Updated environment versions and instructions

Question

Updated environment versions and instructions

Opened this issue 7 years ago · 3 comments

Hi,
I'm trying to train my own dataset but failed to deploy after many different CUDA, CUDNN, CuPy, Python version trials on Ubuntu 16.04 on NVIDIA GTX-950M.

Is there any updated environment versions for CUDA, CUDNN, CuPy, Python, Chainer, NumPy?

Cheers,

Answer 1 · 2018-03-17T17:03:19.000Z

@Vol-i
Thank you for using my code. To be honest, currently, I don't have enough time to update the versions of packages used in this repo. How about using Nvidia docker to construct the environment? This comment might be useful: #20 (comment)

Answer 2 · 2018-03-17T18:31:04.000Z

Here's a Dockerfile created by @Skylion007: #5

Answer 3 · 2018-03-17T20:59:00.000Z

Thanks for the code and info, I'll give it a try.
Cheers!