The hdmf-zarr library implements a Zarr backend for HDMF as well as convenience classes for integration of Zarr with PyNWB to support writing of NWB files to Zarr.
ZarrIO and NWBZarrIO read and write the Zarr v3 format. Legacy Zarr v2 files are read with ZarrV2IO and NWBZarrV2IO, which also provide helpers to convert those files to Zarr v3.
Status: The Zarr backend is under development and may still change. See the overview page for an overview of the available features and known limitations of hdmf-zarr.
Latest release:
Dev branch:
If you use HDMF or hdmf_zarr in your research, please use the following citation:
- A. J. Tritt, O. Ruebel, B. Dichter, R. Ly, D. Kang, E. F. Chang, L. M. Frank, K. Bouchard, "HDMF: Hierarchical Data Modeling Framework for Modern Science Data Standards," 2019 IEEE International Conference on Big Data (Big Data), Los Angeles, CA, USA, 2019, pp. 165-179, doi: 10.1109/BigData47090.2019.9005648.
- HDMF-Zarr, RRID:SCR_022709
See the hdmf-zarr documentation for details: https://hdmf-zarr.readthedocs.io
The library is intended to be used in conjunction with HDMF. hdmf-zarr mainly provides
with the ZarrIO class an alternative to the HDF5IO I/O backend that ships with HDMF.
To support customization of I/O settings, hdmf-zarr provides ZarrDataIO (similar to
H5DataIO in HDMF). Using ZarrIO and ZarrDataIO works much in the same way as HDF5IO.
To ease integration with the NWB data standard and PyNWB, hdmf-zarr provides the NWBZarrIO
class as alternative to pynwb.NWBHDF5IO. See the tutorials included with the documentation for more details.