def entry_point():
"""
Application starting point which parses command line.
Can be used from other modules too.
"""
import plac
plac.call(main)
def test_kwargs():
def main(opt, arg1, *args, **kw):
print(opt, arg1)
return args, kw
main.__annotations__ = dict(opt=('Option', 'option'))
argskw = plac.call(main, ['arg1', 'arg2', 'a=1', 'b=2'])
assert argskw == [('arg2',), {'a': '1', 'b': '2'}], argskw
argskw = plac.call(main, ['arg1', 'arg2', 'a=1', '-o', '2'])
assert argskw == [('arg2',), {'a': '1'}], argskw
expect(SystemExit, plac.call, main, ['arg1', 'arg2', 'a=1', 'opt=2'])
def main(verbose, interactive, multiline, serve, batch, test, fname=None,
*extra):
"Runner for plac tools, plac batch files and plac tests"
baseparser = plac.parser_from(main)
if fname is None:
baseparser.print_help()
elif sys.argv[1] == fname: # script mode
plactool = plac.import_main(fname)
plactool.prog = os.path.basename(sys.argv[0]) + ' ' + fname
out = plac.call(plactool, sys.argv[2:], eager=False)
if plac.iterable(out):
for output in out:
print(output)
else:
print(out)
elif interactive or multiline or serve:
plactool = plac.import_main(fname, *extra)
plactool.prog = ''
i = plac.Interpreter(plactool)
if interactive:
i.interact(verbose=verbose)
elif multiline:
i.multiline(verbose=verbose)
elif serve:
i.start_server(serve)
elif batch:
run((fname,) + extra, 'execute', verbose)
elif test:
run((fname,) + extra, 'doctest', verbose)
print('run %s plac test(s)' % (len(extra) + 1))
else:
baseparser.print_usage()
def call(main, default_level='INFO', stream=sys.stdout, process_names=False):
'''Enable logging and start up a main method.
Parameters
----------
main : callable
The main method to invoke after initialization.
default_level : str, optional
Logging level. Defaults to INFO.
stream : file-like, optional
Stream for logging output. Defaults to ``sys.stdout``.
process_names : bool, optional
If True, include process names in logging output. Defaults to False.
'''
enable_default_logging(
default_level=default_level,
stream=stream,
process_names=process_names)
from . import flags
if flags.PARSER is None:
return plac.call(main)
args, rest = parse_known_args()
if rest:
logging.debug('unknown arguments: %s', rest)
logging.debug('running with arguments:')
kwargs = vars(args)
for k in sorted(kwargs):
logging.debug('--%s = %s', k, kwargs[k])
return main(args)
def manage():
@plac.annotations(
config_file=plac.Annotation("paster config file", 'positional', None, str),
include_file=plac.Annotation("include file, containing test file to run (for pycharm interaction)", 'option',
'i', str),
)
def _manage(config_file, include_file=None, *args):
"""
manage entrypoint for django, using the paster configfile
``*args`` must be kept here to allow dynamic arguments in plac
"""
# when we're operating in a cli, we don't have a request id, but still want to have an unique id.
# therefore we'll create one here
# print ('create new uuid: %s' % fallback_uuid())
# filter out all arguments that have been applied by plac already
leftover_args = [arg for arg in sys.argv if not arg in (config_file, include_file, '-i', '--include-file')]
# override ``sys.argv`` as nose or django will directly access them
sys.argv = leftover_args
log.debug('bootstrapping from config: %s', config_file)
boostrap_from_config(config_file)
if include_file:
with file(include_file) as f:
additional_file_arg = f.read().splitlines()[0]
leftover_args.append(additional_file_arg)
log.debug('leftover arguments: %s', leftover_args)
from django.core import management
from django.db.utils import OperationalError
try:
management.execute_from_command_line(leftover_args)
except OperationalError as e:
log.error('db error: %s', e)
from django.db import connection
cursor = connection.cursor()
cursor.execute("SELECT * FROM pg_stat_activity")
for row in cursor.fetchall():
log.error(' %s', row)
log.error('ERROR: %s', e)
# TODO amb: dunno why this happens, but it constantly fails the build
# raise e
plac.call(_manage, eager=False)
def entry_point():
"""
Application starting point which parses command line.
Can be used from other modules too.
"""
exit_code = plac.call(main)
sys.exit(exit_code)
def call_plac(f):
"Decorator to create a simple CLI from `func` using `plac`"
name = inspect.currentframe().f_back.f_globals['__name__']
if name == '__main__':
import plac
res = plac.call(f)
if callable(res): res()
else: return f
def entry_point():
"""
Enter the via setup.py entry_point declaration.
Handle UNIX style application exit values
"""
import plac
exitcode = plac.call(main)
sys.exit(exitcode)
def main(interactive, *subcommands):
"""
This script works both interactively and non-interactively.
Use .help to see the internal commands.
"""
if interactive:
plac.Interpreter(ishelve.main).interact()
else:
for out in plac.call(ishelve.main, subcommands):
print(out)
def main():
"""
Defined as the `gossamer` command in setup.py.
Runs the argument parser and passes settings to
:func:`gossamer.main.dispatcher`.
"""
try:
sys.exit(plac.call(initialize))
except KeyboardInterrupt:
sys.stdout.write('\n')
sys.stdout.flush()
sys.exit(1)
def call(obj, arglist=sys.argv[1:], eager=True, config=None,
default_section=None, gnu=True):
if gnu:
plac.ArgumentParser.add_argument = add_gnu_argument
if config is None:
return plac.call(obj, arglist=arglist, eager=eager)
argparser = plac.parser_from(obj)
argnames = argparser.argspec.args
defaults = argparser.argspec.defaults
cp = config_parser_from(obj, config, default_section)
cfg = dict(zip_longest(argnames, defaults))
ini_values = _read_config(cp, config, default_section)
for k in obj.__annotations__.keys():
a = plac.Annotation.from_(obj.__annotations__[k])
if a.type and k in ini_values:
if a.type is type(True):
try:
ini_values[k] = cp._convert_to_boolean(ini_values[k])
except ValueError:
argparser.print_usage(sys.stderr)
_print_exit(
"{}: error: {}={} failed conversion to <type 'bool'> in:\n{}\n".format(
argparser.prog, k, ini_values[k], config))
else:
try:
ini_values[k] = a.type(ini_values[k])
except ValueError:
argparser.print_usage(sys.stderr)
_print_exit(
'{}: error: {}={} failed conversion to {} in:\n{}\n'.format(
argparser.prog, k, ini_values[k], a.type, config))
cfg.update(ini_values)
if sys.version >= '3':
items = cfg.items()
else:
items = cfg.iteritems()
argparser.set_defaults(**dict((k, v) for k, v in items))
cmd, result = argparser.consume(arglist)
if plac.iterable(result) and eager: # listify the result
return list(result)
return result
def main():
plac.call(entry_point)
yield "%dM iterations" % n
x, y = random(), random()
if x * x + y * y < 1:
counts += 1
yield (4.0 * counts) / npoints
def run(self):
tasks = self.i.tasks()
for t in tasks:
t.run()
try:
total = 0
for task in tasks:
total += task.result
except: # the task was killed
print(tasks)
return
return total / self.n_cpu
if __name__ == "__main__":
pc = plac.call(PiCalculator)
pc.submit_tasks()
try:
import time
t0 = time.time()
print("%f in %f seconds " % (pc.run(), time.time() - t0))
finally:
pc.close()
def cli_main():
#The argument parser is inferred - it also deals with too few or too many func args
plac.call(main)
weights = weightsn / len(refs)
distance_matrix = np.zeros((len(wvoc), len(wvoc)), dtype=np.double)
for i, o in enumerate(vc.get_feature_names()):
for j, c in enumerate(vc.get_feature_names()):
distance_matrix[i, j] = np.sqrt(
np.sum(((weights[i] * wvvecs[vocabdict[o]]) -
(weights[j] * wvvecs[vocabdict[c]]))**2))
if np.sum(distance_matrix) == 0.0:
return float('inf')
v_obj = v_obj.astype(np.double)
v_desc = v_desc.astype(np.double)
v_obj /= v_obj.sum()
v_desc /= v_desc.sum()
distance_matrix = distance_matrix.astype(np.double)
# distance_matrix /= distance_matrix.max()
score = emd(v_obj, v_desc, distance_matrix)
return score
if __name__ == '__main__':
import plac
plac.call(objdescrefs)
def main():
plac.call(cli)
# Minimum number of words per sentence
if len(tokens) < MIN_TOKEN_COUNT:
return False
# Most tokens should be words
if sum([t.is_alpha for t in tokens]) / len(tokens) < MIN_WORD_TOKENS_RATIO:
return False
text = spacy_sentence.text
# Most characters should be letters, not numbers and not special characters
if sum([c.isalpha() for c in text]) / len(text) < MIN_LETTER_CHAR_RATIO:
return False
return True
def process_paper_record(paper_record, out_file, with_body):
sentences = _paper_record_to_sentences(paper_record, with_body)
# print(" paper {} with {} sents".format(paper_record['id'], len(sentences)))
for s in sentences:
s_text = re.sub('\s+', ' ', s.text).strip()
if s_text != "":
out_file.write("{}\n".format(s_text))
if len(sentences) > 0:
out_file.write("\n")
plac.call(main, sys.argv[1:])
logger.debug("pip package: {}".format(pip_package))
cmd = [
sys.executable,
"-m",
"pip",
"install",
"--no-deps",
"--no-cache-dir",
pip_package,
]
logger.debug(" ".join(cmd))
if subprocess.call(cmd, env=os.environ.copy()) == 0:
logger.debug("linking: {} to {}".format(model_name, lang))
package_path = get_package_path(model_name)
link(model_name, lang, force=True, model_path=package_path)
else:
raise Exception("Error to download {}".format(lang))
elif lang != value:
logger.debug("downloading {}".format(value))
download(value)
logger.debug("linking: {} to {}".format(value, lang))
package_path = get_package_path(value)
link(value, lang, force=True, model_path=package_path)
else:
logger.debug("downloading {}".format(value))
download(value)
if __name__ == "__main__":
plac.call(download_spacy_models, sys.argv[1:])
def main():
plac.call(run)
# example10.py
import plac
@plac.annotations(
operator=("The name of an operator", 'positional', None, str, ['add', 'mul']),
numbers=("A number", 'positional', None, float, None, "n"))
def main(operator, *numbers):
"A script to add and multiply numbers"
if operator == 'mul':
op = float.__mul__
result = 1.0
else: # operator == 'add'
op = float.__add__
result = 0.0
for n in numbers:
result = op(result, n)
return result
if __name__ == '__main__':
print(plac.call(main))
('http://localhost:59125/process?INPUT_TEXT={}' + '&INPUT_TYPE=TEXT' +
'&OUTPUT_TYPE=PHONEMES&LOCALE=en_US').format(tx))
soup = BeautifulSoup(r.text, 'lxml')
for sent in soup('s'):
for token in sent('t'):
try:
yield dict(text=token.text.strip(),
pos=token['pos'],
phonetics=token['ph'])
except:
pass
def text8_to_phon():
show_every = 5000
dataset = api.load('text8')
total = 0
for chunk in dataset:
for i in range(0, len(chunk), 100):
text = " ".join(chunk[i:i + 100])
if text.strip():
for token in to_phon(text):
print(token['text'], token['pos'], token['phonetics'])
total += 100
if total % show_every == 0:
print(total, text, file=sys.stderr)
if __name__ == "__main__":
plac.call(text8_to_phon)
def cli_main():
plac.call(main)
def main():
plac.call(dispatch)
for file in [
i for s in
[glob(x) for x in ["*.py", "dataloader/*.py", "model/*.py"]] for i in s
]:
shutil.copy(file, wandb.run.dir)
trainer = pl.Trainer(gpus=hparams.gpus,
logger=[wandb_logger]) # , tb_logger],
# ------------------------
# 3 START TESTING
# ------------------------
trainer.test(model)
if __name__ == "__main__":
"""
Script entrypoint.
"""
# Converting dictionary to namespace
hyperparams = Namespace(**plac.call(get_hparams, eager=False))
# Set the evaluation flag in hyperparamters
hyperparams.eval = True
# ---------------------
# RUN TESTING
# ---------------------
main(hyperparams)
frontend=("the frontend to be used to interact with the generation program", 'positional', None, str, frontends),
cudaminer=(__cudaminer_docstring__, 'positional', None, str),
output_scan_interval=(__output_scan_interval_docstring__, 'positional', None, int, ),
output_scan_max_count=(__output_scan_max_count_docstring__, 'positional', None, int),
hash_rate_count=(__hash_rate_count_docstring__, 'positional', None, int, ),
debug=("Turn on debugging output", "flag"),
version=("Print information about the version of the software to stdout and exit", "flag"),
)
def cudaminer_param_checker(frontend=frontend_default, cudaminer=cudaminer_default, output_scan_interval=output_scan_interval_default, output_scan_max_count=output_scan_max_count_default, hash_rate_count=hash_rate_count_default, debug=debug_default, version=False):
"""
@args cudaminer %(__cudaminer_docstring__)s
@args output_scan_interval %(__output_scan_interval_docstring__)s
@args output_scan_max_count %(__output_scan_max_count_docstring__)s
@args hash_rate_count %(__hash_rate_count_docstring__)s""" % {"__cudaminer_docstring__": __cudaminer_docstring__, "__output_scan_interval_docstring__": __output_scan_interval_docstring__, "__output_scan_interval_docstring__": __output_scan_interval_docstring__, "__output_scan_max_count_docstring__": __output_scan_max_count_docstring__, "__hash_rate_count_docstring__": __hash_rate_count_docstring__}
if version:
print(cudaminer_param_checker_globals.app_version_string)
return
if debug == True:
logger.setLevel(logging.DEBUG)
ch.setLevel(logging.DEBUG)
if frontend == FRONTEND_CONSOLE_GUI:
visualize_cudaminer_param_checker_results_console_gui()
elif frontend == FRONTEND_WXPYTHON_GUI:
visualize_cudaminer_param_checker_results_wxpython_gui()
else:
raise ValueError("frontend has to be one of '%s', but is '%s'" % (str(frontends), frontend))
if __name__ == "__main__":
plac.call(cudaminer_param_checker)
for rownum in range(data_start_row, s.nrows):
bind_params = s.row_values(rownum)
log.debug('DML %r, %r', tmp_sql, bind_params)
db_cursor.execute(tmp_sql, bind_params)
db_conn.commit()
#Only do this if we're not working on an externally-opened db
if isinstance(outfile, string_types):
db_cursor.close()
db_conn.close()
def db2xls(infile, outfile):
"""
Convert an sqlite db into an xls file. Not implemented!
Some issues: one needs to be able to figure out what the table names are!
"""
raise NotImplementedError
if __name__ == "__main__":
if sys.version_info >= (3, ):
argv = sys.argv
else:
fse = sys.getfilesystemencoding()
argv = [i.decode(fse) for i in sys.argv]
#Apparently this thing's pretty magical.
import plac
plac.call(xls2db, argv[1:])
def script(main):
"""Call a script main function."""
enable_default_logging()
plac.call(main)
def _to_jsonld(microdata):
context = 'http://schema.org'
properties = 'properties_'
typestr = 'type'
jsonld_data = {}
jsonld_data["@context"] = context
for data in microdata:
data = flatten_json.flatten(data)
for key in data.keys():
value = data[key]
if context in value:
value = value.replace(context + "/", "")
if (properties in key):
keyn = key.replace(properties, "")
jsonld_data[keyn] = value
if (typestr in keyn):
keyn = keyn.replace(typestr, "@" + typestr)
jsonld_data[keyn] = value
if (typestr is key):
keyn = key.replace(typestr, "@" + typestr)
jsonld_data[keyn] = value
del data
jsonld_data = flatten_json.unflatten(jsonld_data)
return [jsonld_data]
if __name__ == "__main__":
res = plac.call(retrieve_claimreview)
print(res)
# example9.py
def main(verbose: ('prints more info', 'flag', 'v'), dsn: 'connection string'):
if verbose:
print('connecting to %s' % dsn)
# ...
if __name__ == '__main__':
import plac; plac.call(main)
year = int(y)
month = int(m)
start = datetime(year, month, 1)
end = datetime(year, month+1, 1)
current = start
n = 0
while current <= end:
files = calipso.l1_files(current.year, current.month, current.day, '*')
if files:
n+=len(files)
for f in files:
try:
linkedfile = target + '/' + os.path.basename(f)
os.symlink(f, linkedfile)
except OSError:
# catch already-linked-error silently
pass
current += timedelta(days=1)
print(n, 'files linked in ', target)
return
if __name__ == '__main__':
import plac
plac.call(rake_caliop_l1)
def main(dry_run: ('Dry run', 'flag', 'd')):
if dry_run:
print('Doing nothing')
else:
print('Doing something')
if __name__ == '__main__':
import plac; plac.call(main)
def plac_call_main():
return plac.call(main)
def main():
plac.call(generate_hashes)
def plac_call_main():
try:
return plac.call(main)
except KeyboardInterrupt:
logging.error("Canceled.")
sys.exit(1)
plt.plot(group['window-size'], group['ar-train-error'], label = ' AR train', color = 'b')
plt.fill_between(group['window-size'].values,
group['np-test-error'].values - group['np-test-std'].values,
group['np-test-error'].values + group['np-test-std'].values,
alpha=0.1, linewidth=0, color = 'r')
plt.plot(group['window-size'], group['np-test-error'], label = ' Eulerian test', color = 'r')
plt.legend(loc = 'center right')
plt.suptitle('%i Hour Forecast Performance v. Window Size' % delta_time, fontsize = 'large')
plt.xlabel('window size', fontsize = 'large')
plt.ylabel('RMSE $W/m^2$', fontsize = 'large')
plt.subplot(212)
for indx, group in gb:
plt.plot(group['window-size'], 100*(group['np-test-error'].values - group['ar-test-error'].values) / group['np-test-error'].values , label = ' AR test', color = 'g')
plt.plot(group['window-size'], 100*(group['np-train-error'].values - group['ar-train-error'].values) / group['np-test-error'].values, label = ' AR train', color = 'b')
plt.xlabel('window size', fontsize = 'large')
plt.ylabel('% better than Eulerian', fontsize = 'large')
plt.legend(loc = 'lower right', fontsize = 'large')
plt.savefig(use_path[:-4] + '.png')
if __name__ == '__main__':
#plac.call(plot_performance)
plac.call(plot_weights)
import jsonlines
from pymongo import MongoClient
import plac
import sys
@plac.annotations(coll_name=("Collectionto load into", "option", "c", str),
db_name=("Database to load into", "option", "d", str))
def delete(coll_name, db_name="gsr"):
# quick script for updating Mongo tasks
client = MongoClient('mongodb://localhost:27017/')
db = client[db_name]
coll = db[coll_name]
count = coll.count()
conf = input(
"You're about to delete the collection {0}, which has {1} records. Please type this name to confirm: "
.format(coll_name, count))
if conf != coll_name:
print("Bye!")
sys.exit(0)
if conf == coll_name:
coll.delete_many({})
print("Deleted all records from ", coll_name)
if __name__ == "__main__":
plac.call(delete)
def plac_call_main():
return plac.call(main)
def main(self, command, debug=False, local=local, *args):
if debug:
logging.basicConfig(level=logging.DEBUG)
self.local = local
self.load()
self.proxies = {}
if command == self.COMMAND_CONFIGURE:
plac.call(self.configure, args)
elif command == self.COMMAND_LOGIN:
plac.call(self.login, args)
elif command == self.COMMAND_CREATE_STACK:
plac.call(self.create_stack, args)
elif command == self.COMMAND_UPDATE_STACK:
plac.call(self.update_stack, args)
elif command == self.COMMAND_UPDATE_STACK_ACL:
plac.call(self.update_stack_acl, args)
elif command == self.COMMAND_CREATE_OR_UPDATE_STACK:
plac.call(self.create_or_update_stack, args)
elif command == self.COMMAND_GET_STACK_ID:
plac.call(self.get_stack_id, args)
elif command == self.COMMAND_UPDATE_REGISTRY:
plac.call(self.update_registry, args)
elif command == self.COMMAND_REQUEST:
plac.call(self.request, args)
def main():
try:
plac.call(run)
except KeyboardInterrupt:
pass
def test():
assert plac.call(ishelve.main, ['.clear']) == ['cleared the shelve']
assert plac.call(ishelve.main, ['a=1']) == ['setting a=1']
assert plac.call(ishelve.main, ['a']) == ['1']
assert plac.call(ishelve.main, ['.delete=a']) == ['deleted a']
assert plac.call(ishelve.main, ['a']) == ['a: not found']
def main():
plac.call(get_val)
# Updated: 2018/11/17 09:29:56 by gmonnier ### ########.fr #
# #
# **************************************************************************** #
import sys
import numpy as np
import plac
import data
@plac.annotations(
dataset=plac.Annotation("Dataset", 'positional'),
alpha=plac.Annotation('Set the learning rate', 'option', 'l', float),
n_epoch=plac.Annotation('Set the number of epochs', 'option', 'e', int),
)
def learning(dataset, alpha, n_epoch):
dat = data.Data(dataset, ',', alpha, n_epoch)
dat.display()
dat.feature_scaling()
print("Starting to learn parameters...")
dat.gradient_descent()
dat.write_thetas()
print("Parameters found and stored in file '.parameters'")
dat.draw_line()
dat.display_gradient_descent()
if __name__ == "__main__":
plac.call(learning)
def main():
plac.call(conv)
if __name__ == '__main__':
import plac
import sys
from spacy.cli import download, link, info, package, train, convert
from spacy.cli import vocab, profile, evaluate, validate
from spacy.util import prints
commands = {
'download': download,
'link': link,
'info': info,
'train': train,
'evaluate': evaluate,
'convert': convert,
'package': package,
'vocab': vocab,
'profile': profile,
'validate': validate
}
if len(sys.argv) == 1:
prints(', '.join(commands), title="Available commands", exits=1)
command = sys.argv.pop(1)
sys.argv[0] = 'spacy %s' % command
if command in commands:
plac.call(commands[command])
else:
prints("Available: %s" % ', '.join(commands),
title="Unknown command: %s" % command,
exits=1)
def entry_point():
exit_code = plac.call(main)
return exit_code
project_base = Path('/home/jeremy/Projects')
r = Redis()
def create_project(name):
project_name = name.lower().replace(' '.'-')
project_key = f'project:{project_name}'
document_path = cur_dir.joinpath('content', filename)
redis_key=f'{project}:content:{snake_case(document_path.stem)}'
args = ['--standalone']
args.append(f'-o {str(document_path)}')
with TemporaryDirectory() as tmpdir:
tmpfile = Path(tmpdir, filename)
tmpfile.write_text(input())
args.append(str(tmpfile))
print(args)
try:
rs = run('pandoc', args, universal_newlines=True)
except Exception as e:
print (e)
return False
print (f'created document {filename}')
print( redis_key )
if __name__ == '__main__':
plac.call(create_document)
def main():
sys.exit(plac.call(_main))
import plac, asyncio
import ffdl
def main(url_or_id: ("Fanfiction.net URL or just the ID")):
story_id = ffdl.get_id(url_or_id)
asyncio.run(ffdl.create_epub(story_id))
if __name__ == "__main__":
plac.call(main, version=ffdl.__version__)
def do_deleted(self, line):
self.addfilter('deleted', useflag('alloc',0))
def do_unalloc(self, line):
self.addfilter('unalloc', useflag('unalloc',1))
def do_reset(self, line):
self.filters = []
self.setprompt()
self.tree.clearfilters()
def do_EOF(self, line):
sys.exit(0)
def emptyline(self):
pass
def main(ddpath, dbpath, mountpoint):
mytskfuse = tskfuse.TskFuse(ddpath, dbpath)
filters = Filters(mytskfuse.tree)
t = multiprocessing.dummy.Process(target=filters.cmdloop)
t.daemon = True
t.start()
fuse.FUSE(mytskfuse, mountpoint, foreground=True, allow_other=True)
os.system('stty sane')
if __name__ == '__main__':
plac.call(main,sys.argv[1:])
#@+leo-ver=5-thin
#@+node:maphew.20120622231339.1567: * @file a1.py
''' First attempt at using Plac module '''
import plac
@plac.annotations(command=("command to run", 'positional', None, str,
['install', 'remove', 'update', 'setup']),
packages=("list of packages", 'positional', None, str, None,
'pkg'))
def main(command, *packages):
"Operate on packages"
yield "Running %s on %s" % (command, packages)
if command == 'install':
yield 'Installing %s' % packages
elif command == 'remove':
yield 'Removing %s' % packages
elif command == 'update':
yield 'Updating package list from mirror'
elif command == 'setup':
yield 'Initializing new Osgeo4W install'
if __name__ == '__main__':
for output in plac.call(main):
print output
#@-leo
if stats['failed'] == 0:
print colored('All %d tests passed' % stats['total'], 'white',
'on_green')
else:
print colored('%d test(s) out of %d failed' % (stats['failed'],
stats['total']), 'white', 'on_red')
def accept_tests(files):
for f in files:
corrected_content = ''
content = open(f).read()
sections = content.split('++++')
i = 0
for section in sections:
if i>0:
corrected_content += "\n++++\n"
code, _unused = map(strip, section.split('----'))
tokens_actual = prettyprint(code).strip()
corrected_content += "%s\n----\n%s" % (code, tokens_actual)
i += 1
open(f, 'w').write(corrected_content)
print colored("Accepted %s" % f, 'green')
matched_files = get_matched_files(files)
run_tests(matched_files) if not accept else accept_tests(matched_files)
if __name__ == '__main__':
import plac
plac.call(run)
with msg.loading(f"Loading model '{model}'..."):
nlp = spacy.load(model)
texts = (eg["text"] for eg in srsly.read_jsonl(data))
n_docs = 0
n_words = 0
start_time = timer()
for doc in nlp.pipe(texts):
n_docs += 1
n_words += len(doc)
end_time = timer()
wps = int(n_words / (end_time - start_time))
result = [
("Docs", f"{n_docs:,}"),
("Words", f"{n_words:,}"),
("Words/s", f"{wps:,}"),
]
msg.table(result, widths=(7, 12), aligns=("l", "r"))
if __name__ == "__main__":
opts = {"train": train_model, "evaluate": evaluate_model, "wps": wps}
cmd = sys.argv.pop(1)
if cmd not in opts:
msg.fail(f"Unknown command: {cmd}",
f"Available: {', '.join(opts)}",
exits=1)
try:
plac.call(opts[cmd])
except KeyboardInterrupt:
msg.warn("Stopped.", exits=1)
from __future__ import with_statement
from Tkinter import *
from importer3 import FakeImporter
def taskwidget(root, task, tick=500):
"A Label widget showing the output of a task every 500 ms"
sv = StringVar(root)
lb = Label(root, textvariable=sv)
def show_outlist():
try:
out = task.outlist[-1]
except IndexError: # no output yet
out = ''
sv.set('%s %s' % (task, out))
root.after(tick, show_outlist)
root.after(0, show_outlist)
return lb
def monitor(tasks):
root = Tk()
for task in tasks:
task.run()
taskwidget(root, task).pack()
root.mainloop()
if __name__ == '__main__':
import plac
with plac.Interpreter(plac.call(FakeImporter)) as i:
tasks = [i.submit('import_file f1'), i.submit('import_file f2')]
monitor(tasks)
def overlap_tokens(doc, other_doc):
"""Get the tokens from the original Doc that are also in the comparison Doc.
"""
overlap = []
other_tokens = [token.text for token in other_doc]
for token in doc:
if token.text in other_tokens:
overlap.append(token)
return overlap
if __name__ == '__main__':
plac.call(main)
@plac.annotations(
model=("Model name. Defaults to blank 'en' model.", "option", "m", str),
output_dir=("Optional output directory", "option", "o", Path),
n_iter=("Number of training iterations", "option", "n", int))
def main(model=None, output_dir=None, n_iter=100):
"""Load the model, set up the pipeline and train the entity recognizer."""
if model is not None:
nlp = spacy.load(model) # load existing spaCy model
print("Loaded model '%s'" % model)
else:
nlp = spacy.blank('en') # create blank Language class
print("Created blank 'en' model")
# create the built-in pipeline components and add them to the pipeline
# nlp.create_pipe works for built-ins that are registered with spaCy
if 'ner' not in nlp.pipe_names:
ner = nlp.create_pipe('ner')
def entry_point():
plac.call(main)
test_tags_fpath = os.path.join(cross_val_folder, 'user_test_tags.dat')
user_test_tags = load_dict_from_file(test_tags_fpath)
return user_items_to_filter, user_validation_tags, user_test_tags
@plac.annotations(db_fpath=plac.Annotation('H5 database file', type=str),
db_name=plac.Annotation('H5 database name', type=str),
cross_val_folder=plac.Annotation(
'Folder with cross validation files', type=str),
probs_folder=plac.Annotation('Probabilities Folder',
type=str))
def main(db_fpath, db_name, cross_val_folder, probs_folder):
#get cross validation dicts
user_items_to_filter, user_validation_tags, user_test_tags = \
load_train_test_validation(cross_val_folder)
with AnnotReader(db_fpath) as reader:
reader.change_table(db_name)
annot_filter = FilteredUserItemAnnotations(user_items_to_filter)
est = PrecomputedEstimator(probs_folder)
run_exp(user_validation_tags, user_test_tags, est, annot_filter,
reader)
if __name__ == '__main__':
sys.exit(plac.call(main))
val_tags_fpath = os.path.join(cross_val_folder, 'user_val_tags.dat')
user_validation_tags = load_dict_from_file(val_tags_fpath)
test_tags_fpath = os.path.join(cross_val_folder, 'user_test_tags.dat')
user_test_tags = load_dict_from_file(test_tags_fpath)
return user_items_to_filter, user_validation_tags, user_test_tags
@plac.annotations(
db_fpath = plac.Annotation('H5 database file', type=str),
db_name = plac.Annotation('H5 database name', type=str),
cross_val_folder = plac.Annotation('Folder with cross validation files',
type=str),
probs_folder = plac.Annotation('Probabilities Folder', type=str))
def main(db_fpath, db_name, cross_val_folder, probs_folder):
#get cross validation dicts
user_items_to_filter, user_validation_tags, user_test_tags = \
load_train_test_validation(cross_val_folder)
with AnnotReader(db_fpath) as reader:
reader.change_table(db_name)
annot_filter = FilteredUserItemAnnotations(user_items_to_filter)
est = PrecomputedEstimator(probs_folder)
run_exp(user_validation_tags, user_test_tags, est, annot_filter, reader)
if __name__ == '__main__':
sys.exit(plac.call(main))
rvals = defaultdict(lambda: defaultdict(int))
plvals = defaultdict(lambda: defaultdict(int))
prvals = defaultdict(lambda: defaultdict(int))
roots = defaultdict(int)
seen_pos = set(['ROOT', 'NONE'])
for sent in sents:
rdeps = defaultdict(list)
for i, (w, p, h, l) in enumerate(sent):
seen_pos.add(p)
if i > int(h):
rdeps[int(h)].append(i)
for head, children in rdeps.items():
if head == -1:
head_pos = 'ROOT'
else:
head_pos = sent[head][1]
sib_pos = 'NONE'
children.sort()
for i, child in enumerate(children):
rvals[head_pos][(sib_pos, sent[child][1])] += 1
sib_pos = sent[child][1]
seen_pos = list(sorted(seen_pos))
for head in seen_pos:
for sib in seen_pos:
for child in seen_pos:
print head, sib, child, rvals[head][(sib, child)]
if __name__ == '__main__':
plac.call(main)
import plac
import matplotlib.pyplot as plt
from os import path
@plac.annotations(
in_loc=('Location of input file'),
out_loc=('Location to save output file')
)
def main(in_loc, out_loc):
# Empty lists for values of k and corresponding WSSSEs
ks = []
WSSSEs = []
with open(in_loc, 'r') as f:
# Iterate through lines in file, adding k and WSSSE values to lists
for line in f:
line = line.split()
ks.append(int(line[0]))
WSSSEs.append(float(line[1]))
# Plot and save figure
plt.plot(ks, WSSSEs)
plt.savefig(path.join(out_loc, 'WSSSEs.png'),
bbox_inches='tight')
if __name__ == '__main__':
plac.call(main)
def test_yield():
def main():
for i in (1, 2, 3):
yield i
assert plac.call(main, []) == [1, 2, 3]
