Why getting Execute failed: Lenght mus be >= 1 in Word Vector Apply node

Using same data by a Partitioning node, when I try to use Word Vector Apply with Calculate Document Mean Vector? checked, I'm getting this Execute failed: Lenght must be >= 1.

I'm trying to do the same as in this example.

May it something about the Classification set having some rows with just words that are not in my Vocabulary Extractor ? How could I solve it ?

This question is a duplicate of http://stackoverflow.com/questions/41002897/why-getting-execute-failed-lenght-mus-be-1-in-word-vector-apply-node

I thought I could get more help in here.

Hi wesley,

can you share a workflow. The example on the example server workfs fine for me.

Cheers, Kilian

Kilian,

The example worked to me too.

You may find my workflow on stackoverflow post.

Hi Wesley,

I could see the picture of the workflow but could you please export the workflow and attache the file. This make debugging the workflow much easier. I assume the node fails because of the missing values ("?") in your data tables but am I am not sure. It would be usefule to have a closer look at the workflow and the data.

You can replace missing value using the Missing Values node or filter them out using the Row Filter node.

Cheers, Kilian