| dataset["train"].features | |
| Here's what the individual fields represent: | |
| * id: the example's id | |
| * image: a PIL.Image.Image object containing the document image | |
| * query: the question string - natural language asked question, in several languages | |
| * answers: a list of correct answers provided by human annotators | |
| * words and bounding_boxes: the results of OCR, which we will not use here | |
| * answer: an answer matched by a different model which we will not use here | |
| Let's leave only English questions, and drop the answer feature which appears to contain predictions by another model. |