v1alpha3: Add suggestion CRD #1
Conversation
* Adding initial v1alpha2 controller * Adding logs * Adding comments * Adding template functions for experiment * Adding error checks
Signed-off-by: Ce Gao <gaoce@caicloud.io>
Signed-off-by: Ce Gao <gaoce@caicloud.io>
|
@gaocegege: Adding the "do-not-merge/release-note-label-needed" label because no release-note block was detected, please follow our release note process to remove it. Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
caicloud-bot
added
do-not-merge/release-note-label-needed
|
/release-note-none |
caicloud-bot
added
release-note-none
|
/approve |
|
@gaocegege: you cannot LGTM your own PR. In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
caicloud-bot
added
the
approved
Signed-off-by: Ce Gao <gaoce@caicloud.io>
|
/cc @ddysher |
|
/lgtm |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: ddysher, gaocegege The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
|
|
||
| func init() { | ||
| // Register the types with the Scheme so the components can map objects to GroupVersionKinds and back | ||
| AddToSchemes = append(AddToSchemes, v1alpha2.SchemeBuilder.AddToScheme) |
There was a problem hiding this comment.
experiment: v1alpha3
trial: v1alpha2 (?)
suggestion: v1alpha2 (?)
There was a problem hiding this comment.
API
- trial v1alpha2
- experiment v1alpha2
- suggestion v1alpha2
GRPC API
- suggestion v1alpha3
controller
- trial v1alpha3
- experiment v1alpha3
- suggestion v1alpha3
| // Total number of trials to run. | ||
| MaxTrialCount int `json:"maxTrialCount,omitempty"` | ||
| // Max completed trials to mark experiment as succeeded | ||
| MaxTrialCount *int `json:"maxTrialCount,omitempty"` |
There was a problem hiding this comment.
Name as MinTrialCount? Minimum trail count for success..
BTW, what is the benefit of pointer *int rather than int, which is changed in several places?
| @@ -86,6 +90,9 @@ type ExperimentStatus struct { | |||
|
|
|||
| // How many trials are currently pending. | |||
| TrialsPending int `json:"trialsPending,omitempty"` | |||
|
|
|||
| // How many trials are currently running. | |||
| TrialsRunning int `json:"trialsRunning,omitempty"` | |||
There was a problem hiding this comment.
And where do we record periodical metric?
| Type SuggestionType `json:"type,omitempty"` | ||
|
|
||
| // Defines if the suggestion needs previous results. | ||
| NeedHistory bool `json:"needHistory,omitempty"` |
There was a problem hiding this comment.
If NeedHistory == true, what does it mean for ES and HP? For HP, it means it needs trial final metrics (opposite: random search), while in ES, it needs periodical metrics?
No description provided.