Tuner
🚨 Experimental¶
automodel
¶
AutoModelV2
¶
Search Hyperparameter for your Model
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 | Type | Description | Default |
---|---|---|---|
learner |
tuner. |
required | |
optimization_metric |
metric on which to optimize model on |
required | |
mode |
max or min for optimization_metric |
required |
build_model(self, hparams, tuner)
¶
build and compile Model
State (Enum)
¶
An enumeration.
tuner
¶
ComplexObject
¶
Class to store and retrieve large size objects and convert it to ray.tune.Domain
.
Objects will be stored with ray.put
and retrieved with ray.get
.
to_choice(self)
¶
converts to ray.tune Domain
Tuner
¶
Supports ray.tune
methods and provide an easy way for tuning large size complex objects like Models.
choice(self, key, *values)
¶
Tune for categorical values
get_complex_object(self, key, idx)
¶
Get registered complex object value from key at given index
scalar(self, key, value)
¶
This sets a scalar value and will not be used for tuning
suggest_complex(self, key, *values)
¶
Use this method when you want to search models or any large object. It will also update search space with the provided key.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
key |
str |
hyperparameter name |
required |
*values |
Sequence |
values for the hyperparameter |
() |
Returns:
Type | Description |
---|---|
ComplexObject |
|
union(self, tuner)
¶
Inplace Merge of two Tuners
update_search_space(self, k, v)
¶
Update search space with value ray.tune(...)
or gradsflow.tuner.ComplexObject
Parameters:
Name | Type | Description | Default |
---|---|---|---|
k |
str |
hyperparameter name |
required |
v |
Union[ray.tune.sample.Domain, gradsflow.tuner.tuner.ComplexObject] |
hyperparameter value - |
required |