mirror of
https://github.com/vale981/ray
synced 2025-03-06 10:31:39 -05:00
b1308b1c8c
This fixes the previous problems from team column revert. This has 2 additional changes; alert handler receives the team argument, which was the root cause of breakage; https://github.com/ray-project/ray/pull/21289 Previously, tests without a team column were raising an exception, but I made the condition weaker (warning logs). I will eventually change it to raise an exception, but for smoother transition, we will log warning instead for a short time
11 lines
No EOL
334 B
YAML
11 lines
No EOL
334 B
YAML
# Test multi-node, multi-GPU Ray SGD example.
|
|
- name: sgd_gpu
|
|
team: ml
|
|
cluster:
|
|
app_config: sgd_gpu/sgd_gpu_app_config.yaml
|
|
compute_template: sgd_gpu/sgd_gpu_compute.yaml
|
|
|
|
run:
|
|
timeout: 3000
|
|
prepare: python wait_cluster.py 2 600
|
|
script: python sgd_gpu/sgd_gpu_test.py --num-workers=2 --use-gpu --address=auto |