hyperactive 0.3.4

A hyperparameter optimization toolbox for convenient and fast prototyping

Hyperactive

A hyperparameter optimization toolbox for convenient and fast prototyping.

---

Overview | Performance | Installation | Examples | Hyperactive API

---

Overview:

• Optimize hyperparameters of machine- or deep-learning models
• Choose from a variety of different optimization techniques to improve your model
• Never lose progress of previous optimizations: Just pass one or more models as start points and continue optimizing
• Use transfer learning during the optimization process to build a more accurate model, while saving training and optimization time
• Utilize multiprocessing for machine learning or your gpu for deep learning models

Optimization Techniques	Supported Packages
Local Search: Hill Climbing Stochastic Hill Climbing Tabu Search (beta) Random Methods: Random Search Random Restart Hill Climbing Random Annealing Markov Chain Monte Carlo: Simulated Annealing Stochastic Tunneling Parallel Tempering Population Methods: Particle Swarm Optimizer Evolution Strategy Sequential Methods: Bayesian Optimization (beta)	Machine Learning: Scikit-learn XGBoost Deep Learning: Keras Distribution: Multiprocessing

Performance

The bar chart below shows, that the optimization process itself represents only a small fraction (<0.6%) of the computation time. The 'No Opt'-bar shows the training time of a default Gradient-Boosting-Classifier normalized to 1. The other bars show the computation time relative to 'No Opt'. Each optimizer did 30 runs of 300 iterations, to get a good statistic.

Installation

pip install hyperactive

Examples

Basic sklearn example:

from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split

from hyperactive import SimulatedAnnealingOptimizer

iris_data = load_iris()
X = iris_data.data
y = iris_data.target

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33)

# this defines the model and hyperparameter search space
search_config = {
    "sklearn.ensemble.RandomForestClassifier": {
        "n_estimators": range(10, 100, 10),
        "max_depth": [3, 4, 5, 6],
        "criterion": ["gini", "entropy"],
        "min_samples_split": range(2, 21),
        "min_samples_leaf": range(2, 21),
    }
}

Optimizer = SimulatedAnnealingOptimizer(search_config, n_iter=100, n_jobs=4)

# search best hyperparameter for given data
Optimizer.fit(X_train, y_train)

# predict from test data
prediction = Optimizer.predict(X_test)

# calculate accuracy score
score = Optimizer.score(X_test, y_test)

Example with a convolutional neural network in keras:

import numpy as np

from keras.datasets import mnist
from keras.utils import to_categorical

from hyperactive import RandomSearchOptimizer

(X_train, y_train), (X_test, y_test) = mnist.load_data()

X_train = X_train.reshape(60000, 28, 28, 1)
X_test = X_test.reshape(10000, 28, 28, 1)

y_train = to_categorical(y_train)
y_test = to_categorical(y_test)


# this defines the structure of the model and the search space in each layer
search_config = {
    "keras.compile.0": {"loss": ["categorical_crossentropy"], "optimizer": ["adam"]},
    "keras.fit.0": {"epochs": [20], "batch_size": [500], "verbose": [2]},
    "keras.layers.Conv2D.1": {
        "filters": [32, 64, 128],
        "kernel_size": range(3, 4),
        "activation": ["relu"],
        "input_shape": [(28, 28, 1)],
    },
    "keras.layers.MaxPooling2D.2": {"pool_size": [(2, 2)]},
    "keras.layers.Conv2D.3": {
        "filters": [16, 32, 64],
        "kernel_size": [3],
        "activation": ["relu"],
    },
    "keras.layers.MaxPooling2D.4": {"pool_size": [(2, 2)]},
    "keras.layers.Flatten.5": {},
    "keras.layers.Dense.6": {"units": range(30, 200, 10), "activation": ["softmax"]},
    "keras.layers.Dropout.7": {"rate": list(np.arange(0.4, 0.8, 0.1))},
    "keras.layers.Dense.8": {"units": [10], "activation": ["softmax"]},
}

Optimizer = RandomSearchOptimizer(search_config, n_iter=20)

# search best hyperparameter for given data
Optimizer.fit(X_train, y_train)

# predict from test data
prediction = Optimizer.predict(X_test)

# calculate accuracy score
score = Optimizer.score(X_test, y_test)

Hyperactive API

Classes:

HillClimbingOptimizer(search_config, n_iter, metric="accuracy", n_jobs=1, cv=5, verbosity=1, random_state=None, warm_start=False, memory=True, hyperband_init=False, eps=1, r=1e-6)
StochasticHillClimbingOptimizer(search_config, n_iter, metric="accuracy", n_jobs=1, cv=5, verbosity=1, random_state=None, warm_start=False, memory=True, hyperband_init=False)
TabuOptimizer(search_config, n_iter, metric="accuracy", n_jobs=1, cv=5, verbosity=1, random_state=None, warm_start=False, memory=True, hyperband_init=False, eps=1, tabu_memory=[3, 6, 9])

RandomSearchOptimizer(search_config, n_iter, metric="accuracy", n_jobs=1, cv=5, verbosity=1, random_state=None, warm_start=False, memory=True, hyperband_init=False)
RandomRestartHillClimbingOptimizer(search_config, n_iter, metric="accuracy", n_jobs=1, cv=5, verbosity=1, random_state=None, warm_start=False, memory=True, hyperband_init=False, n_restarts=10)
RandomAnnealingOptimizer(search_config, n_iter, metric="accuracy", n_jobs=1, cv=5, verbosity=1, random_state=None, warm_start=False, memory=True, hyperband_init=False, eps=100, t_rate=0.98)

SimulatedAnnealingOptimizer(search_config, n_iter, metric="accuracy", n_jobs=1, cv=5, verbosity=1, random_state=None, warm_start=False, memory=True, hyperband_init=False, eps=1, t_rate=0.98)
StochasticTunnelingOptimizer(search_config, n_iter, metric="accuracy", n_jobs=1, cv=5, verbosity=1, random_state=None, warm_start=False, memory=True, hyperband_init=False, eps=1, t_rate=0.98, n_neighbours=1, gamma=1)
ParallelTemperingOptimizer(search_config, n_iter, metric="accuracy", n_jobs=1, cv=5, verbosity=1, random_state=None, warm_start=False, memory=True, hyperband_init=False, eps=1, t_rate=0.98, n_neighbours=1, system_temps=[0.1, 0.2, 0.01], n_swaps=10)

ParticleSwarmOptimizer(search_config, n_iter, metric="accuracy", n_jobs=1, cv=5, verbosity=1, random_state=None, warm_start=False, memory=True, hyperband_init=False, n_part=4, w=0.5, c_k=0.5, c_s=0.9)
EvolutionStrategyOptimizer(search_config, n_iter, metric="accuracy", n_jobs=1, cv=5, verbosity=1, random_state=None, warm_start=False, memory=True, hyperband_init=False, individuals=10, mutation_rate=0.7, crossover_rate=0.3)

BayesianOptimizer(search_config, n_iter, metric="accuracy", n_jobs=1, cv=5, verbosity=1, random_state=None, warm_start=False, memory=True, hyperband_init=False)

General positional argument:

Argument	Type	Description
search_config	dict	hyperparameter search space to explore by the optimizer
n_iter	int	number of iterations to perform

General keyword arguments:

Argument	Type	Default	Description
metric	str	"accuracy"	metric for model evaluation
n_jobs	int	1	number of jobs to run in parallel (-1 for maximum)
cv	int	5	cross-validation
verbosity	int	1	Shows model and metric information
random_state	int	None	The seed for random number generator
warm_start	dict	None	Hyperparameter configuration to start from
memory	bool	True	Stores explored evaluations in a dictionary to save computing time
hyperband_init	int	False	Chooses better initial position by training on multiple random positions with smaller training dataset (split into int subsets)

Specific keyword arguments (hill climbing):

Argument	Type	Default	Description
eps	int	1	epsilon

Specific keyword arguments (stochastic hill climbing):

Argument	Type	Default	Description
eps	int	1	epsilon
r	float	1e-6	acceptance factor

Specific keyword arguments (tabu search):

Argument	Type	Default	Description
eps	int	1	epsilon
tabu_memory	list	[3, 6, 9]	length of short/mid/long-term memory

Specific keyword arguments (random restart hill climbing):

Argument	Type	Default	Description
eps	int	1	epsilon
n_restarts	int	10	number of restarts

Specific keyword arguments (random annealing):

Argument	Type	Default	Description
eps	int	100	epsilon
t_rate	float	0.98	cooling rate

Specific keyword arguments (simulated annealing):

Argument	Type	Default	Description
eps	int	1	epsilon
t_rate	float	0.98	cooling rate

Specific keyword arguments (stochastic tunneling):

Argument	Type	Default	Description
eps	int	1	epsilon
t_rate	float	0.98	cooling rate
gamma	float	1	tunneling factor

Specific keyword arguments (parallel tempering):

Argument	Type	Default	Description
eps	int	1	epsilon
t_rate	float	0.98	cooling rate
system_temps	list	[0.1, 0.2, 0.01]	initial temperatures (number of elements defines number of systems)
n_swaps	int	10	number of swaps

Specific keyword arguments (particle swarm optimization):

Argument	Type	Default	Description
n_part	int	1	number of particles
w	float	0.5	intertia factor
c_k	float	0.8	cognitive factor
c_s	float	0.9	social factor

Specific keyword arguments (evolution strategy optimization):

Argument	Type	Default	Description
individuals	int	10	number of individuals
mutation_rate	float	0.7	mutation rate
crossover_rate	float	0.3	crossover rate

General methods:

fit(self, X_train, y_train)

Argument	Type	Description
X_train	array-like	training input features
y_train	array-like	training target

predict(self, X_test)

Argument	Type	Description
X_test	array-like	testing input features

score(self, X_test, y_test)

Argument	Type	Description
X_test	array-like	testing input features
y_test	array-like	true values

export(self, filename)

Argument	Type	Description
filename	str	file name and path for model export