Helpful functions for Data Science
Project description
DSfun
This package contains useful loss function class for training algorithms. These are f1-loss related functions. The main features:
- It is differentiable so it works with tensorflow
- It is more eficient than standard implementations
- Great for task that require to optimize F1-score
- Works with missing data *(TO DO)
- Can be modified to perform arbitrary differential functions on confusion matrix (TO DO)
Limitations:
- As any machine learning framework, this loss function shouldn't be used without proper validation as it is not deeply understood
- If calculating on batches, it will give a biased estimation of global loss
- If there are no representatives of a class in a batch, it might not converge properly
Instalation
pip install dsfun
Usage
TO DO
import tensorflow as tf
from dsfun import f1_loss, f1_score
y_true = tf.constant([[1.0, 0.0], [1.0, 1.0], [0.0, 1.0], [0.0, 1.0]])
y_pred = tf.constant([[0.5, 0.5], [0.5, 0.5], [1.0, 0.0], [0.0, 1.0]])
f1_loss(y_true, y_pred, 'macro')
> ?
f1_score(y_true, y_pred, 'macro')
> ?
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
dsfun-0.0.3.tar.gz
(3.0 kB
view hashes)
Built Distribution
dsfun-0.0.3-py3-none-any.whl
(4.9 kB
view hashes)