alisterburt
commented
4 years ago This sounds like a great idea, would you mind sending me a large RELION model star file for testing?
I will probably try use stringio because I have used these before but if you have experience with mmap and would like to have a go it would be good to compare both!
The idea about an option for just reading the first few blocks is a good one and should work with the way things are set up - I’ll look into it over the next week or so.
Cheers,
Alister
On 12 Oct 2020, at 10:02, Xhark notifications@github.com wrote:
When coping with a star file contains thousands of sub data blocks(.e.g. 3D classification's model file), is it possible to use a mmap or stringio buffer object to feed pandas.read_csv's argument in _read_loop_data to increase the performance? Otherwise it might take hours to read a model star file. Furthermore, add an option to read just first several blocks would be a nice addition to speed up specific case. Cheers.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe.
inter1965
When coping with a star file contains thousands of sub data blocks(.e.g. 3D classification's model file), is it possible to use a mmap or stringio buffer object to feed pandas.read_csv's argument in _read_loop_data to increase the performance? Otherwise it might take hours to read a model star file. Furthermore, add an option to read just first several blocks would be a nice addition to speed up specific case. Cheers.