The tophub package can’t be download, is it necessary for auto-tuning convolutional networks?
1 Like
I fall into the same question 。Although it has no influence on the runng of my program, I want to solve it!
I faced this warning too
I faced this warning too
Are you in China，I have resolved this warning by VNP
...100%, 0.02 MB, 49 KB/s, 0 seconds passed
it is VPN? emmm,what is the name of VPN?HALLO HALLO HALLO
You can git clone from https://github.com/uwsampl/tvm-distro and mv tophub/ to ~/.tvm/tophub.
4 Likes
wow，thank you！can i add your friends？My wechat is 17671689743.
but there is not better result to improve speed.
work for me! But i encounter a warning as below: Cannot find config for target=cuda -keys=cuda,gpu -max_num_threads=1024 -model=unknown -thread_warp_size=32, workload=(‘dense_small_batch.cuda’, (‘TENSOR’, (1, 512), ‘float32’), (‘TENSOR’, (1000, 512), ‘float32’), None, ‘float32’). A fallback configuration is used, which may bring great performance regression.
Excuse me,where should I put the “tophub” file?
Nice！！！Thanks！ I solved this problem.
for guys in china could use this repo: https://gitee.com/lanyuflying/tvm-distro
after cloned, in the tvm-distro dir, excute mv tophub/ ~/.tvm/