Tuner

🚨 Experimental¶

`automodel` ¶

`AutoModelV2` ¶

Searches Hyperparameter

Examples:

tuner = Tuner()
cnns = tuner.suggest_complex("learner", cnn1, cnn2)
optimizers = tuner.choice("optimizer", "adam", "sgd")
loss = "crossentropyloss"
model = AutoModelV2(cnns)
model.hp_tune(tuner, autodataset, max_epochs=10)

Parameters:

Name	Description	Default
`learner`	tuner.	required
`optimization_metric`	metric on which to optimize model on	required
`mode`	max or min for optimization_metric	required

`tuner` ¶

`ComplexObject` ¶

`to_choice(self)` ¶

converts to ray.tune Domain

`Tuner` ¶

`choice(self, key, *values)` ¶

Tune for categorical values

`get_complex_object(self, key, idx)` ¶

Get registered complex object value from key at given index

`union(self, tuner)` ¶

Inplace Merge of two Tuners

Last update: September 29, 2021

Tuner

🚨 Experimental¶

automodel ¶

AutoModelV2 ¶

tuner ¶

ComplexObject ¶

to_choice(self) ¶

Tuner ¶

choice(self, key, *values) ¶

get_complex_object(self, key, idx) ¶

union(self, tuner) ¶

Comments