lattice/pyerrors
1
0
Fork 0
mirror of https://github.com/fjosw/pyerrors.git synced 2025-05-16 12:33:41 +02:00
Code Issues Projects Releases Packages Wiki Activity

Merge branch 'develop' into documentation

Browse source
This commit is contained in:
fjosw 2022-01-16 15:54:32 +00:00
commit 399547a957
2 changed files with 6 additions and 75 deletions
Download patch file Download diff file Expand all files Collapse all files

33
pyerrors/input/openQCD.py
View file

@ -1,6 +1,3 @@
#!/usr/bin/env python
# coding: utf-8
import os import os
import fnmatch import fnmatch
import re import re
@ -42,7 +39,6 @@ def read_rwms(path, prefix, version='2.0', names=None, **kwargs):
if 'files' in kwargs: if 'files' in kwargs:
ls = kwargs.get('files') ls = kwargs.get('files')
else: else:
# Exclude files with different names
for exc in ls: for exc in ls:
if not fnmatch.fnmatch(exc, prefix + '*' + postfix + '.dat'): if not fnmatch.fnmatch(exc, prefix + '*' + postfix + '.dat'):
ls = list(set(ls) - set([exc])) ls = list(set(ls) - set([exc]))
@ -69,7 +65,6 @@ def read_rwms(path, prefix, version='2.0', names=None, **kwargs):
print('Read reweighting factors from', prefix[:-1], ',', print('Read reweighting factors from', prefix[:-1], ',',
replica, 'replica', end='') replica, 'replica', end='')
# Adjust replica names to new bookmarking system
if names is None: if names is None:
rep_names = [] rep_names = []
for entry in ls: for entry in ls:
@ -88,7 +83,6 @@ def read_rwms(path, prefix, version='2.0', names=None, **kwargs):
tmp_array = [] tmp_array = []
with open(path + '/' + ls[rep], 'rb') as fp: with open(path + '/' + ls[rep], 'rb') as fp:
# header
t = fp.read(4) # number of reweighting factors t = fp.read(4) # number of reweighting factors
if rep == 0: if rep == 0:
nrw = struct.unpack('i', t)[0] nrw = struct.unpack('i', t)[0]
@ -97,7 +91,6 @@ def read_rwms(path, prefix, version='2.0', names=None, **kwargs):
for k in range(nrw): for k in range(nrw):
deltas.append([]) deltas.append([])
else: else:
# little weird if-clause due to the /2 operation needed.
if ((nrw != struct.unpack('i', t)[0] and (not version == '2.0')) or (nrw != struct.unpack('i', t)[0] / 2 and version == '2.0')): if ((nrw != struct.unpack('i', t)[0] and (not version == '2.0')) or (nrw != struct.unpack('i', t)[0] / 2 and version == '2.0')):
raise Exception('Error: different number of reweighting factors for replicum', rep) raise Exception('Error: different number of reweighting factors for replicum', rep)
@ -110,8 +103,6 @@ def read_rwms(path, prefix, version='2.0', names=None, **kwargs):
for i in range(nrw): for i in range(nrw):
t = fp.read(4) t = fp.read(4)
nfct.append(struct.unpack('i', t)[0]) nfct.append(struct.unpack('i', t)[0])
# print('nfct: ', nfct) # Hasenbusch factor,
# 1 for rat reweighting
else: else:
for i in range(nrw): for i in range(nrw):
nfct.append(1) nfct.append(1)
@ -124,7 +115,6 @@ def read_rwms(path, prefix, version='2.0', names=None, **kwargs):
if not struct.unpack('i', fp.read(4))[0] == 0: if not struct.unpack('i', fp.read(4))[0] == 0:
print('something is wrong!') print('something is wrong!')
# body
while 0 < 1: while 0 < 1:
t = fp.read(4) t = fp.read(4)
if len(t) < 4: if len(t) < 4:
@ -220,7 +210,6 @@ def extract_t0(path, prefix, dtr_read, xmin,
if not ls: if not ls:
raise Exception('Error, directory not found') raise Exception('Error, directory not found')
# Exclude files with different names
for exc in ls: for exc in ls:
if not fnmatch.fnmatch(exc, prefix + '*.ms.dat'): if not fnmatch.fnmatch(exc, prefix + '*.ms.dat'):
ls = list(set(ls) - set([exc])) ls = list(set(ls) - set([exc]))
@ -232,7 +221,6 @@ def extract_t0(path, prefix, dtr_read, xmin,
r_start = kwargs.get('r_start') r_start = kwargs.get('r_start')
if len(r_start) != replica: if len(r_start) != replica:
raise Exception('r_start does not match number of replicas') raise Exception('r_start does not match number of replicas')
# Adjust Configuration numbering to python index
r_start = [o - 1 if o else None for o in r_start] r_start = [o - 1 if o else None for o in r_start]
else: else:
r_start = [None] * replica r_start = [None] * replica
@ -251,7 +239,6 @@ def extract_t0(path, prefix, dtr_read, xmin,
for rep in range(replica): for rep in range(replica):
with open(path + '/' + ls[rep], 'rb') as fp: with open(path + '/' + ls[rep], 'rb') as fp:
# Read header
t = fp.read(12) t = fp.read(12)
header = struct.unpack('iii', t) header = struct.unpack('iii', t)
if rep == 0: if rep == 0:
@ -270,7 +257,6 @@ def extract_t0(path, prefix, dtr_read, xmin,
Ysl = [] Ysl = []
# Read body
while 0 < 1: while 0 < 1:
t = fp.read(4) t = fp.read(4)
if(len(t) < 4): if(len(t) < 4):
@ -334,12 +320,6 @@ def _parse_array_openQCD2(d, n, size, wa, quadrupel=False):
return arr return arr
# mimic the read_array routine of openQCD-2.0.
# fp is the opened file handle
# returns the dict array
# at this point we only parse a 2d array
# d = 2
# n = [nfct[irw], 2*nsrc[irw]]
def _read_array_openQCD2(fp): def _read_array_openQCD2(fp):
t = fp.read(4) t = fp.read(4)
d = struct.unpack('i', t)[0] d = struct.unpack('i', t)[0]
@ -400,13 +380,11 @@ def read_qtop(path, prefix, c, dtr_cnfg=1, version="1.2", **kwargs):
last configurations that need to be read (per replicum) last configurations that need to be read (per replicum)
files: list files: list
specify the exact files that need to be read specify the exact files that need to be read
from path, pratical if e.g. only one replicum is needed from path, practical if e.g. only one replicum is needed
names: list names: list
Alternative labeling for replicas/ensembles. Alternative labeling for replicas/ensembles.
Has to have the appropriate length Has to have the appropriate length
""" """
# one could read L from the header in case of sfQCD
# c = 0.35
known_versions = ["1.0", "1.2", "1.4", "1.6", "2.0", "sfqcd"] known_versions = ["1.0", "1.2", "1.4", "1.6", "2.0", "sfqcd"]
if version not in known_versions: if version not in known_versions:
@ -426,11 +404,9 @@ def read_qtop(path, prefix, c, dtr_cnfg=1, version="1.2", **kwargs):
r_start = kwargs.get("r_start") r_start = kwargs.get("r_start")
if "r_stop" in kwargs: if "r_stop" in kwargs:
r_stop = kwargs.get("r_stop") r_stop = kwargs.get("r_stop")
# if one wants to read specific files with this method...
if "files" in kwargs: if "files" in kwargs:
files = kwargs.get("files") files = kwargs.get("files")
else: else:
# find files in path
found = [] found = []
files = [] files = []
for (dirpath, dirnames, filenames) in os.walk(path + "/"): for (dirpath, dirnames, filenames) in os.walk(path + "/"):
@ -441,14 +417,12 @@ def read_qtop(path, prefix, c, dtr_cnfg=1, version="1.2", **kwargs):
if fnmatch.fnmatch(f, prefix + "*" + ".ms.dat"): if fnmatch.fnmatch(f, prefix + "*" + ".ms.dat"):
files.append(f) files.append(f)
print(files) print(files)
# now that we found our files, we dechiffer them...
rep_names = [] rep_names = []
deltas = [] deltas = []
idl = [] idl = []
for rep, file in enumerate(files): for rep, file in enumerate(files):
with open(path + "/" + file, "rb") as fp: with open(path + "/" + file, "rb") as fp:
# header
t = fp.read(12) t = fp.read(12)
header = struct.unpack('<iii', t) header = struct.unpack('<iii', t)
# step size in integration steps "dnms" # step size in integration steps "dnms"
@ -477,7 +451,6 @@ def read_qtop(path, prefix, c, dtr_cnfg=1, version="1.2", **kwargs):
Q = [] Q = []
ncs = [] ncs = []
while 0 < 1: while 0 < 1:
# int nt
t = fp.read(4) t = fp.read(4)
if(len(t) < 4): if(len(t) < 4):
break break
@ -524,8 +497,6 @@ def read_qtop(path, prefix, c, dtr_cnfg=1, version="1.2", **kwargs):
if len(Q_round) != len(ncs) // dtr_cnfg: if len(Q_round) != len(ncs) // dtr_cnfg:
raise Exception("qtops and ncs dont have the same length") raise Exception("qtops and ncs dont have the same length")
# replica = len(files)
truncated_file = file[:-7] truncated_file = file[:-7]
print(truncated_file) print(truncated_file)
idl_start = 1 idl_start = 1
@ -553,7 +524,6 @@ def read_qtop(path, prefix, c, dtr_cnfg=1, version="1.2", **kwargs):
rep_names = names rep_names = names
deltas.append(np.array(Q_round)) deltas.append(np.array(Q_round))
idl.append(range(idl_start, idl_stop)) idl.append(range(idl_start, idl_stop))
# print(idl)
result = Obs(deltas, rep_names, idl=idl) result = Obs(deltas, rep_names, idl=idl)
return result return result
@ -594,7 +564,6 @@ def read_qtop_sector(target=0, **kwargs):
dtr_cnfg = 1 dtr_cnfg = 1
qtop = read_qtop(path, prefix, c, dtr_cnfg=dtr_cnfg, qtop = read_qtop(path, prefix, c, dtr_cnfg=dtr_cnfg,
version=version, **kwargs) version=version, **kwargs)
# unpack to original values, project onto target sector
names = qtop.names names = qtop.names
print(names) print(names)
print(qtop.deltas.keys()) print(qtop.deltas.keys())

48
pyerrors/input/sfcf.py
View file

@ -43,15 +43,16 @@ def read_sfcf(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0,
replaces the name of the ensemble replaces the name of the ensemble
version: str version: str
version of SFCF, with which the measurement was done. version of SFCF, with which the measurement was done.
if the compact output option (-c) was spectified, if the compact output option (-c) was specified,
append a "c" to the version (e.g. "1.0c") append a "c" to the version (e.g. "1.0c")
if the append output option (-a) was specified, if the append output option (-a) was specified,
append an "a" to the version append an "a" to the version. Currently supported versions
are "0.0", "1.0", "2.0", "1.0c", "2.0c", "1.0a" and "2.0a".
replica: list replica: list
list of replica to be read, default is all list of replica to be read, default is all
files: list files: list
list of files to be read per replica, default is all. list of files to be read per replica, default is all.
for non-conpact ouztput format, hand the folders to be read here. for non-compact output format, hand the folders to be read here.
check_configs: check_configs:
list of list of supposed configs, eg. [range(1,1000)] list of list of supposed configs, eg. [range(1,1000)]
for one replicum with 1000 configs for one replicum with 1000 configs
@ -77,17 +78,12 @@ def read_sfcf(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0,
if "replica" in kwargs: if "replica" in kwargs:
reps = kwargs.get("replica") reps = kwargs.get("replica")
# due to higher usage in current projects,
# compact file format is default
compact = True compact = True
appended = False appended = False
# get version string
known_versions = ["0.0", "1.0", "2.0", "1.0c", "2.0c", "1.0a", "2.0a"] known_versions = ["0.0", "1.0", "2.0", "1.0c", "2.0c", "1.0a", "2.0a"]
if version not in known_versions: if version not in known_versions:
raise Exception("This version is not known!") 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"): if(version[-1] == "c"):
appended = False appended = False
compact = True compact = True
@ -119,7 +115,6 @@ def read_sfcf(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0,
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:
# New version, to cope with ids, etc.
ls.sort(key=lambda x: int(re.findall(r'\d+', x[len(prefix):])[0])) ls.sort(key=lambda x: int(re.findall(r'\d+', x[len(prefix):])[0]))
if not appended: if not appended:
@ -135,8 +130,6 @@ def read_sfcf(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0,
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 = [] new_names = []
if not appended: if not appended:
for entry in ls: for entry in ls:
@ -157,7 +150,6 @@ def read_sfcf(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0,
ls.sort(key=lambda x: int(re.findall(r'\d+', x)[-1])) ls.sort(key=lambda x: int(re.findall(r'\d+', x)[-1]))
for entry in ls: for entry in ls:
myentry = entry[:-len(name) - 1] myentry = entry[:-len(name) - 1]
# print(myentry)
try: try:
idx = myentry.index('r') idx = myentry.index('r')
except Exception: except Exception:
@ -167,7 +159,6 @@ def read_sfcf(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0,
new_names.append(kwargs.get('ens_name') + '|' + myentry[idx:]) new_names.append(kwargs.get('ens_name') + '|' + myentry[idx:])
else: else:
new_names.append(myentry[:idx] + '|' + myentry[idx:]) new_names.append(myentry[:idx] + '|' + myentry[idx:])
# print(new_names)
idl = [] idl = []
if not appended: if not appended:
for i, item in enumerate(ls): for i, item in enumerate(ls):
@ -183,7 +174,6 @@ def read_sfcf(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0,
sub_ls.extend(dirnames) sub_ls.extend(dirnames)
break break
# print(sub_ls)
for exc in sub_ls: for exc in sub_ls:
if compact: if compact:
if not fnmatch.fnmatch(exc, prefix + '*'): if not fnmatch.fnmatch(exc, prefix + '*'):
@ -194,7 +184,6 @@ def read_sfcf(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0,
if not fnmatch.fnmatch(exc, 'cfg*'): if not fnmatch.fnmatch(exc, 'cfg*'):
sub_ls = list(set(sub_ls) - set([exc])) sub_ls = list(set(sub_ls) - set([exc]))
sub_ls.sort(key=lambda x: int(x[3:])) sub_ls.sort(key=lambda x: int(x[3:]))
# print(sub_ls)
rep_idl = [] rep_idl = []
no_cfg = len(sub_ls) no_cfg = len(sub_ls)
for cfg in sub_ls: for cfg in sub_ls:
@ -206,26 +195,19 @@ def read_sfcf(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0,
except Exception: except Exception:
raise Exception("Couldn't parse idl from directroy, problem with file " + cfg) raise Exception("Couldn't parse idl from directroy, problem with file " + cfg)
rep_idl.sort() rep_idl.sort()
# maybe there is a better way to print the idls
print(item, ':', no_cfg, ' configurations') print(item, ':', no_cfg, ' configurations')
idl.append(rep_idl) idl.append(rep_idl)
# here we have found all the files we need to look into.
if i == 0: if i == 0:
# here, we want to find the place within the file,
# where the correlator we need is stored.
if compact: if compact:
# to do so, the pattern needed is put together
# from the input values
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)
# and the file is parsed through to find the pattern
with open(path + '/' + item + '/' + sub_ls[0], 'r') as file: with open(path + '/' + item + '/' + sub_ls[0], 'r') as file:
content = file.read() content = file.read()
match = re.search(pattern, content) match = re.search(pattern, content)
if match: if match:
# the start and end point of the correlator # the start and end point of the correlator
# in quaetion is extracted for later use in # in question is extracted for later use in
# the other files # the other files
start_read = content.count('\n', 0, match.start()) + 5 + b2b start_read = content.count('\n', 0, match.start()) + 5 + b2b
end_match = re.search(r'\n\s*\n', content[match.start():]) end_match = re.search(r'\n\s*\n', content[match.start():])
@ -248,7 +230,6 @@ def read_sfcf(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0,
pattern += ", wf_2 " + str(wf2) pattern += ", wf_2 " + str(wf2)
qs = quarks.split(" ") qs = quarks.split(" ")
pattern += " : " + qs[0] + " - " + qs[1] pattern += " : " + qs[0] + " - " + qs[1]
# print(pattern)
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:
@ -265,30 +246,19 @@ def read_sfcf(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0,
start = k + 7 + b2b start = k + 7 + b2b
T -= b2b T -= b2b
print(str(T) + " entries found.") print(str(T) + " entries found.")
# we found where the correlator
# that is to be read is in the files
# after preparing the datastructure
# the correlators get parsed into...
deltas = [] deltas = []
for j in range(T): for j in range(T):
deltas.append([]) deltas.append([])
for t in range(T): for t in range(T):
deltas[t].append(np.zeros(no_cfg)) deltas[t].append(np.zeros(no_cfg))
# ...the actual parsing can start.
# we iterate through all measurement files in the path given... # we iterate through all measurement files in the path given...
if compact: 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:
lines = fp.readlines() lines = fp.readlines()
# check, if the correlator is in fact
# printed completely
if(start_read + T > len(lines)): if(start_read + T > len(lines)):
raise Exception("EOF before end of correlator data! Maybe " + path + '/' + item + '/' + sub_ls[cfg] + " is corrupted?") raise Exception("EOF before end of correlator data! Maybe " + path + '/' + item + '/' + sub_ls[cfg] + " is corrupted?")
# and start to read the correlator.
# the range here is chosen like this,
# since this allows for implementing
# a security check for every read correlator later...
for k in range(start_read - 6, start_read + T): for k in range(start_read - 6, start_read + T):
if k == start_read - 5 - b2b: if k == start_read - 5 - b2b:
if lines[k].strip() != 'name ' + name: if lines[k].strip() != 'name ' + name:
@ -300,10 +270,6 @@ def read_sfcf(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0,
else: else:
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:
# since the non-compatified files
# are typically not so long,
# we can iterate over the whole file.
# here one can also implement the chekc from above.
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()))
@ -320,7 +286,6 @@ def read_sfcf(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0,
if not fnmatch.fnmatch(exc, prefix + '*.' + name): if not fnmatch.fnmatch(exc, prefix + '*.' + name):
ls = list(set(ls) - set([exc])) ls = list(set(ls) - set([exc]))
ls.sort(key=lambda x: int(re.findall(r'\d+', x)[-1])) ls.sort(key=lambda x: int(re.findall(r'\d+', x)[-1]))
# print(ls)
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)
@ -335,7 +300,6 @@ def read_sfcf(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0,
if len(set([data_starts[i] - data_starts[i - 1] for i in if len(set([data_starts[i] - data_starts[i - 1] for i in
range(1, len(data_starts))])) > 1: range(1, len(data_starts))])) > 1:
raise Exception("Irregularities in file structure found, not all runs have the same output length") raise Exception("Irregularities in file structure found, not all runs have the same output length")
# first chunk of data
chunk = content[:data_starts[1]] chunk = content[:data_starts[1]]
for linenumber, line in enumerate(chunk): for linenumber, line in enumerate(chunk):
if line.startswith("gauge_name"): if line.startswith("gauge_name"):
@ -354,12 +318,10 @@ def read_sfcf(path, prefix, name, quarks='.*', noffset=0, wf=0, wf2=0,
deltas.append([]) deltas.append([])
for t in range(T): for t in range(T):
deltas[t].append(np.zeros(len(data_starts))) deltas[t].append(np.zeros(len(data_starts)))
# all other chunks should follow the same structure
for cnfg in range(len(data_starts)): for cnfg in range(len(data_starts)):
start = data_starts[cnfg] start = data_starts[cnfg]
stop = start + data_starts[1] stop = start + data_starts[1]
chunk = content[start:stop] chunk = content[start:stop]
# meta_data = {}
try: try:
rep_idl.append(int(chunk[gauge_line].split("n")[-1])) rep_idl.append(int(chunk[gauge_line].split("n")[-1]))
except Exception: except Exception: