Create new project and virtual environment
mkdir new-project
cd new-project
python3 -m venv auto-sklearn-venv
source auto-sklearn-venv/bin/activate
Install auto-sklearn with SMOTE
git clone --recurse-submodules https://github.com/Karilli/auto-sklearn.git
cd auto-sklearn
git checkout SMOTE-version-6
pip install -e " .[test,doc,examples]"
Check whether the installation was successful
Create a python script that uses auto-sklearn with SMOTE
Copy-paste the following code into example.py
from sklearn .datasets import load_breast_cancer
from sklearn .model_selection import train_test_split
from autosklearn .classification import AutoSklearnClassifier
X , y = load_breast_cancer (return_X_y = True )
X_train , X_test , y_train , y_test = train_test_split (
X , y , random_state = 1
)
automl = AutoSklearnClassifier (
time_left_for_this_task = 120 ,
include = {"balancing" : ["SVMSMOTE" ]},
per_run_time_limit = 30 ,
initial_configurations_via_metalearning = 0 ,
).fit (X_train , y_train , dataset_name = "breast_cancer" )
print (automl .show_models ())
Run the script, you should see SVMSMOTE in all balancing steps