Ensemble model#1535
Conversation
| @Inject | ||
| lateinit var config: EMConfig | ||
|
|
||
| @Inject |
There was a problem hiding this comment.
why are you using @Inject in this class? this class is not instantiated with an injector, isn't it?
There was a problem hiding this comment.
Yes, but I think we need it because the injected fields are populated afterward viatest.injector.injectMembers(test)
| * among all the other used models based on the average of key metrics. | ||
| * For the single-model scenario, the function simply returns the single model. | ||
| */ | ||
| private fun selectBestModel(endpoint: Endpoint): AIModel? { |
There was a problem hiding this comment.
should return AIModel not AIModel?
| // Return the single model if there is only one delegate. | ||
| if (delegates.size == 1) return delegates.first() | ||
|
|
||
| return delegates.maxByOrNull { model -> |
There was a problem hiding this comment.
if there was no delegate, ie delegates.isEmpty, then should had crashed at init time
|
|
||
| val bestModel = selectBestModel(input.endpoint) | ||
|
|
||
| val result = bestModel?.classify(input) ?: AIResponseClassification() |
There was a problem hiding this comment.
bestModel should never be null
| "Ensemble model is a combination of a comma-separated list, e.g., GLM,NN,KDE.") | ||
| var aiModelForResponseClassification: String = "NONE" | ||
|
|
||
| fun setAIModels(vararg models: AIResponseClassifierModel) { |
There was a problem hiding this comment.
these functions shouldn't mixed up in the list of Cfg declaration. move for example to bottom of file with the other support functions
| @Test | ||
| fun testRunDeterministic(){ | ||
| testRunEM(AIResponseClassifierModel.DETERMINISTIC) | ||
| testRunEM("DETERMINISTIC") |
There was a problem hiding this comment.
sorry, but don't like this. should avoid magic constant, and rather use enums. you can also have an overriden version of testRunEM where you take as input a list of enums, which then can convert internally to a join string with ,.
|
|
||
| @Test | ||
| fun testRunEnsemble(){ | ||
| testRunEM("GAUSSIAN,GLM,KDE,KNN,NN") |
There was a problem hiding this comment.
testRunEM(listOf(AIResponseClassifierModel.GAUSSIAN,AIResponseClassifierModel.GLM,AIResponseClassifierModel.KDE,AIResponseClassifierModel.KNN,AIResponseClassifierModel.NN))
can also statically import AIResponseClassifierModel.* to avoid repeating it
| @Cfg("Models used to learn input constraints and predict the response status before issuing a request. " + | ||
| "Supports both single-model and ensemble configurations. " + | ||
| "Ensemble model is a combination of a comma-separated list, e.g., GLM,NN,KDE.") | ||
| var aiModelForResponseClassification: String = "NONE" |
There was a problem hiding this comment.
hmmm... converting to string lose all information on type checking, and automated documentation of available entries in the generated docs. need to support list of enums. but maybe easier if i implement it
There was a problem hiding this comment.
Ok, then I’ll leave this part to you.
There was a problem hiding this comment.
@MohsenTaheriShalmani i ll do it on a different branch
2 participants
Implementation of the ensemble model