forked from evidencebp/commit-classification
-
Notifications
You must be signed in to change notification settings - Fork 0
/
model_evaluation.py
61 lines (49 loc) · 2 KB
/
model_evaluation.py
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
import pandas as pd
import sys
ANALYSIS_PATH = '/Users/idan/src/analysis_utils'
sys.path.append(ANALYSIS_PATH)
from confusion_matrix import ConfusionMatrix
def classifiy_commits_df(df
, classification_column
, classification_function):
df[classification_column] = df.message.map(lambda x: classification_function(x) > 0)
return df
def evaluate_performance(df
, classification_column
, concept_column):
g = df.groupby(
[classification_column, concept_column]
, as_index=False).agg({'commit' : 'count'})
cm = ConfusionMatrix(g_df=g
,classifier=classification_column
,concept=concept_column
,count='commit')
return cm.summarize()
def evaluate_regex_results(labels_file
, classification_column
, classification_function
, concept_column
):
df = pd.read_csv(labels_file
, engine='python')
df = classifiy_commits_df(df
, classification_column
, classification_function
)
df.to_csv(labels_file
, index=False)
return evaluate_performance(df
, classification_column
, concept_column)
def evaluate_regex_results_on_df(df
, classification_column
, classification_function
, concept_column
):
df = classifiy_commits_df(df
, classification_column
, classification_function
)
return evaluate_performance(df
, classification_column
, concept_column)