{"href":"https://api.simplecast.com/oembed?url=https%3A%2F%2Fpytorch-dev-podcast.simplecast.com%2Fepisodes%2Fdynamic-library-structure-kvyf_a_J","width":444,"version":"1.0","type":"rich","title":"Dynamic library structure","thumbnail_width":300,"thumbnail_url":"https://image.simplecastcdn.com/images/8cefde76-fb46-406a-8d87-ab0df67f3423/92f11400-2dad-49b4-8b14-cce35f5ab765/pytorch-symbol-02-orangeondark.jpg","thumbnail_height":300,"provider_url":"https://simplecast.com","provider_name":"Simplecast","html":"<iframe src=\"https://player.simplecast.com/cd27c78d-cc06-48c3-96d3-08b7561e3b41\" height=\"200\" width=\"100%\" title=\"Dynamic library structure\" frameborder=\"0\" scrolling=\"no\"></iframe>","height":200,"description":"Why is PyTorch split into so many libraries? What's the point of these splits? What do Windows, mobile and CUDA have to do with the library splits?"}