Normalize
usage: normalize [-h] [-s SCALE] [--affine] [--sample SAMPLE]
[--niters NITERS] [-a ALPHA] [-b BETA] [--metadata]
[-d DEVICE] [-t NUM_WORKERS] [-j NUM_THREADS] [-o DESTDIR]
[--format FORMAT_] [-v]
files [files ...]
Positional Arguments
- files
Named Arguments
- -s, --scale
downsample images by this factor (default: 1)
Default: 1
- --affine
use standard normalization (x-mu)/std of whole image rather than GMM normalization
Default: False
- --sample
pixel sampling factor for model fit. speeds up estimation of parameters but introduces sample error if set >1. (default: 10)
Default: 10
- --niters
maximum number of EM iterations to run for model fit (default: 100)
Default: 100
- -a, --alpha
alpha parameter of the beta distribution prior on the mixing proportion (default: 900)
Default: 900
- -b, --beta
beta parameter of the beta distribution prior on the mixing proportion (default: 1)
Default: 1
- --metadata
if set, save parameter metadata for each micrograph
Default: False
- -d, --device
which device to use, set to -1 to force CPU. >=0 specifies GPU number (default: -1)
Default: -1
- -t, --num-workers
number of parallel processes to use, 0 specifies main process only (default: 0)
Default: 0
- -j, --num-threads
number of threads for pytorch, 0 uses pytorch defaults, <0 uses all cores (default: 0)
Default: 0
- -o, --destdir
output directory
- --format
image format(s) to write. choices are mrc, tiff, and png. images can be written in multiple formats by specifying each in a comma separated list, e.g. mrc,png would write mrc and png format images (default: mrc)
Default: “mrc”
- -v, --verbose
verbose output
Default: False