Homec4science
Diffusion blackdynamite (features/zeo) Browse

R3127/BlackDynamite/BDstat.py
faf1a2a8a353features/zeo

Last Change
Raw File

BDstat.py
View Options

#!/usr/bin/env python
# -*- coding: utf-8 -*-
# -*- py-which-shell: "python"; -*-
from __future__ import print_function
__all__ = [ "BDStat" ]
import numpy as np
import job
import copy
class BDStat(object):
"""
"""
def average(self,quantity,run_list,entries_to_average):
result = dict()
run_info = dict()
cpt = dict()
run_counter = 0
nb_runs = len(run_list)
entries_to_average.append("id")
entries_to_consider = []
if (run_list is not []):
for e in run_list[0][1].entries:
if e not in entries_to_average:
entries_to_consider.append(e)
entries_to_consider = sorted(entries_to_consider)
#print (entries_to_consider)
for r,j in run_list:
q = r.getScalarQuantity(quantity)
run_counter += 1
if (q is None):
continue
print ("{0:<5} {1:<15} {2:>5}/{3:<5} {4:.1f}%".format(r.id,q.shape,run_counter,nb_runs,1.*run_counter/nb_runs*100))
ent = [j[i] for i in entries_to_consider]
ent = tuple(ent)
if (ent not in result.keys()):
# print (ent)
result[ent] = np.zeros([q.shape[0],3])
cpt[ent] = 0
# print (result[ent].shape)
# print (q.shape)
sz1 = q.shape[0]
sz2 = result[ent].shape[0]
if (sz1 > sz2):
q = q[:sz2]
elif (sz2 > sz1):
result[ent] = result[ent][:sz1]
result[ent][:,0] += q[:,0]
result[ent][:,1] += q[:,1]
result[ent][:,2] += q[:,1]**2
cpt[ent] += 1
for ent in result.keys():
# print (ent)
myjob = job.Job(self.base)
for i in range(0,len(entries_to_consider)):
myjob.entries[entries_to_consider[i]] = ent[i]
result[ent] /= cpt[ent]
result[ent][:,2] -= result[ent][:,1]**2
result[ent][:,2] = np.maximum(np.zeros(result[ent][:,2].shape[0]),result[ent][:,2])
result[ent][:,2] = np.sqrt(result[ent][:,2])
result[ent] = {"ref_job":myjob ,
"averaged_number":cpt[ent],
"data":result[ent]
}
return result
def averageVector(self,quantity,run_list,entries_to_average):
result = dict()
all_steps = dict()
run_info = dict()
cpt = dict()
run_counter = 0
nb_runs = len(run_list)
entries_to_average.append("id")
entries_to_consider = []
if (run_list is not []):
for e in run_list[0][1].entries:
if e not in entries_to_average:
entries_to_consider.append(e)
for r,j in run_list:
steps,data = r.getAllVectorQuantity(quantity)
run_counter += 1
if (data is None):
continue
print ("{0:<5} {1:<15} {2:>5}/{3:<5} {4:.1f}%".format(r.id,data.shape,run_counter,nb_runs,1.*run_counter/nb_runs*100))
ent = [j[i] for i in entries_to_consider]
ent = tuple(ent)
if (ent not in result.keys()):
# print (ent)
result[ent] = np.zeros([data.shape[0],data.shape[1],2])
all_steps[ent] = steps
cpt[ent] = 0
# print (result[ent].shape)
# print (q.shape)
sz1 = data.shape[0]
sz2 = result[ent].shape[0]
if (sz1 > sz2):
data = data[:sz2]
steps = steps[:sz2]
elif (sz2 > sz1):
result[ent] = result[ent][:sz1]
all_steps[ent] = all_steps[ent][:sz1]
result[ent][:,:,0] += data[:,:]
result[ent][:,:,1] += data[:,:]**2
cpt[ent] += 1
for ent in result.keys():
# print (ent)
myjob = job.Job(self.base)
for i in range(0,len(entries_to_consider)):
myjob.entries[entries_to_consider[i]] = ent[i]
result[ent] /= cpt[ent]
result[ent][:,:,1] -= result[ent][:,:,0]**2
result[ent][:,:,1] = np.sqrt(result[ent][:,:,1])
result[ent] = {"ref_job":myjob ,
"averaged_number":cpt[ent],
"steps":all_steps[ent],
"data":result[ent]}
return result
def __init__ (self,base,**params):
self.base = base
c4science · Help