yzhliu commented on issue #1996: [RFC][WIP] Tensor Expression level automatic differentiation URL: https://github.com/apache/incubator-tvm/issues/1996#issuecomment-595904266 @MarisaKirisame could you elaborate "The jacobian of Y, W, should be dW times jacobian Y W" ? not sure I correctly understand the symbol you use. I think the main challenge is to infer the bound for Jacobian's axis, under the scenario where output axes can be arbitrary linear combination of its input tensors' axes
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
