def get_data(sim_file, halo_index):
HUBBLE_CONST = 0.688062
sdf_data = load_sdf(sim_file)
x = float(sdf_data['x'][halo_index] / HUBBLE_CONST)
y = float(sdf_data['y'][halo_index] / HUBBLE_CONST)
z = float(sdf_data['z'][halo_index] / HUBBLE_CONST)
vx = float(sdf_data['vx'][halo_index])
vy = float(sdf_data['vy'][halo_index])
vz = float(sdf_data['vz'][halo_index])
mvir = float(sdf_data['mvir'][halo_index] / HUBBLE_CONST)
r200b = float(sdf_data['r200b'][halo_index] / HUBBLE_CONST)
hid = int(sdf_data['id'][halo_index])
hpid = int(sdf_data['pid'][halo_index])
data = {
'x': x,
'y': y,
'z': z,
'vx': vx,
'vy': vy,
'vz': vz,
'mvir': mvir,
'r200b': r200b,
'id': hid,
'pid': hpid
}
return data
from yt.utilities.sdf import load_sdf
import math
HUBBLE_CONST = 0.688062
sim_file = '/media/jsnguyen/JK-PEXHD/ds14_a_halos_1.0000'
save_dir = '/home/jsnguyen/DSS_data/'
load_data_fn = 'reduced_5Mpc_mass_filter_subhalos_1e+14.txt'
save_data_fn = 'full_data_' + load_data_fn
sdf_data = load_sdf(sim_file)
f_pairs_data = open(save_dir + save_data_fn, 'w')
#header describes the format of how a pair is stored
f_pairs_data.write('# pair_id\n')
f_pairs_data.write('# ax ay az avx avy avz amvir ar200b aid apid\n')
f_pairs_data.write('# bx by bz bvx bvy bvz bmvir br200b bid bpid\n')
f_pairs_data.close()
f_pairs = open(save_dir + load_data_fn, 'r')
f_pairs.next() #skip header line
i = 0
for line in f_pairs:
halo_a = int(line.split()[0])
halo_b = int(line.split()[1])
import time
import numpy as np
import yt
import sklearn
from sklearn.cluster import MiniBatchKMeans
import sys
for arg in sys.argv:
k=arg
k = int(k)
yt.funcs.mylog.setLevel(50) #coerce output null
print "{} clusters test".format(k)
num =100000000
print "Loading {} Particles".format(num)
from yt.utilities.sdf import load_sdf
path = 'http://darksky.slac.stanford.edu/simulations/ds14_a/ds14_a_1.0000'
data = load_sdf(path)
x = data['x'][:num]
print x.nbytes
y = data['y'][:num]
z = data['z'][:num]
idx = data['ident'][:num]
#ds = yt.load("../../ds14_scivis_0128_e4_dt04_1.0000")
#ad = ds.all_data()
#x = ad[("all","particle_position_x")]
#y = ad[("all","particle_position_y")]
#z = ad[("all","particle_position_z")]
#idx = ad[("all","particle_index")]
train = np.array([idx,x,y,z]).T
np.savetxt("train.txt",train)
avrg = open('avrg_dens.txt', 'a')
timef = open('time.txt','a')
import struct
from random import *
from math import *
import sys
if len(sys.argv) < 2:
print "usage: %s <filename>" % sys.argv[0]
quit()
print "data%s.xyzb" % sys.argv[1][-4:]
fileToOpen = sys.argv[1]
fileToSave = "data%s.xyzb" % sys.argv[1][-4:]
sdfdata = load_sdf(fileToOpen)
file = open(fileToSave, 'wb')
'''
for ix in range(0,sdfdata['x'].size):
if (sdfdata['vx'][ix] > 0):
maxVX = minVX = sdfdata['vx'][ix]
print (ix)
break
for iy in range(0,sdfdata['x'].size):
if (sdfdata['vy'][iy] > 0):
maxVY = minVY = sdfdata['vy'][iy]
print (iy)
