如何合并 tf.data.Dataset 的元素？

Question

A quick disclaimer: this is not a noob question about tf.data.Dataset.concat() , tf.data.Dataset.zip() , or tf.data.Dataset.interleave() . This is about combining the rows of a single dataset, not merging two seperate datasets.

I wanted to build a Named Entity Recognition (NER) model in TensorFlow. I found This Dataset for this task. But each word is on a separate row of the dataset, and I wanted to combine all of the words in a sentence into their own row (It would be more like [Sentence 1, Sentence 2, ...] instead of [Sentence 1, Sentence 1, Sentence 1, ...] ) I don't know if I'm supposed to do NER that way, but I don't know how to merge the rows of the dataset. I did try, but as you can see in this gist , it gives me a strange error. How would I merge the rows?

Answer 1

Yes, For Named Entity Recognition (NER) dataset looks exactly same as in This Dataset . In the dataset each word is annoated with person, location,organization, time etc.
For NER the above dataset can be used as it is without any modification in the rows or columns. If you still would like to merge/concatenate the rows, please let us know we will help you to resolve. Thanks