def __init__(
self,
workdir,
fmu_path,
inp,
known,
est,
ideal,
lp_n=None,
lp_len=None,
lp_frame=None,
vp=None,
ic_param=None,
methods=("MODESTGA", "PS"),
ga_opts={},
ps_opts={},
scipy_opts={},
modestga_opts={},
ftype="RMSE",
default_log=True,
logfile="modestpy.log",
):
# Default logging configuration?
if default_log:
config_logger(filename=logfile, level="WARNING")
self.logger = logging.getLogger(type(self).__name__)
# Sanity checks
assert inp.index.equals(
ideal.index), "inp and ideal indexes are not matching"
init, lo, hi = 0, 1, 2 # Init. value, lower bound, upper bound indices
for v in est:
assert (est[v][init] >= est[v][lo]) and (
est[v][init] <=
est[v][hi]), "Initial value out of limits ({})".format(v)
# Input data
self.workdir = workdir
self.fmu_path = fmu_path
self.inp = inp
self.known = known
self.est = est
self.ideal = ideal
self.methods = methods
self.ftype = ftype
# Results placeholders
self.best_per_run = pd.DataFrame()
self.final = pd.DataFrame()
# Estimation options
# GA options
self.GA_OPTS = {
"maxiter": 50,
"pop_size": max((4 * len(est.keys()), 20)),
"tol": 1e-6,
"mut": 0.10,
"mut_inc": 0.33,
"uniformity": 0.5,
"look_back": 50,
"lhs": False,
"ftype": ftype,
} # Default
# Default
self.GA_OPTS["trm_size"] = max(self.GA_OPTS["pop_size"] // 6, 2)
# User options
self.GA_OPTS = self._update_opts(self.GA_OPTS, ga_opts, "GA_LEGACY")
# PS options
self.PS_OPTS = {
"maxiter": 500,
"rel_step": 0.02,
"tol": 1e-11,
"try_lim": 1000,
"ftype": ftype,
} # Default
# User options
self.PS_OPTS = self._update_opts(self.PS_OPTS, ps_opts, "PS")
# SCIPY options
self.SCIPY_OPTS = {
"solver": "L-BFGS-B",
"options": {
"disp": True,
"iprint": 2,
"maxiter": 150,
"full_output": True
},
"ftype": ftype,
} # Default
# User options
self.SCIPY_OPTS = self._update_opts(self.SCIPY_OPTS, scipy_opts,
"SCIPY")
# MODESTGA options
self.MODESTGA_OPTS = {
"workers": 3, # CPU cores to use
"generations": 50, # Max. number of generations
"pop_size": 30, # Population size
"mut_rate": 0.01, # Mutation rate
"trm_size": 3, # Tournament size
"tol": 1e-3, # Solution tolerance
"inertia": 100, # Max. number of non-improving generations
"ftype": ftype,
} # Default
# User options
self.MODESTGA_OPTS = self._update_opts(self.MODESTGA_OPTS,
modestga_opts, "MODESTGA")
# Method dictionary
self.method_dict = {
"MODESTGA": (MODESTGA, self.MODESTGA_OPTS),
"GA_LEGACY": (GA, self.GA_OPTS),
"PS": (PS, self.PS_OPTS),
"SCIPY": (SCIPY, self.SCIPY_OPTS),
} # Key -> method name, value -> (method class, method options)
# List of learning periods (tuples with start, stop)
self.lp = self._select_lp(lp_n, lp_len, lp_frame)
# Validation period (a tuple with start, stop)
if vp is not None:
self.vp = vp
else:
self.vp = (ideal.index[0], ideal.index[-1])
# Initial condition parameters
# Take first value from time series from 'ideal' column
self.ic_param = ic_param # dict (par_name: ideal_column_name)
#!/usr/bin/env python
from modestpy.test import run
from modestpy.loginit import config_logger
if __name__ == "__main__":
config_logger(filename='unit_tests.log', level='DEBUG')
run.tests()
random.seed(1)
np.random.seed(1)
ga = GA(
self.fmu_path,
self.inp,
self.known,
self.est,
self.ideal,
maxiter=self.gen,
lhs=True,
)
indiv = ga.pop.individuals
par2 = list()
for i in indiv:
par2.append(i.get_estimates(as_dict=True))
for d1, d2 in zip(par1, par2):
self.assertDictEqual(d1, d2)
def suite():
suite = unittest.TestSuite()
suite.addTest(TestGA("test_ga"))
suite.addTest(TestGA("test_init_pop"))
return suite
if __name__ == "__main__":
config_logger(filename="unit_tests.log", level="DEBUG")
unittest.main()
def __init__(self,
workdir,
fmu_path,
inp,
known,
est,
ideal,
lp_n=None,
lp_len=None,
lp_frame=None,
vp=None,
ic_param=None,
methods=('MODESTGA', 'PS'),
ga_opts={},
ps_opts={},
scipy_opts={},
modestga_opts={},
ftype='RMSE',
default_log=True,
logfile='modestpy.log'):
# Default logging configuration?
if default_log:
config_logger(filename=logfile, level='WARNING')
self.logger = logging.getLogger(type(self).__name__)
# Sanity checks
assert inp.index.equals(ideal.index), \
'inp and ideal indexes are not matching'
init, lo, hi = 0, 1, 2 # Init. value, lower bound, upper bound indices
for v in est:
assert (est[v][init] >= est[v][lo]) \
and (est[v][init] <= est[v][hi]), \
'Initial value out of limits ({})'.format(v)
# Input data
self.workdir = workdir
self.fmu_path = fmu_path
self.inp = inp
self.known = known
self.est = est
self.ideal = ideal
self.methods = methods
self.ftype = ftype
# Results placeholders
self.best_per_run = pd.DataFrame()
self.final = pd.DataFrame()
# Estimation options
# GA options
self.GA_OPTS = {
'maxiter': 50,
'pop_size': max((4 * len(est.keys()), 20)),
'tol': 1e-6,
'mut': 0.10,
'mut_inc': 0.33,
'uniformity': 0.5,
'look_back': 50,
'lhs': False,
'ftype': ftype
} # Default
# Default
self.GA_OPTS['trm_size'] = max(self.GA_OPTS['pop_size'] // 6, 2)
# User options
self.GA_OPTS = self._update_opts(self.GA_OPTS, ga_opts, 'GA_LEGACY')
# PS options
self.PS_OPTS = {
'maxiter': 500,
'rel_step': 0.02,
'tol': 1e-11,
'try_lim': 1000,
'ftype': ftype
} # Default
# User options
self.PS_OPTS = self._update_opts(self.PS_OPTS, ps_opts, 'PS')
# SCIPY options
self.SCIPY_OPTS = {
'solver': 'L-BFGS-B',
'options': {
'disp': True,
'iprint': 2,
'maxiter': 150,
'full_output': True
},
'ftype': ftype
} # Default
# User options
self.SCIPY_OPTS = \
self._update_opts(self.SCIPY_OPTS, scipy_opts, 'SCIPY')
# MODESTGA options
self.MODESTGA_OPTS = {
'workers': 3, # CPU cores to use
'generations': 50, # Max. number of generations
'pop_size': 30, # Population size
'mut_rate': 0.01, # Mutation rate
'trm_size': 3, # Tournament size
'tol': 1e-3, # Solution tolerance
'inertia': 100, # Max. number of non-improving generations
'ftype': ftype
} # Default
# User options
self.MODESTGA_OPTS = \
self._update_opts(self.MODESTGA_OPTS, modestga_opts, 'MODESTGA')
# Method dictionary
self.method_dict = {
'MODESTGA': (MODESTGA, self.MODESTGA_OPTS),
'GA_LEGACY': (GA, self.GA_OPTS),
'PS': (PS, self.PS_OPTS),
'SCIPY': (SCIPY, self.SCIPY_OPTS)
} # Key -> method name, value -> (method class, method options)
# List of learning periods (tuples with start, stop)
self.lp = self._select_lp(lp_n, lp_len, lp_frame)
# Validation period (a tuple with start, stop)
if vp is not None:
self.vp = vp
else:
self.vp = (ideal.index[0], ideal.index[-1])
# Initial condition parameters
# Take first value from time series from 'ideal' column
self.ic_param = ic_param # dict (par_name: ideal_column_name)
def __init__(self,
workdir,
fmu_path,
inp,
known,
est,
ideal,
lp_n=None,
lp_len=None,
lp_frame=None,
vp=None,
ic_param=None,
methods=('GA', 'PS'),
ga_opts={},
ps_opts={},
scipy_opts={},
fmi_opts={},
ftype='RMSE',
seed=None,
default_log=True,
logfile='modestpy.log'):
"""
Index in DataFrames ``inp`` and ``ideal`` must be named 'time'
and given in seconds. The index name assertion check is
implemented to avoid situations in which a user reads DataFrame
from a csv and forgets to use ``DataFrame.set_index(column_name)``
(it happens quite often...). TODO: Check index name assertion.
Currently available estimation methods:
- GA - genetic algorithm
- PS - pattern search (Hooke-Jeeves)
- SCIPY - interface to algorithms available through
scipy.optimize.minimize()
Parameters:
-----------
workdir: str
Output directory, must exist
fmu_path: str
Absolute path to the FMU
inp: pandas.DataFrame
Input data, index given in seconds and named ``time``
known: dict(str: float)
Dictionary with known parameters (``parameter_name: value``)
est: dict(str: tuple(float, float, float))
Dictionary defining estimated parameters,
(``par_name: (guess value, lo limit, hi limit)``)
ideal: pandas.DataFrame
Ideal solution (usually measurements),
index in seconds and named ``time``
lp_n: int or None
Number of learning periods, one if ``None``
lp_len: float or None
Length of a single learning period, entire ``lp_frame`` if ``None``
lp_frame: tuple of floats or None
Learning period time frame, entire data set if ``None``
vp: tuple(float, float) or None
Validation period, entire data set if ``None``
ic_param: dict(str, str) or None
Mapping between model parameters used for IC and variables from
``ideal``
methods: tuple(str, str)
List of methods to be used in the pipeline
ga_opts: dict
Genetic algorithm options
ps_opts: dict
Pattern search options
scipy_opts: dict
SciPy solver options
fmi_opts: dict
Additional options to be passed to the FMI model
(e.g. solver tolerance)
ftype: string
Cost function type. Currently 'NRMSE' (advised for multi-objective
estimation) or 'RMSE'.
seed: None or int
Random number seed. If None, current time or OS specific
randomness is used.
default_log: bool
If true, use default logging settings. Use false if you want to
use own logging.
logfile: str
If default_log=True, this argument can be used to specify the log
file name
"""
# Default logging configuration?
if default_log:
config_logger(filename=logfile, level='DEBUG')
self.logger = logging.getLogger(type(self).__name__)
# Sanity checks
assert inp.index.equals(ideal.index), \
'inp and ideal indexes are not matching'
init, lo, hi = 0, 1, 2 # Init. value, lower bound, upper bound indices
for v in est:
assert (est[v][init] >= est[v][lo]) \
and (est[v][init] <= est[v][hi]), \
'Initial value out of limits ({})'.format(v)
# Random seed
if seed is not None:
self.logger.info('Setting random seed: {}'.format(seed))
random.seed(seed)
np.random.seed(seed) # Important for other libraries, like pyDOE
# Input data
self.workdir = workdir
self.fmu_path = fmu_path
self.inp = inp
self.known = known
self.est = est
self.ideal = ideal
self.methods = methods
self.ftype = ftype
# Results placeholders
self.best_per_run = pd.DataFrame()
self.final = pd.DataFrame()
# Estimation options
# GA options
self.GA_OPTS = {
'maxiter': 50,
'pop_size': max((4 * len(est.keys()), 20)),
'tol': 1e-6,
'mut': 0.10,
'mut_inc': 0.33,
'uniformity': 0.5,
'look_back': 50,
'lhs': False,
'ftype': ftype,
'fmi_opts': fmi_opts
} # Default
# Default
self.GA_OPTS['trm_size'] = max(self.GA_OPTS['pop_size'] // 6, 2)
# User options
self.GA_OPTS = self._update_opts(self.GA_OPTS, ga_opts, 'GA')
# PS options
self.PS_OPTS = {
'maxiter': 500,
'rel_step': 0.02,
'tol': 1e-11,
'try_lim': 1000,
'ftype': ftype,
'fmi_opts': fmi_opts
} # Default
# User options
self.PS_OPTS = self._update_opts(self.PS_OPTS, ps_opts, 'PS')
# SCIPY options
self.SCIPY_OPTS = {
'solver': 'L-BFGS-B',
'options': {
'disp': True,
'iprint': 2,
'maxiter': 150,
'full_output': True
},
'ftype': ftype,
'fmi_opts': fmi_opts
} # Default
# User options
self.SCIPY_OPTS = \
self._update_opts(self.SCIPY_OPTS, scipy_opts, 'SCIPY')
# Method dictionary
self.method_dict = {
'GA': (GA, self.GA_OPTS),
'PS': (PS, self.PS_OPTS),
'SCIPY': (SCIPY, self.SCIPY_OPTS)
} # Key -> method name, value -> (method class, method options)
# List of learning periods (tuples with start, stop)
self.lp = self._select_lp(lp_n, lp_len, lp_frame)
# Validation period (a tuple with start, stop)
if vp is not None:
self.vp = vp
else:
self.vp = (ideal.index[0], ideal.index[-1])
# Initial condition parameters
# Take first value from time series from 'ideal' column
self.ic_param = ic_param # dict (par_name: ideal_column_name)
This code is licensed under BSD 2-clause license.
See LICENSE file in the project root for license terms.
"""
import json
import os
import pandas as pd
from modestpy import Estimation
from modestpy.utilities.sysarch import get_sys_arch
from modestpy.loginit import config_logger
if __name__ == "__main__":
"""
This file is supposed to be run from the root directory.
Otherwise the paths have to be corrected.
"""
config_logger(filename='simple.log', level='DEBUG')
# DATA PREPARATION ==============================================
# Resources
platform = get_sys_arch()
assert platform, 'Unsupported platform type!'
fmu_file = 'Simple2R1C_ic_' + platform + '.fmu'
fmu_path = os.path.join('examples', 'simple', 'resources', fmu_file)
inp_path = os.path.join('examples', 'simple', 'resources', 'inputs.csv')
ideal_path = os.path.join('examples', 'simple', 'resources', 'result.csv')
est_path = os.path.join('examples', 'simple', 'resources', 'est.json')
known_path = os.path.join('examples', 'simple', 'resources', 'known.json')
# Working directory
workdir = os.path.join('examples', 'simple', 'workdir')
"""
import json
import os
import pandas as pd
from modestpy import Estimation
from modestpy.loginit import config_logger
from modestpy.utilities.sysarch import get_sys_arch
if __name__ == "__main__":
"""
This file is supposed to be run from the root directory.
Otherwise the paths have to be corrected.
"""
config_logger(filename="simple.log", level="DEBUG")
# DATA PREPARATION ==============================================
# Resources
platform = get_sys_arch()
assert platform, "Unsupported platform type!"
fmu_file = "Simple2R1C_ic_" + platform + ".fmu"
fmu_path = os.path.join("examples", "simple", "resources", fmu_file)
inp_path = os.path.join("examples", "simple", "resources", "inputs.csv")
ideal_path = os.path.join("examples", "simple", "resources", "result.csv")
est_path = os.path.join("examples", "simple", "resources", "est.json")
known_path = os.path.join("examples", "simple", "resources", "known.json")
# Working directory
workdir = os.path.join("examples", "simple", "workdir")