-
Notifications
You must be signed in to change notification settings - Fork 2
/
ard-plotter.py
44 lines (33 loc) · 1.94 KB
/
ard-plotter.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
import numpy as np
import pylab as pl
import random
import scipy.io
files = ["/Users/nickp/Dropbox/Repos/mkl/matlab/hia.txt.0.mat.sav", "/Users/nickp/Dropbox/Repos/mkl/matlab/hia.txt.1.mat.sav", "/Users/nickp/Dropbox/Repos/mkl/matlab/hia.txt.2.mat.sav", "/Users/nickp/Dropbox/Repos/mkl/matlab/hia.txt.3.mat.sav", "/Users/nickp/Dropbox/Repos/mkl/matlab/hia.txt.4.mat.sav", "/Users/nickp/Dropbox/Repos/mkl/matlab/pgp.txt.0.mat.sav", "/Users/nickp/Dropbox/Repos/mkl/matlab/pgp.txt.1.mat.sav", "/Users/nickp/Dropbox/Repos/mkl/matlab/pgp.txt.2.mat.sav", "/Users/nickp/Dropbox/Repos/mkl/matlab/pgp.txt.3.mat.sav", "/Users/nickp/Dropbox/Repos/mkl/matlab/pgp.txt.4.mat.sav", "/Users/nickp/Dropbox/Repos/mkl/matlab/tdp.txt.0.mat.sav", "/Users/nickp/Dropbox/Repos/mkl/matlab/tdp.txt.1.mat.sav", "/Users/nickp/Dropbox/Repos/mkl/matlab/tdp.txt.2.mat.sav", "/Users/nickp/Dropbox/Repos/mkl/matlab/tdp.txt.3.mat.sav", "/Users/nickp/Dropbox/Repos/mkl/matlab/tdp.txt.4.mat.sav", "/Users/nickp/Dropbox/Repos/mkl/matlab/synthredundant.txt.0.mat.sav"]
#files = [files[-1]]
for i, f in enumerate(files):
mat = []
spr = []
raw = open(f).read().strip().split("\n")
raw = filter(lambda h: len(h) > 0, raw)
raw = filter(lambda h: h[0] != '#', raw)
for r in xrange(len(raw)-1):
mat.append(map(float, raw[r].strip().split(" ")))
spr = map(float, raw[len(raw)-1].strip().split(" "))
mat = np.array(mat)
spr = np.array(spr)
# for k in xrange(100):
# for j in xrange(20):
# mat[j, k]= 1.0 - random.random() / 5.0
# for k in [23, 43, 65, 34, 78, 13, 1, 45, 67, 94]:
# for j in xrange(20):
# mat[j, k]= random.random() / 10.0
N = mat.shape[1]
pl.figure()
pl.bar(np.arange(1, N+1), np.mean(mat, axis=0), 0.4, color='r', yerr=np.std(mat, axis=0) / 2.0)
pl.ylim(0, np.mean(mat + np.std(mat, axis=0) / 2.0, axis=0).max())
pl.xlim(1, N+1)
pl.xlabel("Feature")
pl.ylabel("Unimportance")
pl.savefig("/Users/nickp/Dropbox/PhD/Thesis/img/ardgp-%d.eps" % i)
pl.show()
print mat.shape