liuyunpu

I am using 1.7.0 taichi, but the problem can also be reproduced in 1.6.0

Hope this can be fixed soon! Transferring array to scipy from taichi is really slow

Is there any plan to support cdna or cdna2 architecture? I hope to run my taichi code on MI series gpu.

How? I thought the current taichi doesn't support amd gpu

> As a workaround, you may manually divide tasks and use process pool to launch mutilple kernels to utilize multi-cpu, which is how our project works for now. thanks, that...