spatialdata
spatialdata copied to clipboard
scverse
An open and interoperable data framework for spatial omics data
We have received feedback that our intro notebooks "Use SpatialData with your data": 1. [I. Use SpatialData with your data: the SpatialData object](https://spatialdata.scverse.org/en/latest/tutorials/notebooks/notebooks/examples/models1.html) 2. [II. Use SpatialData with your data:...
- [ ] new notebook similar to [Pytorch dataset loading](https://spatialdata.scverse.org/en/stable/tutorials/notebooks/notebooks/examples/densenet.html) - [ ] bencharking the spatialdata tutorial against saving an intermediate file (Huggingface dataset, HDF5) - [ ] resource for...
Spatial partitioning with Dask geopandas is described here: https://dask-geopandas.readthedocs.io/en/stable/guide/spatial-partitioning.html
CC @melonora @berombau Working with Zarr files on a HPC may create problems due to Zarr v2. One of the challenges for the hackathon will be to setup a set...
**Is your feature request related to a problem? Please describe.** A lot of documentation is focused on one dataset or method. Method developers need to benchmark multiple datasets or methods...
While extending SOPA for Visium HD data, @quentinblampey encountered a performance bottleneck with `aggregate()` that he could improve using pure `geopandas` code. Since we use `geopandas` internally, this could be...
`aggregate()` is not exploiting dask partitions and parallelization; using it may improve performance. We could make some experiments/benchmarks.
When using `spatialdata-plot`, I don't have auto-completion on VS Code. It seems to be the same in PyCharm. Also, one area to explore would be to import spatialdata-plot (if installed)...
The [transformation notebook](https://spatialdata.scverse.org/en/latest/tutorials/notebooks/notebooks/examples/transformations.html) is centered on showing how the API works and not in telling a biological story (for instance it uses a raccoon image image instead of a biological...
It would be convenient to know the data extent, or the extent of some elements, and being able to query and save a smaller object without having to open Python.