-
Notifications
You must be signed in to change notification settings - Fork 53
Conversation
The test failure seems not related to PR. |
I will take a look at the test |
Thanks :) |
Otherwise, LGTM :) |
Codecov Report
@@ Coverage Diff @@
## master #103 +/- ##
==========================================
+ Coverage 58.70% 59.17% +0.46%
==========================================
Files 99 100 +1
Lines 5936 6075 +139
==========================================
+ Hits 3485 3595 +110
- Misses 2103 2122 +19
- Partials 348 358 +10
Continue to review full report at Codecov.
|
@swiftdiaries sorry for the late reply. |
LGTM |
This is really cool 😬 Great to see this come along |
Hey I thought this was merged. One check seems |
Test failure seems unrelated. |
return nil, flyteerr.Errorf(flyteerr.BadTaskSpecification, "Unable to create pod spec: [%v]", err.Error()) | ||
} | ||
|
||
overrideDefaultContainerName(taskCtx, podSpec) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Do you need to do this?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Tensorflow operator will also force pod to have a container named tensorflow
, I've updated the comment.
hmm ya the test is not related, it is interesting to me that your branch has this error, but master does not? |
I've moved some common code between PyTorch and TensorFlow to |
I got the same error in master |
Let me review today and we can merge soon |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
Some files are not go-imported. Once they are done, the linter should be green and let us merge it afterwards. I do not have permissions to your fork. |
Codecov Report
@@ Coverage Diff @@
## master #103 +/- ##
==========================================
- Coverage 61.98% 60.01% -1.98%
==========================================
Files 109 129 +20
Lines 6256 6955 +699
==========================================
+ Hits 3878 4174 +296
- Misses 1984 2361 +377
- Partials 394 420 +26
Flags with carried forward coverage won't be shown. Click here to find out more.
Continue to review full report at Codecov.
|
cc @EngHabu @swiftdiaries @kumare3 Can you help me review this, Thanks. However, I don't why
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Awesome work, @pingsutw. Just have a few comments and we can merge!
@EngHabu Thanks for your review. I will update ASAP. |
Co-authored-by: Haytham Abuelfutuh <[email protected]>
I think I am +1 on the issue, will wait for @EngHabu to approve. |
Let's address this: #103 (comment) then let's get it merged! |
okay, I'm working on it. |
* Tensorflow plugin implementaion * Fix checkstyle * Update go/tasks/plugins/k8s/kfoperators/common/common_operator.go Co-authored-by: Haytham Abuelfutuh <[email protected]> * Address comments * Fix lint error * Address comment * Fix lint * Update to running Co-authored-by: Haytham Abuelfutuh <[email protected]> Signed-off-by: Katrina Rogan <[email protected]>
* Tensorflow plugin implementaion * Fix checkstyle * Update go/tasks/plugins/k8s/kfoperators/common/common_operator.go Co-authored-by: Haytham Abuelfutuh <[email protected]> * Address comments * Fix lint error * Address comment * Fix lint * Update to running Co-authored-by: Haytham Abuelfutuh <[email protected]>
TL;DR
TensorFlow plugin for Flyte that leverages https://github.com/kubeflow/tf-operator to run distributed tensorflow jobs on k8s.
Type
Are all requirements met?
Complete description
related to flyteorg/flyteidl#71
Tracking Issue
flyteorg/flyte#115
Follow-up issue
NA