Created
March 9, 2016 09:19
-
-
Save hnishi/17c84719b41a5362ee61 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# prjnishi.py v 1.0 | |
# projection into PC space with a PCA eigenvector and average. | |
import sys | |
import numpy as np | |
from optparse import OptionParser | |
############ argument ############## | |
parser = OptionParser | |
def get_options(): | |
p = OptionParser() | |
p.add_option('--i-trj', dest='fn_trj', | |
help="file name for trajectory of structure components") | |
p.add_option('--i-ave', dest='fn_ave', | |
help="file name for average of structure components") | |
p.add_option('--i-egv', dest='fn_egv', | |
help="file name for eigen vector of VCV by PCA") | |
p.add_option('--o-pcc', dest='fn_pcc', | |
help="file name for output coordinates in PC space") | |
opts, args = p.parse_args() | |
print "----------------------------" | |
p.print_help() | |
print "----------------------------" | |
return opts, args | |
############ main ############## | |
print "projnishi.py" | |
print "Projection from principle component analysis (PCA)" | |
opts, args = get_options() | |
fn_trj = 'coord.dat' | |
if opts.fn_trj: | |
fn_trj = opts.fn_trj | |
f = open(fn_trj) | |
a = np.array(f.read().split(),dtype=float) | |
f.close() | |
fn_ave = 'aveq.dat' | |
if opts.fn_ave: | |
fn_ave = opts.fn_ave | |
f = open(fn_ave) | |
a2 = np.array(f.read().split(),dtype=float) | |
f.close() | |
fn_egv = 'e1.dat' | |
if opts.fn_egv: | |
fn_egv = opts.fn_egv | |
f = open(fn_egv) | |
a3 = np.array(f.read().split(),dtype=float) | |
f.close() | |
if len(a2) != len (a3): | |
print "ERROR: the num of date in fn_ave and fn_egv are not the same" | |
sys.exit(1) | |
dim = len(a3) | |
structure = len(a)/dim | |
print "dimension: ",dim | |
print "the number of structures: ", structure | |
##### CALCULATE INNER PRODUCT | |
ccc = a.reshape(len(a)/dim,dim) | |
#print np.shape(ccc) | |
dots = [] #dots:PC coordinates | |
for i in xrange(len(a)/dim): | |
#print ccc[i] - a2 | |
#print np.dot(a3,ccc[i] - a2) | |
dots.append(np.dot(a3,ccc[i] - a2)) | |
#print np.shape(dots) | |
#print dots | |
##### OUTPUT PC COORDINATES OF ALL STRUCTURES | |
fn_pcc = 'c1.dat' | |
if opts.fn_pcc: | |
fn_pcc = opts.fn_pcc | |
output = open(fn_pcc,'w') | |
for i in dots: | |
print >> output, i | |
output.close() | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment