'''
JSON output module (for general JSON writing operations).
.. moduleauthor:: Chris Fournier <*****@*****.**>
'''
from __future__ import absolute_import
import json
import os
import codecs
from segeval.util.lang import enum
Field = enum(
# Property fields
segmentation_type='segmentation_type',
# Structural fields
items='items')
SegmentationType = enum(linear='linear')
def __write_json__(filepath, data):
'''
Write a JSON file using the given data.
'''
# Create a default filename if a dir is specified
if os.path.isdir(filepath):
filepath = os.path.join(filepath, 'output.json')
# Open file
json_file = codecs.open(filepath, 'w+', 'utf-8')
try:
json.dump(data, fp=json_file, sort_keys=True, indent=4)
"""
Segmentation encoding format converstion utilities.
.. moduleauthor:: Chris Fournier <*****@*****.**>
"""
from __future__ import absolute_import
from itertools import groupby
from segeval.util.lang import enum
BoundaryFormat = enum(position="position", mass="mass", sets="sets", nltk="nltk")
def convert_positions_to_masses(positions):
"""
Convert an ordered sequence of boundary position labels into a
sequence of segment masses, e.g., ``[1,1,1,1,1,2,2,2,3,3,3,3,3]`` becomes
``[5,3,5]``.
:param segments: Ordered sequence of which segments a unit belongs to.
:type segments: tuple
.. deprecated:: 1.0
"""
return tuple([len(list(group)) for _, group in groupby(positions)])
def convert_masses_to_positions(masses):
"""
Converts a sequence of segment masses into an ordered sequence of section
labels for each unit, e.g., ``[5,3,5]`` becomes
'''
Machine learning metric package. This package a variety of traditional machine
learning metrics that have been adapted for use in segmentation.
.. moduleauthor:: Chris Fournier <*****@*****.**>
'''
from __future__ import absolute_import, division
from decimal import Decimal
from collections import defaultdict
from segeval.util import SegmentationMetricError
from segeval.util.math import mean
from segeval.util.lang import enum
Average = enum('micro', 'macro')
def __value_micro_macro__(fnc, arguments, classification=None,
version=Average.micro):
def __compute__(fnc, classes, arguments, classification, version):
if classification is None:
if version is Average.micro:
# Micro-average
numerator, denominator = 0, 0
for classification in classes:
arguments['classification'] = classification
arguments['return_parts'] = True
class_numerator, class_denominator = fnc(**arguments)
numerator += class_numerator
denominator += class_denominator
'''
Machine learning metric package. This package a variety of traditional machine
learning metrics that have been adapted for use in segmentation.
.. moduleauthor:: Chris Fournier <*****@*****.**>
'''
from __future__ import absolute_import, division
from decimal import Decimal
from collections import defaultdict
from segeval.util import SegmentationMetricError
from segeval.util.math import mean
from segeval.util.lang import enum
Average = enum('micro', 'macro')
def __value_micro_macro__(fnc,
arguments,
classification=None,
version=Average.micro):
def __compute__(fnc, classes, arguments, classification, version):
if classification is None:
if version is Average.micro:
# Micro-average
numerator, denominator = 0, 0
for classification in classes:
arguments['classification'] = classification
arguments['return_parts'] = True
class_numerator, class_denominator = fnc(**arguments)
numerator += class_numerator
denominator += class_denominator
'''
JSON output module (for general JSON writing operations).
.. moduleauthor:: Chris Fournier <*****@*****.**>
'''
from __future__ import absolute_import
import json
import os
import codecs
from segeval.util.lang import enum
Field = enum(
# Property fields
segmentation_type='segmentation_type',
# Structural fields
items='items'
)
SegmentationType = enum(linear='linear')
def __write_json__(filepath, data):
'''
Write a JSON file using the given data.
'''
# Create a default filename if a dir is specified
if os.path.isdir(filepath):
filepath = os.path.join(filepath, 'output.json')
# Open file
json_file = codecs.open(filepath, 'w+', 'utf-8')
'''
Segmentation encoding format converstion utilities.
.. moduleauthor:: Chris Fournier <*****@*****.**>
'''
from __future__ import absolute_import
from itertools import groupby
from segeval.util.lang import enum
BoundaryFormat = enum(position='position', mass='mass', sets='sets', nltk='nltk')
def convert_positions_to_masses(positions):
'''
Convert an ordered sequence of boundary position labels into a
sequence of segment masses, e.g., ``[1,1,1,1,1,2,2,2,3,3,3,3,3]`` becomes
``[5,3,5]``.
:param segments: Ordered sequence of which segments a unit belongs to.
:type segments: tuple
.. deprecated:: 1.0
'''
return tuple([len(list(group)) for _, group in groupby(positions)])
def convert_masses_to_positions(masses):
'''
Converts a sequence of segment masses into an ordered sequence of section
labels for each unit, e.g., ``[5,3,5]`` becomes
