combust / combust/mleap-docs

mnist example: serializing the pipeline

Open
#12 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Shell
Stars
15
Forks
21
PR merge metrics
No merged PRs in 30d

Description

Hi, I am running the mnist example on a free databricks account and I cannot get a valid bundle file for the pipeline. The code runs without any errors but the zip file is empty. Switching to a directory based serialization does seem to produce a valid export of the pipeline. The Databricks cluster is Spark 2.2.0, scala 2.11. What am I doing wrong? Thanks

Attached is the bundle zip file I get.
[mnist-spark-pipeline-01.zip](https://github.com/combust/mleap-docs/files/1766714/mnist-spark-pipeline-01.zip)

Contributor guide

No contributing guide indexed for this repository

Research direction

Start with the MNIST example and reproduce the empty bundle on the stated Spark 2.2.0 and Scala 2.11 Databricks setup. Compare its zip-based serialization with the directory-based export described in the issue; done means identifying the cause and documenting a verified resolution.

Written by the indexing model from the issue text.

Assessment

Tech stack
scala, spark
Domain
machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
30/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.