stephane.jhtd package¶
Examples¶
Here is an example of reading and using data sample
Reading 3d data from JHTDB data¶
import numpy as np
import stephane.jhtd.get as jhtd_get
import stephane.display.graphes as graphes
import glob #smart module to access datafile
import os.path
"""
Code example to read, process and plot data from JHTD
"""
rootdir='/Volumes/labshared/JHTDB_Data/3d_volume/'
fileList = glob.glob(rootdir+'jhtdb_data*.h5')
print(len(fileList))
print(fileList)
file = fileList[0]
print(file)
#read the data into a h5py format
data = jhtd_get.read(file)
# dump example : compute the distribution of velocity from individual components
U = jhtd_get.vlist(data) #data are polluted by the presence of nan values. Currently fixed by removing any NaN from the data list
graphes.hist(U,log=True,fignum=1)
graphes.legende('V (a.u.)','pdf of V','')
#save the image file in the same folder
filename = os.path.dirname(os.path.abspath(__file__))+'/Figures/velocity_histogramm'
#filename = './velocity_histogramm'
graphes.save_fig(0,filename,frmt='png')
graphes.save_fig(0,filename,frmt='pdf')
print(filename)
Read collection of small dataset from JHTDB data Use them to compute geometrical quantities ——–
import numpy as np
import stephane.jhtd.get as jhtd_get
import stephane.display.graphes as graphes
import stephane.jhtd.geom as geom
import stephane.tools.dict2list as dict2list
import glob
import matplotlib.pyplot as plt
"""
load the data file by file, and compute vorticity, eigenvectors of the strain tensor and geometrical angles for each single point
Call plots() and average_values() as outputs
"""
Datadir = '/Volumes/labshared/JHTDB_Data/JHTDB_lib/'#/Users/stephane/Documents/JHT_Database/Programs_JH/JHTDB_lib'+'/'
#fileList = glob.glob(Datadir+'*.h5') #entire fileList
names = ['jhtdb_data_zl_10_yl_10_xl_10_t0_1001_tl_1_y0_401_x0_601_z0_501.h5',
'jhtdb_data_zl_10_yl_10_xl_10_t0_101_tl_1_y0_501_x0_201_z0_1.h5',
'jhtdb_data_zl_10_yl_10_xl_10_t0_101_tl_1_y0_701_x0_1001_z0_201.h5',
'jhtdb_data_zl_10_yl_10_xl_10_t0_101_tl_1_y0_801_x0_1_z0_801.h5',
'jhtdb_data_zl_10_yl_10_xl_10_t0_150_tl_1_y0_250_x0_150_z0_450.h5'] #subsecltion : compute from only 5 jhtd files
fileList = [Datadir + name for name in names]
print(fileList)
N=len(fileList)
print('Number of files found : '+str(N))
#N=10000 #number of files to process
step = 10 #N/step corresponds to the number of bins for the output
E_t=[] #energy v
omega_t=[] #vorticity
cosine_t={} #dict, cosine angle between various geometrical quantities
eigen_t={} # eigenvalues of the strain tensor lambda_i, and associated eigenvectors Lambda_i
count=0 #counter for the number of corrupted files
for i,file in enumerate(fileList[:N]):
if i%100==0:
print(str(i*100//N)+' %')
# file ='/Users/stephane/Documents/JHT_Database/Data_sample_2015_10_07/Serie/isotropic1024coarse_'+str(i)+'.h5'
try:
data = jhtd_get.read(file)
except:
count+=1
data=None
# print("Datafile not opened : "+file)
if data is not None:
E,eigen,omega,cosine = geom.process(data)
#print(eigen)
E_t += E
omega_t += omega
eigen_t = dict2list.add_list_to_dict(eigen_t,eigen)
cosine_t = dict2list.add_list_to_dict(cosine_t,cosine)
print("Percentage of data files corrupted : "+str((100*count)/N)+" %")
print(eigen_t.keys())
geom.average_values(eigen_t,omega_t,cosine_t)
figs = geom.plots(eigen_t,omega_t,cosine_t,step)
graphes.save_figs(figs,savedir='./Figures/',suffix='',frmt='pdf')
#plt.show(block=True)
#conditionnal pdfs :
#still in progress
T=0.9
condition={}
condition[T]='lambda_omega_1'
eigen_extract,omega_extract,cosine_extract=geom.subset(eigen_t,omega_t,cosine_t,condition[T],T)
figs = geom.plots(eigen_extract,omega_extract,cosine_extract,step)
graphes.save_figs(figs,savedir='./Figures/',suffix='_Conditionnal_'+condition[T]+'gt_0_9',frmt='pdf')
#plt.show(block=True)