Code

Multi-host GPU tests on GKE#111

Code

Merged

will-cromar merged 30 commits intomasterGoogleCloudPlatform/ml-auto-solutions:masterfrom wcromar/multi-gpu-testGoogleCloudPlatform/ml-auto-solutions:wcromar/multi-gpu-testCopy head branch name to clipboardFeb 22, 2024

+406-50

Commits

Commits on Feb 15, 2024

• Multi-GPU tests on GKE

  Show description for 72c4afc
  will-cromar

  committed
  72c4afc

• Fit new test type into task API

  Show description for 6c4dcfd
  will-cromar

  committed
  6c4dcfd

• split up deploy_job

  Show description for dbb523f
  will-cromar

  committed
  dbb523f

• limit epochs

  Show description for 5eff685
  will-cromar

  committed
  5eff685

Commits on Feb 16, 2024

• Make cluster name a parameter

  Show description for 4842cdf
  will-cromar

  committed
  4842cdf

• remove terrform from this PR

  Show description for d3cd9c6
  will-cromar

  committed
  d3cd9c6

• use new cluster

  Show description for 57dc55d
  will-cromar

  committed
  57dc55d

• comment

  Show description for 074c4e2
  will-cromar

  committed
  074c4e2

• set timeout

  Show description for 9d74f72
  will-cromar

  committed
  9d74f72

• move to main dag

  Show description for 24d030c
  will-cromar

  committed
  24d030c

• format

  Show description for c2e0d3d
  will-cromar

  committed
  c2e0d3d

• factor out get_authenticated_client

  Show description for 3a75b0a
  will-cromar

  committed
  3a75b0a

• naming

  Show description for 61a28dd
  will-cromar

  committed
  61a28dd

• formatting

  Show description for bad8b38
  will-cromar

  committed
  bad8b38

• format with correct version

  Show description for b08623e
  will-cromar

  committed
  b08623e

Commits on Feb 20, 2024

• add commented dep

  Show description for c59a53f
  will-cromar

  committed
  c59a53f

• docstrings

  Show description for eebc648
  will-cromar

  committed
  eebc648

• update some comments

  Show description for 33a7de4
  will-cromar

  committed
  33a7de4

• implement timeout

  Show description for 191f582
  will-cromar

  committed
  191f582

• typo

  Show description for 9907376
  will-cromar

  committed
  9907376

• formatting

  Show description for bb27148
  will-cromar

  committed
  bb27148

• format with correct pyink version

  Show description for 4fa2653
  will-cromar

  committed
  4fa2653

• remove extra dag

  Show description for 3c55699
  will-cromar

  committed
  3c55699

• fix data type

  Show description for bd4099b
  will-cromar

  committed
  bd4099b

• formatting

  Show description for e3de889
  will-cromar

  committed
  e3de889

• return -> raise

  Show description for 163a809
  will-cromar

  committed
  163a809

Commits on Feb 22, 2024

• add myself to a TODO

  Show description for c11f120
  will-cromar

  committed
  c11f120

• TODO for big class

  Show description for ce82e82
  will-cromar

  committed
  ce82e82

• run_model

  Show description for 397d7cb
  will-cromar

  committed
  397d7cb

• Merge branch 'master' into wcromar/multi-gpu-test

  will-cromar

  authored
  b0ecd54