mirror of
https://github.com/vale981/ray
synced 2025-03-09 21:06:39 -04:00
f96078687f
* [xgboost/release] Add GPU connect user test * Use scaling cluster * typo * Increase xgboost placement group timeout * Much higher timeout * Move os environment timeout * Move os environ * [dev] install xgboost-ray from master * GPU xgboost master * Remove master install after new xgboost release * Install latest * Add master test
115 lines
2.5 KiB
YAML
115 lines
2.5 KiB
YAML
- name: train_small
|
|
cluster:
|
|
app_config: app_config.yaml
|
|
compute_template: tpl_cpu_small.yaml
|
|
|
|
run:
|
|
use_connect: True
|
|
autosuspend_mins: 10
|
|
timeout: 600
|
|
prepare: python wait_cluster.py 4 600
|
|
script: python workloads/train_small.py
|
|
|
|
- name: train_moderate
|
|
cluster:
|
|
app_config: app_config.yaml
|
|
compute_template: tpl_cpu_moderate.yaml
|
|
|
|
run:
|
|
timeout: 600
|
|
prepare: python wait_cluster.py 32 600
|
|
script: python workloads/train_moderate.py
|
|
|
|
- name: train_gpu
|
|
cluster:
|
|
app_config: app_config_gpu.yaml
|
|
compute_template: tpl_gpu_small.yaml
|
|
|
|
run:
|
|
timeout: 600
|
|
prepare: python wait_cluster.py 5 600
|
|
script: python workloads/train_gpu.py
|
|
|
|
- name: train_gpu_connect_latest
|
|
cluster:
|
|
app_config: app_config_gpu.yaml
|
|
compute_template: tpl_gpu_small_scaling.yaml
|
|
|
|
run:
|
|
use_connect: True
|
|
timeout: 1200
|
|
script: python workloads/train_gpu_connect.py
|
|
|
|
- name: train_gpu_connect_master
|
|
cluster:
|
|
app_config: app_config_gpu_master.yaml
|
|
compute_template: tpl_gpu_small_scaling.yaml
|
|
|
|
run:
|
|
use_connect: True
|
|
timeout: 1200
|
|
script: python workloads/train_gpu_connect.py
|
|
|
|
- name: distributed_api_test
|
|
cluster:
|
|
app_config: app_config.yaml
|
|
compute_template: tpl_cpu_small.yaml
|
|
results:
|
|
|
|
run:
|
|
timeout: 600
|
|
prepare: python wait_cluster.py 4 600
|
|
script: python workloads/distributed_api_test.py
|
|
results: ""
|
|
|
|
- name: ft_small_elastic
|
|
cluster:
|
|
app_config: app_config.yaml
|
|
compute_template: tpl_cpu_small.yaml
|
|
|
|
run:
|
|
timeout: 900
|
|
prepare: python wait_cluster.py 4 600
|
|
script: python workloads/ft_small_elastic.py
|
|
results: ""
|
|
|
|
- name: ft_small_non_elastic
|
|
cluster:
|
|
app_config: app_config.yaml
|
|
compute_template: tpl_cpu_small.yaml
|
|
|
|
run:
|
|
timeout: 900
|
|
prepare: python wait_cluster.py 4 600
|
|
script: python workloads/ft_small_non_elastic.py
|
|
results: ""
|
|
|
|
- name: tune_small
|
|
cluster:
|
|
app_config: app_config.yaml
|
|
compute_template: tpl_cpu_small.yaml
|
|
|
|
run:
|
|
timeout: 600
|
|
prepare: python wait_cluster.py 4 600
|
|
script: python workloads/tune_small.py
|
|
|
|
- name: tune_32x4
|
|
cluster:
|
|
app_config: app_config.yaml
|
|
compute_template: tpl_cpu_moderate.yaml
|
|
|
|
run:
|
|
timeout: 900
|
|
prepare: python wait_cluster.py 32 600
|
|
script: python workloads/tune_32x4.py
|
|
|
|
- name: tune_4x32
|
|
cluster:
|
|
app_config: app_config.yaml
|
|
compute_template: tpl_cpu_moderate.yaml
|
|
|
|
run:
|
|
timeout: 900
|
|
prepare: python wait_cluster.py 32 600
|
|
script: python workloads/tune_4x32.py
|