Kai Zhang
Kai Zhang
Are you interested in developing a DSL on top of `funflow` to help user create complicated workflows? Here is a [real world workflow](https://github.com/Taiji-pipeline/Taiji/blob/master/Taiji.png) which is hard to specify by hand...
I add a simple test to my fork. It passes on my local computer but fails on many platforms, see CI results here: https://github.com/kaizhang/hdf5-rust/actions/runs/2168628529 The test code is: ```rust #[cfg(test)]...
Version: 0.8.0rc1 Example: ```python >>> import anndata as ad >>> a = ad.AnnData(shape=(10, 20)) >>> a AnnData object with n_obs × n_vars = 10 × 20 >>> a.write("1.h5ad") >>> b...
The HDF5 file object is not explicitly closed here. In some cases, it causes the file to be locked even after `adata.file.close()`. It can be fixed by using the context...
When `isbacked == True and is_view == True` and the destination file is the same file, anndata will skip writing the `X` and cause unexpected behaviors. ## Code to reproduce:...
SnapATAC2 uses the backed AnnData to save memory and speed up the processing of large files. But for small datasets, the backed AnnData is less convenient to work with than...
- [ ] Improve the API documentation - [ ] Add descriptions of the algorithms used in the package
## 1: Make similarity matrix calculation more time- and space-efficient We can use the approximate nearest neighbor search to speed up and sparsify the graph Laplacian. Here are some good...
- [x] add an option to filter peaks based on blacklist. - [ ] provide additional peak filtering/correction methods, such as the score per million correction. - [x] call reproducible...
When the input is a sparse array, the `.chunk_X()` function in anndata package still returns a dense array instead of a sparse array. This will result in out-of-memory errors for...