Benchmarks for different forms of target encoder with 10-fold cross validation.
| dataset_name | categorical feaatures | n_features | n_samples | is_classification | openml_url | |
|---|---|---|---|---|---|---|
| 0 | telco | 16 | 19 | 7043 | True | https://www.openml.org/d/42178 |
| 1 | amazon_access | 9 | 9 | 32769 | True | https://www.openml.org/d/4135 |
| 2 | kicks | 18 | 32 | 72983 | True | https://www.openml.org/d/41162 |
| 3 | taxi | 9 | 18 | 581835 | False | https://www.openml.org/d/42729 |
| 4 | ames | 43 | 79 | 1460 | False | https://www.openml.org/d/42165 |
| 5 | churn | 4 | 20 | 5000 | True | https://www.openml.org/d/40701 |
| 6 | adult | 12 | 14 | 48842 | True | https://www.openml.org/d/179 |
| 7 | dresses_sales | 11 | 12 | 500 | True | https://www.openml.org/d/23381 |
| 8 | phishing_websites | 30 | 30 | 11055 | True | https://www.openml.org/d/4534 |
- Clone repo:
git clone http://github.com/thomasjpfan/sk_encoder_cv
cd sk_encoder_cv- Create virtualenv and install
sk_encoder_cv
conda create -n sk_encoder_cv python=3.8 # or use venv
conda activate sk_encoder_cv
python setup.py develop- Run single benchmarks:
python benchmark.py --cv 5 --n-jobs 8 single adult SKTargetEncoder- Or run all benchmarks
python benchmark.py --cv 5 --n-jobs 8 allThe results will be written into the results directory.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent