Dit bericht verscheen eerder bij FOSSlife
PyTorch developers have announced version 2.0 of the popular open source machine learning framework, with a major new feature called torch.compile.
According to the announcement, the torch.compile feature “pushes PyTorch performance to new heights and starts the move for parts of PyTorch from C++ back into Python.”
With these improvements, “data scientists will be able to do with PyTorch 2.x the same things that they did with 1.x, but they can do them faster and at a larger scale,” said Soumith Chintala, PyTorch lead maintainer.
Dit bericht verscheen eerder bij FOSSlife
