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.
Install the most recent stable version on pypi:
pip install danlingInstall the latest version from source:
pip install git+https://github.com/ZhiyuanChen/DanLingIt works the way it should have worked.
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