A CascadeForestClassifier extension of AutoSklearn classifier
Project description
DF for AutoSklearn
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-0.0.2.tar.gz
(4.4 kB
view hashes)
Built Distribution
Close
Hashes for df_autosk-0.0.2-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | bb8669002c45ac1f07603841a996df71c88f5c6ffe890a6185fcf44e6a9e1f25 |
|
MD5 | 341bd033250d7f2af0fc4b033443d651 |
|
BLAKE2b-256 | e909e3cd0f1209ea478589728ece0976f02c43fadc9e772ee8fce2d6020c6ac4 |