sbodenstein commented on issue #9543: RNN Op Changes (was: Variable Length Support for cuDNN RNN) URL: https://github.com/apache/incubator-mxnet/issues/9543#issuecomment-425101837 @szha: I'm happy to make a PR. One question about the behaviour: the flag `MXNET_CUDA_ALLOW_TENSOR_CORE` is set to `true` by default. Its behaviour is: - Only use TensorCores if `DType` is `float16` and `MXNET_CUDA_ALLOW_TENSOR_CORE` is `true`. Presumably, the use of TensorCores will never worsen `float16` training, so defaulting to TensorCore use seems reasonable. - But if `DType` is `float32`, then we can't assume the user wants to use TensorCores. But they do want a way of opting-in (and setting `CUDNN_TENSOR_OP_MATH_ALLOW_CONVERSION` when available). So how about an environment variable `MXNET_CUDA_TENSOR_OP_MATH_ALLOW_CONVERSION`, which defaults to `false`, and if `true`, will let `float32` or `float64` nets use TensorCores by implicit downcasting?
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on 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
