yiakwy-xpu-ml-framework-team

B.t.w I just found this line when I scan the dataset file: > resized_image = resized_image.astype('float32') The image can be effectively worked with uint8, this allows 8-bit IO workload and...

> @yiakwy-xpu-ml-framework-team you'll need to create a PR, target for dygraph branch. we decided not merging this into release/2.7. I guess I can cherry-pick this commit onto the dygraph branch....

@tink2123 det_cml is just teacher - student (x2) distillation (with KL loss to make the smaller students mimic the teacher output). So you can simply use det_student for baseline. The...

@zhengwy888 @yuvalkirstain bf has already been supported in apex aten::Tensor api https://github.com/NVIDIA/apex/pull/1407/files Hence I believe this issue should be closed ?

> Ok, I've done some debugging, this function is literally designed to allocate GPU memory as max as possible (up to what's set in the `gpu_memory_utilization`). What's the reasoning behind...

您好，邮箱主人会认真阅读！谢谢关注/

Could you update gcc, cuda to gcc-9(10, 11 are also welcome), details can be found in TensorRT OSS repository docker file. gcc-6 is definitely not the right version for Ubuntu-18.04,...

您好，邮箱主人会认真阅读！谢谢关注/

@1049451037 This simply because you were using a m-core model (m-core model has bugs). You switch to legacy model : --use-legacy-model (now m-core is the default)

@htyu @ThomasRaoux is the optimization still on the menu? I am just learning how to auto coalesce global access to SMEM (to make sure data load store continuously). Is there...