search

databricks / spark-deep-learning

Deep Learning Pipelines for Apache Spark
https://databricks.github.io/spark-deep-learning
Apache License 2.0
1.99k stars 494 forks source link

[ML-5678] Use `conda activate` in docker for best compatibility #169

Closed mengxr closed 5 years ago

mengxr commented 5 years ago

Use "conda activate" for the best compatibility with conda. This PR also made the following changes:

  1. Set /mnt/sparkdl as the WORKDIR in Docker.
  2. Mount ~/.ivy2 to accelerate sbt build in docker.
jkbradley commented 5 years ago

LGTM I mainly checked the test logs vs other logs, plus some travis & docker docs for reference.

I'll merge this now.