I changed my h5py from version 2.10.0 to 3.2.1 recently. I found the string dataset output was changed from string type to bytes type. So I should change my codes to fit this situation. There are two ways to do this:
1st. Decode the output bytes
2nd. Save the string in the attribute which supports the string type, and no decode was needed when reading it.
I want to know which way is better and if the attribute will also support the string type in the future.
Thank you in advance.
It is not clear to me what your situation is.
Have you tried asstr() to read HDF strings created by h5py?
f = h5py.File('foo.h5', 'r') ds = f['data'] strings = ds.asstr()[()]
thanks for the awesome information.
Thanks for your reply.
thanks my issue has been fixed.