[FLINK-9664] FlinkML Quickstart Loading Data section example doesn't work as described

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 1.5.0
• Fix Version/s: 1.7.0
• Component/s: Documentation, Library / Machine Learning
• Labels: documentation-update machine_learning ml pull-request-available
• Assignee: Rong Rong
• Reporter: Mano Swerts
• Created: 26/Jun/18 09:21
• Updated: 19/Sep/19 18:57
• Resolved: 14/Aug/18 10:13

Description

The ML documentation example isn't complete: https://ci.apache.org/projects/flink/flink-docs-release-1.5/dev/libs/ml/quickstart.html#loading-data

The referred section loads data from an astroparticle binary classification dataset to showcase SVM. The dataset uses 0 and 1 as labels, which doesn't produce correct results. The SVM predictor expects -1 and 1 labels to correctly predict the label. The documentation, however, doesn't mention that. The example therefore doesn't work without a clue why.

The documentation should be updated with an explicit mention to -1 and 1 labels and a mapping function that shows the conversion of the labels.

Issue Links

• links to: GitHub Pull Request #6425