lattice/pyerrors
1
0
Fork 0
mirror of https://github.com/fjosw/pyerrors.git synced 2025-05-15 20:13:41 +02:00
Code Issues Projects Releases Packages Wiki Activity

first ver. of just one SFCF read method

Browse source
This commit is contained in:
jkuhl-uni 2021-11-29 13:13:13 +01:00
parent 089cbc0783
commit 66cdd46a92
1 changed files with 165 additions and 73 deletions
Download patch file Download diff file Expand all files Collapse all files

154
pyerrors/input/sfcf.py
View file

@ -8,11 +8,11 @@ import numpy as np # Thinly-wrapped numpy
from ..obs import Obs from ..obs import Obs
def read_sfcf(path, prefix, name, **kwargs): def read_sfcf_old(path, prefix, name, quarks, noffset = 0, wf=0, wf2=0, **kwargs):
"""Read sfcf C format from given folder structure. """Read sfcf format (from around 2012) from given folder structure.
Parameters Keyword arguments
---------- -----------------
im -- if True, read imaginary instead of real part of the correlation function. im -- if True, read imaginary instead of real part of the correlation function.
single -- if True, read a boundary-to-boundary correlation function with a single value single -- if True, read a boundary-to-boundary correlation function with a single value
b2b -- if True, read a time-dependent boundary-to-boundary correlation function b2b -- if True, read a time-dependent boundary-to-boundary correlation function
@ -25,16 +25,12 @@ def read_sfcf(path, prefix, name, **kwargs):
im = 0 im = 0
part = 'real' part = 'real'
if kwargs.get('single'):
b2b = 1
single = 1
else:
b2b = 0 b2b = 0
single = 0
if kwargs.get('b2b'): if kwargs.get('b2b'):
b2b = 1 b2b = 1
quarks = quarks.split(" ")
read = 0 read = 0
T = 0 T = 0
start = 0 start = 0
@ -43,7 +39,8 @@ def read_sfcf(path, prefix, name, **kwargs):
ls.extend(dirnames) ls.extend(dirnames)
break break
if not ls: if not ls:
raise Exception('Error, directory not found') print('Error, directory not found')
#sys.exit()
for exc in ls: for exc in ls:
if fnmatch.fnmatch(exc, prefix + '*'): if fnmatch.fnmatch(exc, prefix + '*'):
ls = list(set(ls) - set(exc)) ls = list(set(ls) - set(exc))
@ -56,17 +53,12 @@ def read_sfcf(path, prefix, name, **kwargs):
if len(new_names) != replica: if len(new_names) != replica:
raise Exception('Names does not have the required length', replica) raise Exception('Names does not have the required length', replica)
else: else:
# Adjust replica names to new bookmarking system new_names = ls
new_names = []
for entry in ls:
idx = entry.index('r')
new_names.append(entry[:idx] + '|' + entry[idx:])
print(replica, 'replica') print(replica, 'replica')
for i, item in enumerate(ls): for i, item in enumerate(ls):
print(item) print(item)
sub_ls = [] sub_ls = []
for (dirpath, dirnames, filenames) in os.walk(path + '/' + item): for (dirpath, dirnames, filenames) in os.walk(path+'/'+item):
sub_ls.extend(dirnames) sub_ls.extend(dirnames)
break break
for exc in sub_ls: for exc in sub_ls:
@ -75,18 +67,25 @@ def read_sfcf(path, prefix, name, **kwargs):
sub_ls.sort(key=lambda x: int(x[3:])) sub_ls.sort(key=lambda x: int(x[3:]))
no_cfg = len(sub_ls) no_cfg = len(sub_ls)
print(no_cfg, 'configurations') print(no_cfg, 'configurations')
if i == 0: if i == 0:
with open(path + '/' + item + '/' + sub_ls[0] + '/' + name) as fp: with open(path + '/' + item + '/' + sub_ls[0] + '/' + name) as fp:
for k, line in enumerate(fp): for k, line in enumerate(fp):
#check if this is really the right file
pattern = "# "+name+" : offset "+str(noffset)+", wf "+"0"
#if b2b, a second wf is needed
if b2b:
pattern+=", wf_2 "+"0"
pattern+=" : "+quarks[0]+" - "+quarks[1]
if read == 1 and not line.strip() and k > start + 1: if read == 1 and not line.strip() and k > start + 1:
break break
if read == 1 and k >= start: if read == 1 and k >= start:
T += 1 T += 1
if '[correlator]' in line: if pattern in line:
#print(line)
read = 1 read = 1
start = k + 7 + b2b start = k+1
T -= b2b print(str(T)+" entries found.")
deltas = [] deltas = []
for j in range(T): for j in range(T):
@ -97,11 +96,12 @@ def read_sfcf(path, prefix, name, **kwargs):
deltas[j].append(np.zeros(sublength)) deltas[j].append(np.zeros(sublength))
for cnfg, subitem in enumerate(sub_ls): for cnfg, subitem in enumerate(sub_ls):
with open(path + '/' + item + '/' + subitem + '/' + name) as fp: with open(path + '/' + item + '/' + subitem + '/'+name) as fp:
for k, line in enumerate(fp): for k, line in enumerate(fp):
if(k >= start and k < start + T): if(k >= start and k < start + T):
floats = list(map(float, line.split())) floats = list(map(float, line.split()))
deltas[k - start][i][cnfg] = floats[1 + im - single] deltas[k-start][i][cnfg] = floats[im]
result = [] result = []
for t in range(T): for t in range(T):
@ -110,7 +110,7 @@ def read_sfcf(path, prefix, name, **kwargs):
return result return result
def read_sfcf_c(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0, **kwargs): def read_sfcf(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0, **kwargs):
"""Read sfcf c format from given folder structure. """Read sfcf c format from given folder structure.
Parameters Parameters
@ -121,11 +121,11 @@ def read_sfcf_c(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0, **kwarg
wf2 -- ID of the second wavefunction (only relevant for boundary-to-boundary correlation functions) wf2 -- ID of the second wavefunction (only relevant for boundary-to-boundary correlation functions)
im -- if True, read imaginary instead of real part of the correlation function. im -- if True, read imaginary instead of real part of the correlation function.
b2b -- if True, read a time-dependent boundary-to-boundary correlation function b2b -- if True, read a time-dependent boundary-to-boundary correlation function
single -- if True, read time independent boundary to boundary correlation function
names -- Alternative labeling for replicas/ensembles. Has to have the appropriate length names -- Alternative labeling for replicas/ensembles. Has to have the appropriate length
ens_name : str ens_name : str
replaces the name of the ensemble replaces the name of the ensemble
""" """
if kwargs.get('im'): if kwargs.get('im'):
im = 1 im = 1
part = 'imaginary' part = 'imaginary'
@ -133,12 +133,38 @@ def read_sfcf_c(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0, **kwarg
im = 0 im = 0
part = 'real' part = 'real'
if kwargs.get('single'):
b2b = 1
single = 1
else:
if kwargs.get('b2b'): if kwargs.get('b2b'):
b2b = 1 b2b = 1
else: else:
b2b = 0 b2b = 0
single = 0
files = []
if "files" in kwargs:
files = kwargs.get("files")
#due to higher usage in current projects, compact file format is default
compact = True
#get version string
version = "1.0"
known_versions = ["0.0","1.0","2.0","1.0c","2.0c"]
if "version" in kwargs:
version = kwargs.get("version")
if not version in known_versions:
raise Exception("This version is not known!")
#if the letter c is appended to the version, the compact fileformat is used (former read_sfcf_c)
if(version[-1] == "c"):
compact = True
version = version[:-1]
else:
compact = False
read = 0
T = 0 T = 0
start = 0
ls = [] ls = []
for (dirpath, dirnames, filenames) in os.walk(path): for (dirpath, dirnames, filenames) in os.walk(path):
ls.extend(dirnames) ls.extend(dirnames)
@ -146,12 +172,17 @@ def read_sfcf_c(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0, **kwarg
if not ls: if not ls:
raise Exception('Error, directory not found') raise Exception('Error, directory not found')
# Exclude folders with different names # Exclude folders with different names
if len(files) != 0:
ls = files
else:
for exc in ls: for exc in ls:
if not fnmatch.fnmatch(exc, prefix + '*'): if not fnmatch.fnmatch(exc, prefix + '*'):
ls = list(set(ls) - set([exc])) ls = list(set(ls) - set([exc]))
if len(ls) > 1: if len(ls) > 1:
ls.sort(key=lambda x: int(re.findall(r'\d+', x[len(prefix):])[0])) # New version, to cope with ids, etc. ls.sort(key=lambda x: int(re.findall(r'\d+', x[len(prefix):])[0])) # New version, to cope with ids, etc.
replica = len(ls) replica = len(ls)
print('Read', part, 'part of', name, 'from', prefix[:-1], ',', replica, 'replica')
if 'names' in kwargs: if 'names' in kwargs:
new_names = kwargs.get('names') new_names = kwargs.get('names')
if len(new_names) != replica: if len(new_names) != replica:
@ -160,23 +191,35 @@ def read_sfcf_c(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0, **kwarg
# Adjust replica names to new bookmarking system # Adjust replica names to new bookmarking system
new_names = [] new_names = []
for entry in ls: for entry in ls:
try:
idx = entry.index('r') idx = entry.index('r')
except:
idx = len(entry)-2
if 'ens_name' in kwargs: if 'ens_name' in kwargs:
new_names.append(kwargs.get('ens_name') + '|' + entry[idx:]) new_names.append(kwargs.get('ens_name') + '|' + entry[idx:])
else: else:
new_names.append(entry[:idx] + '|' + entry[idx:]) new_names.append(entry[:idx] + '|' + entry[idx:])
print('Read', part, 'part of', name, 'from', prefix[:-1], ',', replica, 'replica')
for i, item in enumerate(ls): for i, item in enumerate(ls):
sub_ls = [] sub_ls = []
for (dirpath, dirnames, filenames) in os.walk(path + '/' + item): for (dirpath, dirnames, filenames) in os.walk(path + '/' + item):
if compact:
sub_ls.extend(filenames) sub_ls.extend(filenames)
else:
sub_ls.extend(dirnames)
break break
#print(sub_ls)
for exc in sub_ls: for exc in sub_ls:
if compact:
if not fnmatch.fnmatch(exc, prefix + '*'): if not fnmatch.fnmatch(exc, prefix + '*'):
sub_ls = list(set(sub_ls) - set([exc])) sub_ls = list(set(sub_ls) - set([exc]))
sub_ls.sort(key=lambda x: int(re.findall(r'\d+', x)[-1])) sub_ls.sort(key=lambda x: int(re.findall(r'\d+', x)[-1]))
else:
if not fnmatch.fnmatch(exc, 'cfg*'):
sub_ls = list(set(sub_ls) - set([exc]))
sub_ls.sort(key=lambda x: int(x[3:]))
if compact:
first_cfg = int(re.findall(r'\d+', sub_ls[0])[-1]) first_cfg = int(re.findall(r'\d+', sub_ls[0])[-1])
last_cfg = len(sub_ls) + first_cfg - 1 last_cfg = len(sub_ls) + first_cfg - 1
@ -188,8 +231,14 @@ def read_sfcf_c(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0, **kwarg
no_cfg = last_cfg - first_cfg + 1 no_cfg = last_cfg - first_cfg + 1
print(item, ':', no_cfg, 'evenly spaced configurations (', first_cfg, '-', last_cfg, ') ,', len(sub_ls) - no_cfg, 'configs omitted\n') print(item, ':', no_cfg, 'evenly spaced configurations (', first_cfg, '-', last_cfg, ') ,', len(sub_ls) - no_cfg, 'configs omitted\n')
else:
no_cfg = len(sub_ls)
print(no_cfg, 'configurations')
#here we have found all the files we need to look into.
if i == 0: if i == 0:
if compact:
pattern = 'name ' + name + '\nquarks ' + quarks + '\noffset ' + str(noffset) + '\nwf ' + str(wf) pattern = 'name ' + name + '\nquarks ' + quarks + '\noffset ' + str(noffset) + '\nwf ' + str(wf)
if b2b: if b2b:
pattern += '\nwf_2 ' + str(wf2) pattern += '\nwf_2 ' + str(wf2)
@ -205,7 +254,36 @@ def read_sfcf_c(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0, **kwarg
print(T, 'entries, starting to read in line', start_read) print(T, 'entries, starting to read in line', start_read)
else: else:
raise Exception('Correlator with pattern\n' + pattern + '\nnot found.') raise Exception('Correlator with pattern\n' + pattern + '\nnot found.')
else:
#print(path + '/' + item + '/')# + sub_ls[0] + '/' + name)
with open(path + '/' + item + '/' + sub_ls[0] + '/' + name) as fp:
for k, line in enumerate(fp):
if version == "0.0":
#check if this is really the right file
pattern = "# "+name+" : offset "+str(noffset)+", wf "+str(wf)
#if b2b, a second wf is needed
if b2b:
pattern+=", wf_2 "+str(wf2)
qs = quarks.split(" ")
pattern+=" : "+qs[0]+" - "+qs[1]
#print(pattern)
if read == 1 and not line.strip() and k > start + 1:
break
if read == 1 and k >= start:
T += 1
if version == "0.0":
if pattern in line:
#print(line)
read = 1
start = k+1
else:
if '[correlator]' in line:
read = 1
start = k + 7 + b2b
T -= b2b
print(str(T)+" entries found.")
#we found where the correlator that is to be read is in the files
deltas = [] deltas = []
for j in range(T): for j in range(T):
deltas.append([]) deltas.append([])
@ -213,16 +291,30 @@ def read_sfcf_c(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0, **kwarg
sublength = no_cfg sublength = no_cfg
for j in range(T): for j in range(T):
deltas[j].append(np.zeros(sublength)) deltas[j].append(np.zeros(sublength))
if compact:
for cfg in range(no_cfg): for cfg in range(no_cfg):
with open(path + '/' + item + '/' + sub_ls[cfg]) as fp: with open(path + '/' + item + '/' + sub_ls[cfg]) as fp:
for k, line in enumerate(fp): lines = fp.readlines()
if(start_read + T>len(lines)):
raise Exception("EOF before end of correlator data! Maybe "+path + '/' + item + '/' + sub_ls[cfg]+" is corrupted?")
for k in range(start_read - 6,start_read + T):
if k == start_read - 5 - b2b: if k == start_read - 5 - b2b:
if line.strip() != 'name ' + name: if lines[k].strip() != 'name ' + name:
raise Exception('Wrong format', sub_ls[cfg]) raise Exception('Wrong format', sub_ls[cfg])
if(k >= start_read and k < start_read + T): if(k >= start_read and k < start_read + T):
floats = list(map(float, line.split())) floats = list(map(float, lines[k].split()))
deltas[k - start_read][i][cfg] = floats[-2:][im] deltas[k - start_read][i][cfg] = floats[-2:][im]
else:
for cnfg, subitem in enumerate(sub_ls):
with open(path + '/' + item + '/' + subitem + '/' + name) as fp:
for k, line in enumerate(fp):
if(k >= start and k < start + T):
floats = list(map(float, line.split()))
if version == "0.0":
deltas[k-start][i][cnfg] = floats[im]
else:
deltas[k - start][i][cnfg] = floats[1 + im - single]
result = [] result = []
for t in range(T): for t in range(T):