[SPARK-22666] Spark datasource for image format

• Type: Improvement
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 2.3.0
• Fix Version/s: 2.4.0
• Component/s: ML
• Target Version/s: 2.4.0
• Assignee: Weichen Xu
• Reporter: Timothy Hunter
• Shepherd: Xiangrui Meng
• Created: 30/Nov/17 22:07
• Updated: 11/Sep/19 01:11
• Resolved: 05/Sep/18 18:59

Description

The current API for the new image format is implemented as a standalone feature, in order to make it reside within the mllib package. As discussed in SPARK-21866, users should be able to load images through the more common spark source reader interface.

This ticket is concerned with adding image reading support in the spark source API, through either of the following interfaces:

• spark.read.format("image")...
• spark.read.image....
  The output is a dataframe that contains images (and the file names for example), following the semantics discussed already in SPARK-21866.

A few technical notes:

• since the functionality is implemented in mllib, calling this function may fail at runtime if users have not imported the spark-mllib dependency
• How to deal with very flat directories? It is common to have millions of files in a single "directory" (like in S3), which seems to have caused some issues to some users. If this issue is too complex to handle in this ticket, it can be dealt with separately.

Issue Links

• blocks: SPARK-25345 Deprecate readImages APIs from ImageSchema
• blocks: SPARK-25524 Spark datasource for image/libsvm user guide
• is duplicated by: SPARK-25157 Streaming of image files from directory
• is related to: SPARK-25349 Support sample pushdown in Data Source V2
• is related to: SPARK-25348 Data source for binary files
• relates to: SPARK-21866 SPIP: Image support in Spark
• links to: [Github] Pull Request #22328 (WeichenXu123)
• links to: GitHub Pull Request #24362