python-caterva
python-caterva copied to clipboard
Blosc
Python wrapper for Caterva. Still preliminary.
Hello and thanks for working on this great project! 👏 I am an Xarray developer. We would really love to be able to store xarray data in Caterva. In order...
If we don't want to maintain a binder notebook, I'd suggest to to remove the button (although keeping it working is kind of cool 😎).
In data science it is much more common to use the noun `array` for referring to multidimensional data containers. We should try to adapt to this convention and stop using...
Hello. I am trying to use the parallel feature of Blosc compression to write (and read) large (~100GB) dataset to disk, similar to h5py's (or zarr's) create_dataset. I could not...
The `ndindex` library (https://github.com/Quansight/ndindex) can be useful for dealing with multidimensional indexes.
The [trailets library](https://github.com/ipython/traitlets) could be useful to handle the important amount of different parameters (compression params, shape-related params...) for the functions.
So, in addition of e.g. `cname='lz4'`, one can pass `cname=cat4py.LZ4` to functions too.
When trying get a serialized frame with `.to_sframe()`, add checks for telling that plain buffers cannot be serialized yet.