Boost Loss

Supported Python versions License

Utilities for easy use of custom losses in CatBoost, LightGBM, XGBoost. This sounds very simple, but in reality it took a lot of work.

Installation

Install this via pip (or your favourite package manager):

pip install boost-loss

Usage

Basic Usage

import numpy as np

from boost_loss import LossBase
from numpy.typing import NDArray


class L2Loss(LossBase):
    def loss(self, y_true: NDArray, y_pred: NDArray) -> NDArray:
        return (y_true - y_pred) ** 2 / 2

    def grad(self, y_true: NDArray, y_pred: NDArray) -> NDArray: # dL/dy_pred
        return - (y_true - y_pred)

    def hess(self, y_true: NDArray, y_pred: NDArray) -> NDArray: # d^2L/dy_pred^2
        return np.ones_like(y_true)

import lightgbm as lgb

from boost_loss import apply_custom_loss
from sklearn.datasets import load_boston


X, y = load_boston(return_X_y=True)
apply_custom_loss(lgb.LGBMRegressor(), L2Loss()).fit(X, y)

Built-in losses are available. ¹

from boost_loss.regression import LogCoshLoss

`torch.autograd` Loss ²

import torch

from boost_loss.torch import TorchLossBase


class L2LossTorch(TorchLossBase):
    def loss_torch(self, y_true: torch.Tensor, y_pred: torch.Tensor) -> torch.Tensor:
        return (y_true - y_pred) ** 2 / 2

Contributors ✨

Thanks goes to these wonderful people (emoji key):

_34j
💻 🤔 📖

This project follows the all-contributors specification. Contributions of any kind welcome!

Inspired by orchardbirds/bokbokbok ↩
Inspired by TomerRonen34/treeboost_autograd ↩

Metadata

Release files for boost-loss 0.5.5

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for boost-loss 0.5.5
File	Size	Uploaded
boost_loss-0.5.5.tar.gz	19.8 kB	Jan 26, 2024	Details

Built distribution (wheel)

Table of built distributions (wheels) for boost-loss 0.5.5
File	Interpreter	ABI	Platform	Reset
boost_loss-0.5.5-py3-none-any.whl 20.9 kB Jan 26, 2024	Python 3	none	any	Details

Total release size: 40.7 kB

Release files / boost_loss-0.5.5.tar.gz

Download URL	boost_loss-0.5.5.tar.gz
Size	19.8 kB
Tags	Source
SHA-256 checksum How to use checksums	`b301e6a8b711a9ef40c02fce478426fcbf7eeba9073b9a6c2dc5d859230baf05`
BLAKE2b-256 checksum How to use checksums	`aecf849acb3e7bd97018d1c9764343e27debd89795e6f900cd89c30f978ab428`
Upload date	Jan 26, 2024
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/3.8.0 pkginfo/1.9.6 readme-renderer/42.0 requests/2.31.0 requests-toolbelt/1.0.0 urllib3/2.1.0 tqdm/4.66.1 importlib-metadata/7.0.1 keyring/24.3.0 rfc3986/2.0.0 colorama/0.4.6 CPython/3.10.13`

Release files / boost_loss-0.5.5-py3-none-any.whl

Download URL	boost_loss-0.5.5-py3-none-any.whl
Size	20.9 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`1967f08dc2f3c3cffdefdf996a1e9c1ddc69c37f41ea818b7cf6a50f35d3bb35`
BLAKE2b-256 checksum How to use checksums	`8248ec77613130aa71b09ca48bbafc0486771542e1c8de666b0b44d713e647e0`
Upload date	Jan 26, 2024
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/3.8.0 pkginfo/1.9.6 readme-renderer/42.0 requests/2.31.0 requests-toolbelt/1.0.0 urllib3/2.1.0 tqdm/4.66.1 importlib-metadata/7.0.1 keyring/24.3.0 rfc3986/2.0.0 colorama/0.4.6 CPython/3.10.13`