mirror of
https://github.com/hanzo-ml/katib.git
synced 2026-09-24 00:31:27 +00:00
Mahdi KhashanandGitHub
19db2d47d4
* move to ghcr Signed-off-by: mahdikhashan <[email protected]> * move images to ghcr Signed-off-by: mahdikhashan <[email protected]> * manifests Signed-off-by: mahdikhashan <[email protected]> * change registry in all path Signed-off-by: mahdikhashan <[email protected]> * update script Signed-off-by: mahdikhashan <[email protected]> * fix Signed-off-by: mahdikhashan <[email protected]> * fix Signed-off-by: mahdikhashan <[email protected]> * slight fix Signed-off-by: mahdikhashan <[email protected]> --------- Signed-off-by: mahdikhashan <[email protected]> Signed-off-by: Mahdi Khashan <[email protected]>
54 lines
1.4 KiB
YAML
54 lines
1.4 KiB
YAML
---
|
|
apiVersion: kubeflow.org/v1beta1
|
|
kind: Experiment
|
|
metadata:
|
|
namespace: kubeflow
|
|
name: simple-pbt
|
|
spec:
|
|
maxTrialCount: 2
|
|
parallelTrialCount: 2
|
|
maxFailedTrialCount: 3
|
|
resumePolicy: FromVolume
|
|
objective:
|
|
type: maximize
|
|
goal: 0.99
|
|
objectiveMetricName: Validation-accuracy
|
|
algorithm:
|
|
algorithmName: pbt
|
|
algorithmSettings:
|
|
- name: suggestion_trial_dir
|
|
value: /var/log/katib/checkpoints/
|
|
- name: n_population
|
|
value: '40'
|
|
- name: truncation_threshold
|
|
value: '0.2'
|
|
parameters:
|
|
- name: lr
|
|
parameterType: double
|
|
feasibleSpace:
|
|
min: '0.0001'
|
|
max: '0.02'
|
|
step: '0.0001'
|
|
trialTemplate:
|
|
primaryContainerName: training-container
|
|
trialParameters:
|
|
- name: learningRate
|
|
description: Learning rate for training the model
|
|
reference: lr
|
|
trialSpec:
|
|
apiVersion: batch/v1
|
|
kind: Job
|
|
spec:
|
|
template:
|
|
spec:
|
|
containers:
|
|
- name: training-container
|
|
image: ghcr.io/kubeflow/katib/simple-pbt:latest
|
|
command:
|
|
- "python3"
|
|
- "/opt/pbt/pbt_test.py"
|
|
- "--epochs=20"
|
|
- "--lr=${trialParameters.learningRate}"
|
|
- "--checkpoint=/var/log/katib/checkpoints/"
|
|
restartPolicy: Never
|