Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

feat: Support random state in random search #873

Merged
merged 2 commits into from
Oct 11, 2019
Merged

feat: Support random state in random search #873

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
dd259f0
feat: Support random search
gaocegege Oct 11, 2019
95a3ede
feat: Update docs
gaocegege Oct 11, 2019
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 7 additions & 2 deletions docs/algorithm-settings.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,7 +35,12 @@ Random sampling is an alternative to grid search when the number of discrete par

### [Hyperopt][]

Algorithm name in katib is `random`.
Algorithm name in katib is `random`, and there are some algortihm settings that we support:

| Setting Name | Description | Example |
|------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|----------|
| random_state | [int]: Set random state to something other than None for reproducible results. | 10 |


## TPE

Expand Down Expand Up @@ -75,7 +80,7 @@ Algorithm name in katib is `skopt-bayesian-optimization`, and there are some alg
| n_initial_points | [int, default=10]: Number of evaluations of `func` with initialization points before approximating it with `base_estimator`. Points provided as `x0` count as initialization points. If len(x0) < n_initial_points additional points are sampled at random. More in [skopt document](https://scikit-optimize.github.io/#skopt.Optimizer) | 10 |
| acq_func | [string, default=`"gp_hedge"`]: Function to minimize over the posterior distribution. More in [skopt document](https://scikit-optimize.github.io/#skopt.Optimizer) | gp_hedge |
| acq_optimizer | [string, "sampling" or "lbfgs", default="auto"]: Method to minimize the acquistion function. The fit model is updated with the optimal value obtained by optimizing acq_func with acq_optimizer. More in [skopt document](https://scikit-optimize.github.io/#skopt.Optimizer) | auto |
| random_state | [int, RandomState instance, or None (default)]: Set random state to something other than None for reproducible results. | 10 |
| random_state | [int]: Set random state to something other than None for reproducible results. | 10 |

## References

Expand Down
5 changes: 3 additions & 2 deletions pkg/suggestion/v1alpha3/hyperopt/base_hyperopt_service.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,7 +8,8 @@
logger = logging.getLogger("BaseHyperoptService")

class BaseHyperoptService(object):
def __init__(self, algorithm_name="tpe"):
def __init__(self, algorithm_name="tpe", random_state=None):
self.random_state = random_state
if algorithm_name == 'tpe':
self.hyperopt_algorithm = hyperopt.tpe.suggest
elif algorithm_name == 'random':
Expand Down Expand Up @@ -42,7 +43,7 @@ def getSuggestions(self, search_space, trials, request_number):
hyperopt_search_space[param.name] = hyperopt.hp.choice(
param.name, param.list)
# New hyperopt variables
hyperopt_rstate = np.random.RandomState()
hyperopt_rstate = np.random.RandomState(self.random_state)
hyperopt_domain = hyperopt.Domain(
None, hyperopt_search_space, pass_expr_memo_ctrl=None)

Expand Down
17 changes: 16 additions & 1 deletion pkg/suggestion/v1alpha3/hyperopt_service.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,12 +17,27 @@ def GetSuggestions(self, request, context):
"""
Main function to provide suggestion.
"""
name, config = OptimizerConfiguration.convertAlgorithmSpec(
request.experiment.spec.algorithm)
base_serice = BaseHyperoptService(
algorithm_name=request.experiment.spec.algorithm.algorithm_name)
algorithm_name=name, random_state=config.random_state)
search_space = HyperParameterSearchSpace.convert(request.experiment)
trials = Trial.convert(request.trials)
new_assignments = base_serice.getSuggestions(
search_space, trials, request.request_number)
return api_pb2.GetSuggestionsReply(
parameter_assignments=Assignment.generate(new_assignments)
)


class OptimizerConfiguration(object):
def __init__(self, random_state=None):
self.random_state = random_state

@staticmethod
def convertAlgorithmSpec(algorithm_spec):
optmizer = OptimizerConfiguration()
for s in algorithm_spec.algorithm_setting:
if s.name == "random_state":
optmizer.random_state = int(s.value)
return algorithm_spec.algorithm_name, optmizer