Skip to main content

A CascadeForestClassifier extension of AutoSklearn classifier

Project description

Deep Forest for Auto-Sklearn

Extension of AutoSklearnClassifier with DF21 - CascadeForestClassifier - a Deep Forest implementantion. Based on an example extension from Auto-Sklearn documentation.

Requirements

  • Linux operating system (requirement of Auto-Sklearn)
  • numpy version <= 1.19
  • installation of CascadeForestClassifier
  • installation of Auto-Sklearn
  • input variables have to be converted to numeric without missing values
  • output variables also should be converted to numeric values (you can use sklearn LabelEncoder)

Example use

# import libraries
import sklearn.metrics
import autosklearn.classification
import autosklearn.pipeline.components.classification


# import DFClassifier
from df_autosk.df_autosk import DFClassifier

# add DFClassifier to autosklearn classifier
autosklearn.pipeline.components.classification.add_classifier(DFClassifier)

# initialize and get hyperparameter search space
cs = DFClassifier.get_hyperparameter_search_space()
print(cs)

# load the dataset
from sklearn.model_selection import train_test_split
from sklearn.datasets import load_breast_cancer
X, y = load_breast_cancer(return_X_y=True)
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)

# use the classifier
clf = autosklearn.classification.AutoSklearnClassifier(
        	time_left_for_this_task=5400,
        	include={"classifier": ['DFClassifier']},
        	initial_configurations_via_metalearning=0,
        	memory_limit = 102400,
                # Not recommended for a real implementation
               	smac_scenario_args={"runcount_limit": 2}
                )

clf.fit(X_train, y_train)

#get result
y_pred = clf.predict(X_test)
print("accuracy: ", sklearn.metrics.accuracy_score(y_pred, y_test))

Example based on an example extension from Auto-Sklearn documentation.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

df_autosk-0.0.4.tar.gz (4.4 kB view hashes)

Uploaded Source

Built Distribution

df_autosk-0.0.4-py3-none-any.whl (4.9 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page