In this directory, you can find notebooks that illustrate how to use Microsoft's GIT both for inference as well as fine-tuning on custom data. It currently includes the following notebooks:
- performing inference with GIT for image/video captioning as well as question answering (QA) on images/video.
- fine-tuning GIT on an image captioning dataset.
The docs can be found here: https://huggingface.co/docs/transformers/main/en/model_doc/git
All models can be found on the hub.