Fancy data functions that will make your life as a data scientist easier.
Project description
WhiteBox Utilities Toolkit: Tools to make your life easier
Fancy data functions that will make your life as a data scientist easier.
Installing
To install this library in your Python environment:
pip install whiteboxml
Documentation
Metrics
Classification
- ROC curve / AUC:
import numpy as np
from whiteboxml.modeling.metrics import plot_roc_auc_binary
y_pred = np.random.normal(0, 1, 1000)
y_true = np.random.choice([0, 1], 1000)
ax, fpr, tpr, thr, auc_score = plot_roc_auc_binary(y_pred=y_pred, y_true=y_true, figsize=(8, 8))
ax.get_figure().savefig('roc_curve.png')
- Confusion Matrix:
import numpy as np
from whiteboxml.modeling.metrics import plot_confusion_matrix
y_true = np.random.choice([0, 1, 2, 3], 10000)
y_pred = np.random.choice([0, 1, 2, 3], 10000)
ax, matrix = plot_confusion_matrix(y_pred=y_pred, y_true=y_true,
class_labels=['a', 'b', 'c', 'd'])
ax.get_figure().savefig('confusion_matrix.png')
- Optimal Threshold:
import numpy as np
from whiteboxml.modeling.metrics import get_optimal_thr
y_pred_proba = np.random.normal(0, 1, (100, 1))
y_true = np.random.choice([0, 1], (100, 1))
thr = get_optimal_thr(y_pred=y_pred_proba, y_true=y_true)
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
whiteboxml-0.0.4.tar.gz
(16.9 kB
view hashes)
Built Distribution
whiteboxml-0.0.4-py3-none-any.whl
(17.4 kB
view hashes)
Close
Hashes for whiteboxml-0.0.4-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | cfa62cb20cd5845623581cfa70d68fe9f62722b1ab9949af15fb1057d1c5c1c4 |
|
MD5 | 4fef060cdb2b20a2a0f20366bfe46db6 |
|
BLAKE2b-256 | 08281024ca33d635d2ff14fcbc1d462f0114707176e27f7b66d218ff5a9ce387 |