chenhaoxing / DiffUTE

This repository is the code of our paper "DiffUTE: Universal Text Editing Diffusion Model" (NeurIPS'2023).
Apache License 2.0
106 stars 6 forks source link

Provide sample training data format #9

Open delonleo opened 5 months ago

delonleo commented 5 months ago

Thank you for the meaningful work here.

Could you please provide sample files for doc.csv and doc_select.csv with 1 line of placeholder data in each column? This will greatly assist us in comprehending the training data format. Much appreciated.

chenhaoxing commented 5 months ago

image

delonleo commented 5 months ago

Thank you @chenhaoxing. One last request. It will good if you could also share the JSON file sample.

Ruby-He commented 2 months ago

I found you also used a lot of public datasets, such as ICDAR series competitions, so could you give some detailed data formats of these data. This also can help us in comprehending the training data format.