MarisaKirisame opened a new issue #6136: URL: https://github.com/apache/incubator-tvm/issues/6136
I noticed that tvm only assume a single "main" entry in the code. I dont think this is a good design for training - when training a network, you will switch between training, validation, and maybe even pruning and quantization. If we do not allow multiple entry in tvm module, one will have to manage multiple module himself, which will result in code duplication and code complexity (one module to manage vs many module) then necessary. Can we support multi-entry? If so how much work is it? @tqchen @jroesch ---------------------------------------------------------------- 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]
