def update_config(self):
self.config = PRACMLNConfig()
self.config["use_emln"] = self.use_emln.get()
self.config['mln'] = self.mln_container.selected_file.get().strip(
).lstrip('*')
self.config['emln'] = self.emln_container.selected_file.get().strip(
).lstrip('*')
self.config["db"] = self.db_container.selected_file.get().strip(
).lstrip('*')
self.config["method"] = InferenceMethods.id(
self.selected_method.get().strip())
self.config["params"] = self.params.get().strip()
self.config["queries"] = self.query.get()
self.config["output_filename"] = self.output_filename.get().strip()
self.config["cw"] = self.closed_world.get()
self.config["cw_preds"] = self.cwPreds.get()
self.config['profile'] = self.profile.get()
self.config['logic'] = self.selected_logic.get()
self.config['grammar'] = self.selected_grammar.get()
self.config['multicore'] = self.multicore.get()
self.config['save'] = self.save.get()
self.config['ignore_unknown_preds'] = self.ignore_unknown_preds.get()
self.config['verbose'] = self.verbose.get()
self.config['window_loc'] = self.master.winfo_geometry()
self.config['dir'] = self.dir
self.project.queryconf = PRACMLNConfig()
self.project.queryconf.update(self.config.config.copy())
def update_config(self):
out('update_config')
self.config = PRACMLNConfig()
self.config['mln'] = self.mln_container.selected_file.get().strip().lstrip('*')
self.config["db"] = self.db_container.selected_file.get().strip().lstrip('*')
self.config["output_filename"] = self.output_filename.get()
self.config["params"] = self.params.get().strip()
self.config["method"] = LearningMethods.id(self.selected_method.get().strip())
self.config["pattern"] = self.pattern.get()
self.config["use_prior"] = int(self.use_prior.get())
self.config["prior_mean"] = self.priorMean.get()
self.config["prior_stdev"] = self.priorStdDev.get()
self.config["incremental"] = int(self.incremental.get())
self.config["shuffle"] = int(self.shuffle.get())
self.config["use_initial_weights"] = int(self.use_initial_weights.get())
self.config["qpreds"] = self.queryPreds.get().strip()
self.config["epreds"] = self.evidencePreds.get().strip()
self.config["discr_preds"] = self.discrPredicates.get()
self.config['logic'] = self.selected_logic.get()
self.config['grammar'] = self.selected_grammar.get()
self.config['multicore'] = self.multicore.get()
self.config['profile'] = self.profile.get()
self.config['verbose'] = self.verbose.get()
self.config['ignore_unknown_preds'] = self.ignore_unknown_preds.get()
self.config['ignore_zero_weight_formulas'] = self.ignore_zero_weight_formulas.get()
self.config['save'] = self.save.get()
self.config["output_filename"] = self.output_filename.get().strip()
self.project.learnconf = PRACMLNConfig()
self.project.learnconf.update(self.config.config.copy())
def main():
logger.level = logs.DEBUG
usage = 'PRACMLN Learning Tool'
parser = argparse.ArgumentParser(description=usage)
parser.add_argument("--run", action="store_true", dest="run", default=False, help="run last configuration without showing gui")
parser.add_argument("-d", "--dir", dest="directory", action='store', help="the directory to start the tool from", metavar="FILE", type=str)
parser.add_argument("-i", "--mln-filename", dest="mlnarg", action='store', help="input MLN filename", metavar="FILE", type=str)
parser.add_argument("-t", "--db-filename", dest="dbarg", action='store', help="training database filename", metavar="FILE", type=str)
parser.add_argument("-o", "--output-file", dest="outputfile", action='store', help="output MLN filename", metavar="FILE", type=str)
args = parser.parse_args()
opts_ = vars(args)
# run learning task/GUI
root = Tk()
conf = PRACMLNConfig(DEFAULT_CONFIG)
app = MLNLearnGUI(root, conf, directory=os.path.abspath(args.directory) if args.directory is not None else None)
if args.run:
logger.debug('running mlnlearn without gui')
app.learn(savegeometry=False, options=opts_)
else:
root.mainloop()
def main():
logger.level = logs.DEBUG
usage = 'PRACMLN Query Tool'
parser = argparse.ArgumentParser(description=usage)
parser.add_argument("-i", "--mln", dest="mlnarg", help="the MLN model file to use")
parser.add_argument("-d", "--dir", dest="directory", action='store', help="the directory to start the tool from", metavar="FILE", type=str)
parser.add_argument("-x", "--emln", dest="emlnarg", help="the MLN model extension file to use")
parser.add_argument("-q", "--queries", dest="queryarg", help="queries (comma-separated)")
parser.add_argument("-e", "--evidence", dest="dbarg", help="the evidence database file")
parser.add_argument("-r", "--results-file", dest="outputfile", help="the results file to save")
parser.add_argument("--run", action="store_true", dest="run", default=False, help="run with last settings (without showing GUI)")
args = parser.parse_args()
opts_ = vars(args)
root = Tk()
conf = PRACMLNConfig(DEFAULT_CONFIG)
app = MLNQueryGUI(root, conf, directory=os.path.abspath(args.directory) if args.directory else None)
if args.run:
logger.debug('running mlnlearn without gui')
app.infer(savegeometry=False, options=opts_)
else:
root.mainloop()
def activate_model(self,database, mln):
"""
Returns the learned mln
--Inputs--
database and mln objects in MLN format
"""
DEFAULT_CONFIG = os.path.join(locs.user_data, global_config_filename)
conf = PRACMLNConfig(DEFAULT_CONFIG)
config = {}
config['verbose'] = True
config['discr_preds'] = 0
config['db'] = database
config['mln'] = mln
config['ignore_zero_weight_formulas'] = 1 #0
config['ignore_unknown_preds'] = True #0
config['incremental'] = 1 #0
config['grammar'] = 'PRACGrammar' #StandardGrammar
config['logic'] = 'FirstOrderLogic'
config['method'] = 'BPLL' # BPLL (BPLL, 'pseudo-log-likelihood')
config['optimizer'] = 'bfgs'
config['multicore'] = False
config['profile'] = 0
config['shuffle'] = 0
config['prior_mean'] = 0
config['prior_stdev'] = 10 # 5
config['save'] = True
config['use_initial_weights'] = 0
config['use_prior'] = 0
config['infoInterval'] = 500
config['resultsInterval'] = 1000
conf.update(config)
print('training...')
learn = MLNLearn(conf, mln=mln, db=database)
result = learn.run()
print('finished...')
return result
def activate_model(self, database, mln):
DEFAULT_CONFIG = os.path.join(locs.user_data, global_config_filename)
conf = PRACMLNConfig(DEFAULT_CONFIG)
config = {}
config['verbose'] = True
config['discr_preds'] = 0
config['db'] = database
config['mln'] = mln
config['ignore_zero_weight_formulas'] = 1 # 0
config['ignore_unknown_preds'] = True # 0
config['incremental'] = 1 # 0
config['grammar'] = 'StandardGrammar'
config['logic'] = 'FirstOrderLogic'
config['method'] = 'BPLL' # BPLL
config['multicore'] = False
config['profile'] = 0
config['shuffle'] = 0
config['prior_mean'] = 0
config['prior_stdev'] = 10 # 5
config['save'] = True
config['use_initial_weights'] = 0
config['use_prior'] = 0
# config['output_filename'] = 'learnt.dbpll_cg.student-new-train-student-new-2.mln'
# 亲测无效, 此句没法储存.mln 档案
config['infoInterval'] = 500
config['resultsInterval'] = 1000
conf.update(config)
print('training...')
learn = MLNLearn(conf, mln=mln, db=database)
# learn.output_filename(r'C:\Users\anaconda3 4.2.0\test.mln')
# 亲测无效, 此句没法储存.mln 档案
result = learn.run()
print('finished...')
return result
"--queries",
dest="queryarg",
help="queries (comma-separated)")
parser.add_option("-e",
"--evidence",
dest="dbarg",
help="the evidence database file")
parser.add_option("-r",
"--results-file",
dest="outputfile",
help="the results file to save")
parser.add_option("--run",
action="store_true",
dest="run",
default=False,
help="run with last settings (without showing GUI)")
(opts, args) = parser.parse_args()
options = vars(opts)
root = Tk()
conf = PRACMLNConfig(DEFAULT_CONFIG)
app = MLNQueryGUI(root,
conf,
directory=os.path.abspath(args[0]) if args else None)
if opts.run:
logger.debug('running mlnlearn without gui')
app.infer(savegeometry=False, options=options)
else:
root.mainloop()
def main():
logger.level = logs.DEBUG
usage = 'PRAC Query Tool'
parser = argparse.ArgumentParser(description=usage)
parser.add_argument("instruction", help="The instruction.")
parser.add_argument(
"-i",
"--interactive",
dest="interactive",
default=False,
action='store_true',
help="Starts PRAC inference with an interactive GUI tool.")
parser.add_argument("-v",
"--verbose",
dest="verbose",
default=1,
type=int,
action="store",
help="Set verbosity level {0..3}. Default is 1.")
args = parser.parse_args()
opts_ = vars(args)
sentences = args.instruction
prac = PRAC()
prac.verbose = args.verbose
conf = PRACMLNConfig(DEFAULT_CONFIG)
if args.interactive: # use the GUI
from tkinter import Tk
root = Tk()
# in case we have natural-language parameters, parse them
infer = PRACInference(prac, sentences)
if len(sentences) > 0:
# module = prac.module('nl_parsing')
# prac.run(infer, module)
n = infer.runstep()
# print parsing result
for odb in n.outdbs:
odb.write()
# print input sentence
print(n.nlinstr())
#Started control structure handling
'''
cs_recognition = prac.module('cs_recognition')
prac.run(inference, cs_recognition)
dbs = inference.inference_steps[-1].output_dbs
dbs_ = []
for db in dbs:
dbs_.extend(parser.extract_multiple_action_cores(db))
inference.inference_steps[-1].output_dbs = dbs_
'''
app = PRACQueryGUI(root,
infer.prac,
n,
conf,
directory=args[0] if args else None)
root.mainloop()
exit(0)
# regular PRAC pipeline
infer = PRACInference(prac, sentences)
infer.run()
print(headline('inference results'))
print('instructions:')
for i in infer.root:
print(i)
frames = []
for step in infer.steps():
print(step.frame)
print(prac_heading('cram plans', color='blue'))
for step in infer.steps():
if hasattr(step, 'plan'):
print(step.plan)
# infer.write()
exit(0)
