"""
This module analyzes and manipulates multistate accumulators.
For example, the canonical ensemble average energy for a given number of particles.
"""
from pathlib import Path
import pandas as pd
[docs]def splice_by_max_column(prefix, suffix, column='moment0', crit_prefix=None, crit_suffix=None):
"""
Combine all files with matching prefix and suffix, each with the same number of states,
by assigning rows based on the maximum in the given column.
:param str prefix: Find all files beginning with this prefix.
:param str suffix: Find all files endding with this suffix.
:param str column:
Use the row with the maximum column. For example, moment0 is the highest number of samples.
:param str crit_prefix: Use criteria to read only specific macrostates, if given.
:param str crit_suffix: Use criteria to read only specific macrostates, if given.
>>> import pandas as pd
>>> from pyfeasst import multistate_accumulator
>>> spliced = multistate_accumulator.splice_by_max_column(prefix="../../tests/lj_enn0s",
... suffix='.txt')
>>> en0 = pd.read_csv('../../tests/lj_enn0s00.txt')
>>> en1 = pd.read_csv('../../tests/lj_enn0s01.txt')
>>> assert spliced['average'][136] == en0['average'][136]
>>> assert spliced['average'][136] != en1['average'][136]
>>> assert spliced['average'][137] == en0['average'][137]
>>> assert spliced['average'][137] == en1['average'][137]
>>> assert spliced['average'][138] != en0['average'][138]
>>> assert spliced['average'][138] == en1['average'][138]
"""
spliced = None
if crit_prefix is None:
for filename in Path('.').rglob(prefix+'*'+suffix):
if spliced is None:
spliced = pd.read_csv(filename)
else:
addition = pd.read_csv(filename)
newer = addition[addition[column] > spliced[column]]
if len(newer) > 0:
spliced.loc[newer['state'].values[0]: newer['state'].values[-1]] = newer
else:
rows = list()
for filename in sorted(Path('.').rglob(crit_prefix+'*'+crit_suffix)):
print('filename', filename)
with open(filename, 'r') as file1:
lines = file1.readlines()
exec('iprm={' + lines[0][1:] + '}', globals())
rows.append([iprm['soft_min'], iprm['soft_max']])
#print('rows', rows)
ifile = 0
lines=list()
for filename in sorted(Path('.').rglob(prefix+'*'+suffix)):
#print('filename', filename, 'ifile', ifile)
with open(filename, 'r') as file1:
tlines = file1.readlines()
if ifile == 0: lines += tlines[0]
lines += tlines[rows[ifile][0]+1:rows[ifile][1]+2]
ifile += 1
#print(lines)
with open(prefix+suffix+'_agg.csv', 'w') as file1:
for line in lines: file1.write(line)
spliced = pd.read_csv(prefix+suffix+'_agg.csv', usecols=range(0, 5))
return spliced
[docs]def splice_by_node(prefix, suffix, num_nodes, extra_overlap=0):
"""
Use splice_by_max_column for each node, with prefix=prefix+node.
Then, drop 1+extra_overlap
Combine all files with matching prefix and suffix, each with the same number of states,
by assigning rows based on the maximum in the given column.
:param int num_nodes: The number of nodes to splice.
>>> import pandas as pd
>>> from pyfeasst import multistate_accumulator
>>> from pyfeasst import macrostate_distribution
>>> spliced = splice_by_node(prefix='../../tests/lj_enn', suffix='.txt', num_nodes=2)
>>> round(spliced['average'][375], 8)
-2001.76687973
>>> round(spliced['average'][376], 8)
-2012.46764871
>>> len(spliced)
476
>>> spliced.to_csv('spliced.csv')
>>> lnpi = macrostate_distribution.splice_files(prefix='../../tests/lj_lnpin', suffix='.txt')
>>> lnpi.concat_dataframe(spliced, add_prefix='e_')
>>> round(lnpi.equilibrium(), 8)
-0.31402411
>>> vapor, liquid = lnpi.split()
>>> round(-vapor.ln_prob()[0]*0.7/8**3, 8) # pressure
0.00136904
>>> round(vapor.ensemble_average('e_average')/vapor.average_macrostate(), 8)
-0.02500369
>>> round(liquid.ensemble_average('e_average')/liquid.average_macrostate(), 8)
-6.09838831
"""
node_data = list()
for node in range(num_nodes):
dat = splice_by_max_column(prefix=prefix+str(node), suffix=suffix)
if node > 0:
dat.drop(dat.head(1 + extra_overlap).index, inplace=True)
dat['state'] += node_data[-1]['state'].values[-1]
node_data.append(dat)
data = pd.concat(node_data)
data.reset_index(inplace=True)
block = 0
while 'block'+str(block) in data.columns:
cname = 'block'+str(block)
# if data[cname].isnull().values.any():
# data.drop([cname], inplace=True, axis=1)
block += 1
return data
if __name__ == "__main__":
import doctest
doctest.testmod()