1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
|
# -*- coding: utf-8 -*-
"""ConfusionMatrix handlers."""
from __future__ import division
from typing import Dict, IO
from .class_funcs import class_statistics
from .errors import pycmVectorError, pycmMatrixError
from .overall_funcs import overall_statistics
from .utils import *
from .params import *
import json
import types
import numpy
def __class_stat_init__(cm: "pycm.ConfusionMatrix") -> None:
"""
Init individual class stats.
:param cm: confusion matrix
"""
for stat, field_name in CLASS_PARAMS.items():
setattr(cm, field_name, cm.class_stat[stat])
def __overall_stat_init__(cm: "pycm.ConfusionMatrix") -> None:
"""
Init individual overall stats.
:param cm: confusion matrix
"""
for stat, field_name in OVERALL_PARAMS.items():
setattr(cm, field_name, cm.overall_stat[stat])
def __imbalancement_handler__(cm: "pycm.ConfusionMatrix", is_imbalanced: bool) -> None:
"""
Check if the confusion matrix is imbalanced.
:param cm: confusion matrix
:param is_imbalanced: is imbalanced flag passed to __init__
"""
if cm.imbalance is None:
if is_imbalanced is None:
is_imbalanced = imbalance_check(cm.P)
cm.imbalance = is_imbalanced
def __obj_assign_handler__(
cm: "pycm.ConfusionMatrix",
matrix_param: Tuple[List[Any],
Dict[Any, Dict[Any, int]],
Dict[Any, int],
Dict[Any, int],
Dict[Any, int],
Dict[Any, int]]) -> None:
"""
Assign basic parameters to the input confusion matrix.
:param cm: confusion matrix
:param matrix_param: matrix parameters
"""
cm.classes = matrix_param[0]
cm.table = matrix_param[1]
cm.matrix = cm.table
cm.normalized_table = normalized_table_calc(cm.classes, cm.table)
cm.normalized_matrix = cm.normalized_table
cm.TP = matrix_param[2]
cm.TN = matrix_param[3]
cm.FP = matrix_param[4]
cm.FN = matrix_param[5]
if not cm.metrics_off:
statistic_result = class_statistics(
TP=matrix_param[2],
TN=matrix_param[3],
FP=matrix_param[4],
FN=matrix_param[5],
classes=matrix_param[0],
table=matrix_param[1])
cm.class_stat = statistic_result
cm.overall_stat = overall_statistics(
RACC=statistic_result["RACC"],
RACCU=statistic_result["RACCU"],
TPR=statistic_result["TPR"],
PPV=statistic_result["PPV"],
NPV=statistic_result["NPV"],
F1=statistic_result["F1"],
TP=statistic_result["TP"],
FN=statistic_result["FN"],
ACC=statistic_result["ACC"],
POP=statistic_result["POP"],
P=statistic_result["P"],
TOP=statistic_result["TOP"],
jaccard_list=statistic_result["J"],
classes=cm.classes,
table=cm.table,
CEN_dict=statistic_result["CEN"],
MCEN_dict=statistic_result["MCEN"],
AUC_dict=statistic_result["AUC"],
ICSI_dict=statistic_result["ICSI"],
TNR=statistic_result["TNR"],
TN=statistic_result["TN"],
FP=statistic_result["FP"])
else:
cm.class_stat = dict(zip(CLASS_PARAMS.keys(), len(
CLASS_PARAMS) * [{i: "None" for i in cm.classes}]))
cm.overall_stat = dict(
zip(OVERALL_PARAMS.keys(), len(OVERALL_PARAMS) * ["None"]))
def __obj_file_handler__(cm: "pycm.ConfusionMatrix", file: IO) -> Tuple[List[Any],
Dict[Any, Dict[Any, int]],
Dict[Any, int],
Dict[Any, int],
Dict[Any, int],
Dict[Any, int]]:
"""
Handle object conditions for the input file.
:param cm: confusion matrix
:param file: saved confusion matrix file object
"""
obj_data = json.load(file)
if obj_data["Actual-Vector"] is not None and obj_data[
"Predict-Vector"] is not None:
loaded_weights = obj_data.get("Sample-Weight", None)
matrix_param = matrix_params_calc(obj_data[
"Actual-Vector"],
obj_data[
"Predict-Vector"], loaded_weights)
cm.actual_vector = obj_data["Actual-Vector"]
cm.predict_vector = obj_data["Predict-Vector"]
cm.prob_vector = obj_data.get("Prob-Vector", None)
cm.weights = loaded_weights
else:
try:
loaded_transpose = obj_data["Transpose"]
except Exception:
loaded_transpose = False
cm.transpose = loaded_transpose
loaded_matrix = dict(obj_data["Matrix"])
for i in loaded_matrix:
loaded_matrix[i] = dict(loaded_matrix[i])
matrix_param = matrix_params_from_table(loaded_matrix)
cm.digit = obj_data["Digit"]
cm.imbalance = obj_data.get("Imbalanced", None)
return matrix_param
def __obj_matrix_handler__(
matrix: Dict[Any, Dict[Any, int]],
classes: List[Any],
transpose: bool) -> Tuple[List[Any],
Dict[Any, Dict[Any, int]],
Dict[Any, int],
Dict[Any, int],
Dict[Any, int],
Dict[Any, int]]:
"""
Handle object conditions for the matrix.
:param matrix: the confusion matrix in dict form
:param classes: ordered labels of classes
:param transpose: transpose flag
"""
if matrix_check(matrix):
if class_check(list(matrix)) is False:
raise pycmMatrixError(MATRIX_CLASS_TYPE_ERROR)
matrix_param = matrix_params_from_table(matrix, classes, transpose)
else:
raise pycmMatrixError(MATRIX_FORMAT_ERROR)
return matrix_param
def __obj_array_handler__(
array: Union[List[List[int]], numpy.ndarray],
classes: List[Any],
transpose: bool) -> Tuple[List[Any],
Dict[Any, Dict[Any, int]],
Dict[Any, int],
Dict[Any, int],
Dict[Any, int],
Dict[Any, int]]:
"""
Handle object conditions for the array.
:param matrix: the confusion matrix in array form
:param classes: ordered labels of classes
:param transpose: transpose flag
"""
if classes is not None and len(set(classes)) != len(classes):
raise pycmMatrixError(VECTOR_UNIQUE_CLASS_ERROR)
if classes is None:
classes = list(range(len(array)))
if len(classes) != len(array):
raise pycmMatrixError(CLASSES_LENGTH_ERROR)
matrix = {}
for i in range(len(array)):
matrix[classes[i]] = {classes[j]: x for j, x in enumerate(array[i])}
return __obj_matrix_handler__(matrix, classes, transpose)
def __obj_vector_handler__(
cm: "pycm.ConfusionMatrix",
actual_vector: Union[List[Any], numpy.ndarray],
predict_vector: Union[List[Any], numpy.ndarray],
threshold: Callable,
sample_weight: Union[List[Any], numpy.ndarray],
classes: List[Any]) -> Tuple[List[Any],
Dict[Any, Dict[Any, int]],
Dict[Any, int],
Dict[Any, int],
Dict[Any, int],
Dict[Any, int]]:
"""
Handle object conditions for vectors.
:param cm: confusion matrix
:param actual_vector: actual vector
:param predict_vector: vector of predictions
:param threshold: activation threshold function
:param sample_weight: sample weights list
:param classes: ordered labels of classes
"""
if isinstance(threshold, types.FunctionType):
cm.prob_vector = predict_vector
predict_vector = list(map(threshold, predict_vector))
cm.predict_vector = predict_vector
if not isinstance(actual_vector, (list, numpy.ndarray)) or not \
isinstance(predict_vector, (list, numpy.ndarray)):
raise pycmVectorError(VECTOR_TYPE_ERROR)
if len(actual_vector) != len(predict_vector):
raise pycmVectorError(VECTOR_SIZE_ERROR)
if len(actual_vector) == 0 or len(predict_vector) == 0:
raise pycmVectorError(VECTOR_EMPTY_ERROR)
if classes is not None and len(set(classes)) != len(classes):
raise pycmVectorError(VECTOR_UNIQUE_CLASS_ERROR)
matrix_param = matrix_params_calc(
actual_vector, predict_vector, sample_weight, classes)
if isinstance(sample_weight, (list, numpy.ndarray)):
cm.weights = sample_weight
return matrix_param
|
