PyWarm: A cleaner way to build neural networks for PyTorch

3 Likes

PyWarm is a high-level neural network construction API for PyTorch.

Using PyWarm, you can put all network data flow logic in the forward() method of your model, without the need to define children modules in the __init__() method then call it again in the forward(). This result in a much more readable model definition in fewer lines of code. Check the GitHub repository for code examples and more details!

2 Likes

PyWarm just hit v0.2!

Featuring the warm functional interface to Transformer, and a new example for MobileNet.

See an example of the transformer model (Attention is all you need) in less than 50 lines of code.