mdx-net
lightning-hydra-template
mdx-net | lightning-hydra-template | |
---|---|---|
3 | 9 | |
156 | 3,693 | |
0.0% | - | |
0.0 | 5.1 | |
about 1 year ago | 2 months ago | |
Python | Python | |
MIT License | MIT License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
mdx-net
-
Ultimate Vocal Remover is "holy sh*t" level good
UVR does include support for demucs (including the latest version, v4); however, the model that OP is recommending is mdx-net, which is a completely different AI model. mdx-net can produce superior vocal stems, but it can separate into only two stems (vocals, other).
-
(NameError: name 'trainer' is not defined) when i try to run auto_lr_find
The only thing i changed was auto_lr_find: False i set it to : True , here https://github.com/kuielab/mdx-net/blob/main/configs/trainer/minimal.yaml
lightning-hydra-template
- User-friendly PyTorch Lightning and Hydra template for ML experimentation
-
Best practice for saving logits/activation values of model in PyTorch Lightning
I've been trying to learn PyTorch Lightning and Hydra in order to use/create my own custom deep learning template (e.g. like this) as it would greatly help with my research workflow. A lot of the work I do requires me to analyse metrics based on the logits/activations of the model.
-
[D] Is Pytorch Lightning + Wandb a good combination for research?
I can't say for sure whether it is the best combination for research in the long run, but if you do go down that route I have found this template very useful
-
How research scientists structure their code ?
lightning-hydra-template
-
[D] Any research specific PyTorch based boilerplate code?
This lightning + hydra template is quite complete. Great for learning best practices.
-
Typing and testing for torch
A good example is this project template https://github.com/ashleve/lightning-hydra-template. It uses a lot of cool things such as
-
Our template to kickstart your pytorch projects, with list of best practices. Minimal boilerplate code. Leverages Lightning + Hydra. Focused on scalability, reproducibility and fast experimentation.
and many more! (checkout the #Your Superpowers section of the readme)
-
General and feature-rich PyTorch/Hydra project template for rapid and scalable ML experimentation, with a list of best practices
I write a LightningDatamodule. I found it to be an intuitive way to encapsulate any dataset. LightningDatamodule is a simple abstraction providing methods for data download, split, transforms and exposing dataloaders. Would love to see more researchers try out this concept, even in projects which don't use pytorch lightning. Reading LightningDatamodule makes me immedietely see how the dataset is prepared, while it seems like most data science projects throw around data logic across different parts of the pipeline, making it hard to understand what's going on. You can see example of such datamodule here
-
[P] General and feature-rich PyTorch/Hydra template for rapid and scalable ML research/experimentation, with a list of best practices
I feel like most ML people don't use those tools because they simply don't realize all the advantages (especially Hydra seems like a very useful addition to any deep learning project). I focused on structuring the readme in a way, which (I hope) will give you a quick overview - my hope is it can help to spread the word about those frameworks in a broaded community. It incorporates best practices and tricks I gathered over the last couple of months of playing around with it.
What are some alternatives?
demucs - Code for the paper Hybrid Spectrogram and Waveform Source Separation, but the goddamm motherfucker doesn't work.
lightning-hydra-template - Deep Learning project template best practices with Pytorch Lightning, Hydra, Tensorboard.
ultimatevocalremovergui - GUI for a Vocal Remover that uses Deep Neural Networks.
pytorch_tempest - My repo for training neural nets using pytorch-lightning and hydra
Pointnet2_PyTorch - PyTorch implementation of Pointnet2/Pointnet++
neptune-client - 📘 The MLOps stack component for experiment tracking
stemroller - Isolate vocals, drums, bass, and other instrumental stems from any song
lightning-transformers - Flexible components pairing 🤗 Transformers with :zap: Pytorch Lightning
spleeter - Deezer source separation library including pretrained models.
traingenerator - 🧙 A web app to generate template code for machine learning
nn-template - Generic template to bootstrap your PyTorch project.
neptune-contrib - This library is a location of the LegacyLogger for PyTorch Lightning.