John Cupitt

Hi @rgluskin, Could you share some sample code that shows the problem? How are you saving in HDF5? Is this with `write_to_file('xxx.mat')`, or some numpy feature? As far as I...

> HDF5 is used to store the results on disk, not in memory. I think loading HDF5 from disc will need a lot of memory, won't it? And it'll use...

Great! Thanks for that. I tried: ``` $ VIPS_PROGRESS=1 ./rgluskin.py mapping ... making vips image ... writing ... rgluskin.py temp-2: 192000 x 100000 pixels, 32 threads, 192000 x 1 tiles,...

Hi again, I can't reproduce your problem, so I can't fix it. You need to post a complete, runnable program I can test. What cffi are you using, what python...

Hello @xyuechen, I think that error means there's a 32/64-bit mixup. I would install a 64-bit python from python.org and the 64-bit libvips binary. You can also try with anaconda...

Hello @petroslk, TIFF doesn't support multichannel images that well -- for example: ``` $ vips black x.tif 100 100 --bands 100 $ tiffinfo x.tif === TIFF directory 0 === TIFF...

Ooop! You're right, it should get `num_channels` before splitting the image, sorry about that. I tried this version: ```python #!/usr/bin/env python3 import sys import pyvips im = pyvips.Image.new_from_file(sys.argv[1]) num_channels =...

... I meant to say, `vipsdisp` is useful for testing as well. With that TIFF I see:  You can see the six bands in the info bar at the...

Did you update your `vips2ome.py` script? The first version I posted had a bug that could cause this.

Wait, is your source image already an OME TIFF? But then why are you trying to convert it? Are you loading it, doing some processing, and trying to save again?...