lrx02

error in multiscale_kernel_attn_forward_cuda: an illegal memory access was encountered

2

I want to ask why there exists a cuda memeory error: multiscale_kernel_attn_forward_cuda

Training with a single RTX 3090

2

I trained the method on a single RTX 3090 card and could not achieve results in the paper. I have set args.autoscale_l = True, can you provide parameters for single...

Training with your configurations cannot reach the performance reported in the paper

2

I have trained the model on a single gpu with your latest configurations, but the performance is severely lower than that reported by your paper. I want to ask if...

Have you trained on single GPU?