Custom train for the table tranformer
#4
by
arunavburagohain
- opened
Can anyone help me with the custom training of this table transformer to detect new table samples to increase the accuracy rate?
Yes, some guides you can use are:
- the object detection fine-tuning guide: https://huggingface.co/docs/transformers/tasks/object_detection
- this notebook illustrating how to fine-tune object detection models: https://github.com/NielsRogge/Transformers-Tutorials/blob/master/DETR/Fine_tuning_DetrForObjectDetection_on_custom_dataset_(balloon).ipynb.
Note that you need to replace the appropriate model and image processor, but other than that everything else is the same.
arunavburagohain
changed discussion status to
closed