Selman Özleyen

hi @ivirshup , Besides a conversation, we didn't resolve about the warning/error on concatenation, I added all the necessary features and their tests. Here is a summary of what is...

Hi, thanks for the respond. Btw I love the work of you guys, just wanted to say... Anyways considering this: ```python r = M[i, :] - beta exp_beta = cp.exp(-r...

Hi @ivirshup, I will start working on this. To specify the requirements, I have some questions: - From what I understand, you'd like to let Dask handle the concatenation. Is...

> I don't think this function actually needs to take AnnData objects as arguments right now. I think this could start off with: `concatenate_from_disk(["adata1.h5ad", "adata2.h5ad"], out="result.h5ad")`. Ok, I think I...

> I'm not sure I'm exactly getting the question. What do you mean by "a pair" here? I meant a join pair. > I would assume so? I expect going...

@giovp can you have a look? after this I will switch to sparse geodesic

Ok just to update this issue waits for https://github.com/ott-jax/ott/issues/519 to be closed and then we'd need to **unskip** the test related in ```python class TestSpatioTemporalProblem: ... @pytest.mark.skip(reason="unbalanced does not work...

> Now I am wondering whether this scenario is realistic, as the joint attribute in the case you describe would be the full gene space (no dim. reduction). @MUCDK yeah...

hey @giovp @MUCDK, I ran some quick benchmark to capture the peak memory. From what I checked maximum memory allocated for `solve` is (`batch_size_point_cloud,d`), (`n,1`), or (`m,1`) and `apply_lse_kernel` contributes...

> So seems like `solve` and `apply` require the same memory? Could you please check by setting `batch_size_sparse` to 1? No, solve doesn't exceed 500mb. Here are the results with...