix: add nodepool deletion safeguard
ishaansehgal99 Oct 18, 2023
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
100 changes: 100 additions & 0 deletions .github/matrix-configs.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,100 @@
[
{
"model": {
"runs_on": "self-hosted",
"name": "falcon-7b",
"dockerfile": "docker/presets/falcon/Dockerfile",
"build_args": "--build-arg FALCON_MODEL_NAME=tiiuae/falcon-7b"
},
"shouldBuildFalcon": "true"
},
{
"model": {
"runs_on": "self-hosted",
"name": "falcon-7b-instruct",
"dockerfile": "docker/presets/falcon/Dockerfile",
"build_args": "--build-arg FALCON_MODEL_NAME=tiiuae/falcon-7b-instruct"
},
"shouldBuildFalcon": "true"
},

{
"model": {
"runs_on": "self-hosted",
"name": "falcon-40b",
"dockerfile": "docker/presets/falcon/Dockerfile",
"build_args": "--build-arg FALCON_MODEL_NAME=tiiuae/falcon-40b"
},
"shouldBuildFalcon": "true"
},

{
"model": {
"runs_on": "self-hosted",
"name": "falcon-40b-instruct",
"dockerfile": "docker/presets/falcon/Dockerfile",
"build_args": "--build-arg FALCON_MODEL_NAME=tiiuae/falcon-40b-instruct"
},
"shouldBuildFalcon": "true"
},

{
"model": {
"runs_on": "self-hosted",
"name": "llama-2-7b",
"dockerfile": "docker/presets/llama-2/Dockerfile",
"build_args": "--build-arg LLAMA_VERSION=llama-2-7b --build-arg SRC_DIR=/home/presets/llama-2"
},
"shouldBuildLlama2": "true"
},

{
"model": {
"runs_on": "self-hosted",
"name": "llama-2-13b",
"dockerfile": "docker/presets/llama-2/Dockerfile",
"build_args": "--build-arg LLAMA_VERSION=llama-2-13b --build-arg SRC_DIR=/home/presets/llama-2"
},
"shouldBuildLlama2": "true"
},

{
"model": {
"runs_on": "self-hosted",
"name": "llama-2-70b",
"dockerfile": "docker/presets/llama-2/Dockerfile",
"build_args": "--build-arg LLAMA_VERSION=llama-2-70b --build-arg SRC_DIR=/home/presets/llama-2"
},
"shouldBuildLlama2": "true"
},

{
"model": {
"runs_on": "self-hosted",
"name": "llama-2-7b-chat",
"dockerfile": "docker/presets/llama-2/Dockerfile",
"build_args": "--build-arg LLAMA_VERSION=llama-2-7b-chat --build-arg SRC_DIR=/home/presets/llama-2-chat"
},
"shouldBuildLlama2Chat": "true"
},

{
"model": {
"runs_on": "self-hosted",
"name": "llama-2-13b-chat",
"dockerfile": "docker/presets/llama-2/Dockerfile",
"build_args": "--build-arg LLAMA_VERSION=llama-2-13b-chat --build-arg SRC_DIR=/home/presets/llama-2-chat"
},
"shouldBuildLlama2Chat": "true"
},

{
"model": {
"runs_on": "self-hosted",
"name": "llama-2-70b-chat",
"dockerfile": "docker/presets/llama-2/Dockerfile",
"build_args": "--build-arg LLAMA_VERSION=llama-2-70b-chat --build-arg SRC_DIR=/home/presets/llama-2-chat"
},
"shouldBuildLlama2Chat": "true"
}
]
Loading
Loading