Release CostNav artifacts (models, dataset) on Hugging Face

[NielsRogge]

Hi @hbseong97 🤗

Niels here from the open-source team at Hugging Face. I discovered your work on Arxiv and was wondering whether you would like to submit it to hf.co/papers to improve its discoverability. If you are one of the authors, you can submit it at https://huggingface.co/papers/submit.

The paper page lets people discuss about your paper and lets them find artifacts about it (your models, datasets or demo for instance), you can also claim the paper as yours which will show up on your public profile at HF, and add Github and project page URLs.

I noticed in your GitHub README that you are planning to release the imitation learning baseline checkpoints and the collected teleoperation dataset soon ("Coming up soon" / "What's next?"). It'd be great to make these available on the 🤗 hub to improve their discoverability and visibility within the AI and robotics community.

I also saw the make download-assets-hf command in your README, which suggests you are already using Hugging Face for simulation assets! It would be excellent to host the upcoming ML training data and model weights there as well to keep all project artifacts centralized.

Uploading models

See here for a guide: https://huggingface.co/docs/hub/models-uploading.

In this case, you could leverage the PyTorchModelHubMixin class which adds from_pretrained and push_to_hub to any custom nn.Module. Alternatively, one can leverage the hf_hub_download one-liner to download a checkpoint from the hub.

Uploading dataset

Would be awesome to make the teleoperation dataset available on 🤗 , so that people can do:

from datasets import load_dataset

dataset = load_dataset("your-hf-org-or-username/your-dataset")

See here for a guide: https://huggingface.co/docs/datasets/loading.

Besides that, there's the dataset viewer which allows people to quickly explore the first few rows of the data in the browser.

Let me know if you're interested or need any help regarding this!

Cheers,

Niels
ML Engineer @ HF 🤗

[hbseong97]

Hi @NielsRogge ,

Thanks for the interest from Hugging Face!

Unfortunately, our arXiv paper was released more than 14 days ago, so the system currently doesn't allow us to submit it to hf.co/papers. Since our goal is to build a large-scale, ever-evolving benchmark with many collaborators, we would love to hear if you have any alternative suggestions for improving our discoverability on the platform!

Regarding the assets—yes, you spotted it correctly! We are already hosting our simulation assets on HF here: https://huggingface.co/datasets/maum-ai/CostNav/tree/main.

We definitely plan to upload the models and datasets for our imitation learning baselines as soon as they are finalized.

Thanks again for reaching out!

Best,

Haebin
Senior AI Research Scientist @ WoRV, Maum.AI