Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!

1. The Module.__call__ method register all hooks and call the method Module.forward. In short, when you train the model you should use the method forward, while when you test the …

Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!

Open Neural Network Exchange (ONNX)

https://github.com/onnx

https://github.com/onnx/onnxmltools

Predictive Model Markup Language (PMML)

https://github.com/jpmml/sklearn2pmml

PMML4S is a PMML (Predictive Model Markup …

Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!

https://github.com/Azure/mmlspark/blob/master/docs/lightgbm.md

MMLSpark seems to be the best option to use train models using LightGBM on a Spark cluster. Note that MMLSpark requires …

Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!

Voice Web

Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!

Functions

torch.load

torch.from_numpy

torch.tensor

torch.as_tensor

torch.flatten

torch.numel

torch.utils.data.random_split

torch.nn.functional.relu

torch.nn.functional.leaky_relu

torch.nn.functional.gelu

torch.nn …

Things on this page are fragmentary and immature notes/thoughts of the author. Please read with your own judgement!

https://www.mathworks.com/company/newsletters/articles/what-is-int8-quantization-and-why-is-it-popular-for-deep-neural-networks.html

https://engineering.fb.com/ai-research/floating-point-math/

Rethinking floating point for deep learning

Training Deep Neural Networks with 8-bit Floating Point Numbers

8-Bit …