I had the exact same problem and after like 3 months, I finally found a
solution:
I assume you have installed CUDA and also exported `/usr/local/cuda/bin|lib64`
(-> `nvcc -V` works).
ValueError: arch(sm_xy) **is not passed**, and we cannot detect it from env
This suggests that we can pass it somehow, right? Yes, we can, but it is hidden
deep down in autoTVM:
```python
import tvm
tvm.autotvm.measure.measure_methods.set_cuda_target_arch("sm_62")
```
`"sm_62"` means that the target GPU has `tvm.gpu(0).compute_capability == 6.2`.
Adapt this if necessary, e.g. `compute_capability == 5.1` would be `"sm_51"`
---
[Visit
Topic](http://tracking.discuss.tvm.ai/tracking/click?d=r7Ovn2YCIazbb3o4uSOsltWIHVv_QgycVpSm7uw-ALP6eqevGntyLF0UjtAOn5MfpMpaqSOe-05nsyK95V0ELMVIO5VT1lEYjdeL4pllEzmyjHEy6dTnFMsuRazCsFnVtAkFL6LyFTp06FvE7TE5z-zV_zaEsSuqORPYFBziBg7IorWOdHayLqZMv5fyx0hYrQ2)
to respond.
You are receiving this because you enabled mailing list mode.
To unsubscribe from these emails, [click
here](http://tracking.discuss.tvm.ai/tracking/click?d=7cFgOaAA4XIBVlVKt_oyC07uihTjg4Q6cjeBRNRTiPrG3MbGTNsHPnIXUnAC8DPUy8KaoFtCzHMcHQWSi2YLArvwbeANzk23f58YDOmI954p4FRQIQe2YUI0mtRgLCIcjNmje-f5ZlXeQlG-eZvwHilLIGys--26YBxNr12Bq_ACkPzPbgoRDjk1wdJfSeCXnm5J-aGSg8JTQ4bsrVzYgOa79AFhlNRTCAe0gc0iMWqs0).
Tianqi Chen, UW, Seattle, WA, 98105, United States
http://tracking.discuss.tvm.ai/tracking/unsubscribe?msgid=vU0jN-V_3qzQ2vMtyg75-g2
