{"title":"PyVideo.org - AI deployment","link":[{"@attributes":{"href":"https:\/\/pyvideo.org\/","rel":"alternate"}},{"@attributes":{"href":"https:\/\/pyvideo.org\/feeds\/tag_ai-deployment.atom.xml","rel":"self"}}],"id":"https:\/\/pyvideo.org\/","updated":"2019-10-16T00:00:00+00:00","subtitle":{},"entry":{"title":"Research to Production: PyTorch JIT\/TorchScript Updates","link":{"@attributes":{"href":"https:\/\/pyvideo.org\/pytorch-conference-2019\/research-to-production-pytorch-jittorchscript-updates.html","rel":"alternate"}},"published":"2019-10-16T00:00:00+00:00","updated":"2019-10-16T00:00:00+00:00","author":{"name":"Michael Suo"},"id":"tag:pyvideo.org,2019-10-16:\/pytorch-conference-2019\/research-to-production-pytorch-jittorchscript-updates.html","summary":"<h3>Description<\/h3><p>TorchScript allows developers to create serializable and optimizable models from PyTorch code. Any TorchScript program can be saved from a Python process and loaded in a process where there is no Python dependency. Using TorchScript and the PyTorch JIT compiler, organizations and businesses have been able to use PyTorch \u2026<\/p>","content":"<h3>Description<\/h3><p>TorchScript allows developers to create serializable and optimizable models from PyTorch code. Any TorchScript program can be saved from a Python process and loaded in a process where there is no Python dependency. Using TorchScript and the PyTorch JIT compiler, organizations and businesses have been able to use PyTorch not just for state-of-the-art research, but also to more easily take their trained models to production.<\/p>\n","category":[{"@attributes":{"term":"PyTorch Conference 2019"}},{"@attributes":{"term":"AI"}},{"@attributes":{"term":"AI deployment"}},{"@attributes":{"term":"Artificial Intelligence"}},{"@attributes":{"term":"Facebook"}},{"@attributes":{"term":"JIT Compiler"}},{"@attributes":{"term":"ML"}},{"@attributes":{"term":"ML deployment"}},{"@attributes":{"term":"Machine Learning"}},{"@attributes":{"term":"Production AI"}},{"@attributes":{"term":"Production ML"}},{"@attributes":{"term":"PyTorch"}},{"@attributes":{"term":"PyTorch 1.3"}},{"@attributes":{"term":"TorchScript"}}]}}