Joe Chandler

Several outputs per input

4

We are working on building a model that outputs a list of outputs per input context. In LangChain, we can use output parsers to enforce list-like formats as follows: ```py...

`plot_tracks_v2` has bug when plotting with `trackgroup` argument.

2

I am running [this notebook](https://github.com/google-deepmind/tapnet/blob/main/colabs/tapir_clustering.ipynb) for RoboTAP clustering. After computing the clusters, I am running the following cell: ```py separation_visibility_trim = clustered['separation_visibility'] separation_tracks_trim = clustered['separation_tracks'] pointtrack_video = viz_utils.plot_tracks_v2( (demo_videos[demo_episode_ids[0]]).astype(np.uint8), separation_tracks_trim[demo_episode_ids[0]],...

TAPIR Checkpoint for Training/Finetuning

4

In the previous iteration of TAP-Net, there was a checkpoint file released that had not only model state but also the optimizer state as well as the `global_step`. This was...

Training on a new dataset

3

I am looking to train TAP-Net on a new dataset, in particular modifying the existing datasets (DAVIS, Kubric, RGB Stacking, etc.) to use new keypoints that we generate. It is...

`RAG.search` is not thread-safe

1

From the digging that I've done, it appears that the following code is _not thread safe_ (using Langchain): ```py RAG = RAGPretrainedModel.from_index(index_path) retriever = RAG.as_langchain_retriever(k=k) chain = RunnablePassthrough.assign( passages=itemgetter("query") |...