Update utils/ghg_classifier.py
Browse files- utils/ghg_classifier.py +10 -6
utils/ghg_classifier.py
CHANGED
@@ -10,10 +10,9 @@ from transformers import pipeline
|
|
10 |
|
11 |
# Labels dictionary ###
|
12 |
_lab_dict = {
|
13 |
-
'
|
14 |
-
'
|
15 |
-
'
|
16 |
-
'NA':'NA',
|
17 |
}
|
18 |
|
19 |
|
@@ -74,9 +73,12 @@ def ghg_classification(haystack_doc:pd.DataFrame,
|
|
74 |
"""
|
75 |
logging.info("Working on GHG Extraction")
|
76 |
haystack_doc['GHG Label'] = 'NA'
|
77 |
-
haystack_doc['GHG Score'] =
|
|
|
78 |
temp = haystack_doc[haystack_doc['Target Label'] == 'TARGET']
|
|
|
79 |
df = haystack_doc[haystack_doc['Target Label'] == 'NEGATIVE']
|
|
|
80 |
|
81 |
if not classifier_model:
|
82 |
classifier_model = st.session_state['ghg_classifier']
|
@@ -84,9 +86,11 @@ def ghg_classification(haystack_doc:pd.DataFrame,
|
|
84 |
results = classifier_model(list(temp.text))
|
85 |
labels_= [(l[0]['label'],l[0]['score']) for l in results]
|
86 |
temp['GHG Label'],temp['GHG Score'] = zip(*labels_)
|
|
|
|
|
87 |
df = pd.concat([df,temp])
|
88 |
-
df['GHG Label'] = df['GHG Label'].apply(lambda i: _lab_dict[i])
|
89 |
df = df.reset_index(drop =True)
|
|
|
90 |
df.index += 1
|
91 |
|
92 |
return df
|
|
|
10 |
|
11 |
# Labels dictionary ###
|
12 |
_lab_dict = {
|
13 |
+
'GHG':'GHG',
|
14 |
+
'NOT_GHG':'NON GHG TRANSPORT TARGET',
|
15 |
+
'NEGATIVE':'OTHERS',
|
|
|
16 |
}
|
17 |
|
18 |
|
|
|
73 |
"""
|
74 |
logging.info("Working on GHG Extraction")
|
75 |
haystack_doc['GHG Label'] = 'NA'
|
76 |
+
haystack_doc['GHG Score'] = 0.0
|
77 |
+
# applying GHG Identifier to only 'Target' paragraphs.
|
78 |
temp = haystack_doc[haystack_doc['Target Label'] == 'TARGET']
|
79 |
+
temp = temp.reset_index(drop=True)
|
80 |
df = haystack_doc[haystack_doc['Target Label'] == 'NEGATIVE']
|
81 |
+
df = df.reset_index(drop=True)
|
82 |
|
83 |
if not classifier_model:
|
84 |
classifier_model = st.session_state['ghg_classifier']
|
|
|
86 |
results = classifier_model(list(temp.text))
|
87 |
labels_= [(l[0]['label'],l[0]['score']) for l in results]
|
88 |
temp['GHG Label'],temp['GHG Score'] = zip(*labels_)
|
89 |
+
temp['GHG Label'] = temp['GHG Label'].apply(lambda x: _lab_dict[x])
|
90 |
+
# merge back Target and non-Target dataframe
|
91 |
df = pd.concat([df,temp])
|
|
|
92 |
df = df.reset_index(drop =True)
|
93 |
+
df['GHG Score'] = df['GHG Score'].round(2)
|
94 |
df.index += 1
|
95 |
|
96 |
return df
|