JohnSnowLabs/spark-nlp-workshop

Incorrect column name in explain-document-ml

davinro opened this issue · 1 comments

Description

In explain-document-ml.ipynb the code tries to read a column using the name "lemma", but the correct name is "lemmas".

The cell preceding that cell is throwing an exception in the committed code, as seen on the GitHub page for it.

The version shown in the run committed on GitHub is 2.4.2. Could you please try all the examples with the latest version?

Steps to Reproduce

Run the notebook up to the cell

result.select("lemma.result").show(1, False)

which will fail. Changing the cell to

result.select("lemmas.result").show(1, False)

succeeds.

Your Environment

  • Spark-NLP version: 2.4.5
  • Apache Spark version: 2.4.5
  • Operating System and version: Mint Linux, compatible with Ubuntu 18.04.
  • Deployment (Docker, Jupyter, Scala, pip, conda, etc.): Jupyter

Updated.