Scaffold for experienced Machine Learning Researchers
Project description
DanLing
Introduction
DanLing (丹灵) is a high-level library to help with running neural networks flexibly and transparently.
DanLing is meant to be a scaffold for experienced researchers and engineers who know how to define a training loop, but are bored of writing the same boilerplate code, such as DDP, logging, checkpointing, etc., over and over again.
Therefore, DanLing does not feature complex Runner designs with many pre-defined methods and complicated hooks. Instead, the Runner of DanLing just initialise the essential parts for you, and you can do whatever you want, however you want.
Although many attributes and properties are pre-defined and are expected to be used in DanLing, you have full control over your code.
DanLing also provides some utilities, such as [NestedTensor
][danling.NestedTensor], [LRScheduler
][danling.optim.LRScheduler], [catch
][danling.utils.catch], etc.
Installation
Install the most recent stable version on pypi:
pip install danling
Install the latest version from source:
pip install git+https://github.com/ZhiyuanChen/DanLing
It works the way it should have worked.
License
DanLing is multi-licensed under the following licenses:
- The Unlicense
- GNU Affero General Public License v3.0 or later
- GNU General Public License v2.0 or later
- BSD 4-Clause "Original" or "Old" License
- MIT License
- Apache License 2.0
You can choose any (one or more) of these licenses if you use this work.
SPDX-License-Identifier: Unlicense OR AGPL-3.0-or-later OR GPL-2.0-or-later OR BSD-4-Clause OR MIT OR Apache-2.0
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
File details
Details for the file danling-0.3.12.tar.gz
.
File metadata
- Download URL: danling-0.3.12.tar.gz
- Upload date:
- Size: 6.4 MB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.1.1 CPython/3.12.7
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 754ec38be317c1234a9bf295b265d7f38761f2137f66ef7bdf6c2ee64d331948 |
|
MD5 | 54fd7c6f44360bc2c196ccfb4649dad1 |
|
BLAKE2b-256 | af96b0655bc4551bc01105725f637b9efe697a194594b624cadddd007e736f00 |
File details
Details for the file danling-0.3.12-py3-none-any.whl
.
File metadata
- Download URL: danling-0.3.12-py3-none-any.whl
- Upload date:
- Size: 87.6 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.1.1 CPython/3.12.7
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | f39ee876618e0022f97a15d513b1e27f0e6dd542cecf81fe5ffc4716dcb9c2c9 |
|
MD5 | d3152ec5c80a35e572398b243e93762b |
|
BLAKE2b-256 | cc04dd990e4059b028b7031535c68b995377d6d42e7020b93617394125891504 |