0

I have recently built the h2o4gpu docker image using the Dockerfile-runtime, and managed to run it and log into the Jupyter notebooks.

However, when trying to run 

import h2o4gpu

I get the error that there is no h2o4gpu module. After, I tried installing by adding the below command to the dockerfile.

pip install --extra-index-url https://pypi.anaconda.org/gpuopenanalytics/simple h2o4gpu 
pip install h2o4gpu-0.2.0-cp36-cp36m-linux_x86_64.whl

This also failed, so I was wondering if there were other changes I should make, or if I should be making the docker file from scratch.

Thank you

strudlelion
  • 15
  • 3
  • The way we're making our docker images is a bit convoluted, I'm working this week on simplifying it and making it a bit more user friendly. For the time being have you tried one of the images we link to in the README on Github? – Mateusz Dymczyk Apr 18 '18 at 06:43

1 Answers1

0

To build the project, you can follow this recipe:

git clone https://github.com/h2oai/h2o4gpu.git
cd h2o4gpu
make centos7_cuda9_in_docker

This will work on either an x86_64 or ppc64le host with a modern docker installed. The python .whl file artifact is written to the dist directory.

Even if the build process is significantly refactored, this style of build API is very likely to remain.

TomKraljevic
  • 3,661
  • 11
  • 14
  • The make steps got further, but have prompted two more questions: 1. If I get "Step 1/55 : FROM undefined" is that because I am missing CUDA drivers? 2. Do I need the latest "wget", I am getting a SSL error on another machine when using wget (version 11.14) to sourceforge – strudlelion Apr 18 '18 at 17:06
  • The FROM is nvidia/cuda:9.0-cudnn7-devel-centos7 – TomKraljevic Apr 18 '18 at 17:59