Datasets:
ruanchaves
commited on
Commit
•
7a90f01
1
Parent(s):
9fb2ad1
add mix field
Browse files- hashset_manual.py +6 -10
hashset_manual.py
CHANGED
@@ -46,8 +46,7 @@ class HashSetManual(datasets.GeneratorBasedBuilder):
|
|
46 |
),
|
47 |
"source": datasets.Value("string"),
|
48 |
"gold_position": datasets.Value("int32"),
|
49 |
-
"
|
50 |
-
"hindi": datasets.Value("bool"),
|
51 |
"other": datasets.Value("bool"),
|
52 |
"ner": datasets.Value("bool"),
|
53 |
"annotator_id": datasets.Value("int32"),
|
@@ -85,18 +84,16 @@ class HashSetManual(datasets.GeneratorBasedBuilder):
|
|
85 |
except json.decoder.JSONDecodeError:
|
86 |
record = {"choices": [field]}
|
87 |
|
88 |
-
|
89 |
-
hindi = False
|
90 |
other = False
|
91 |
ner = False
|
92 |
if mix_label in record["choices"]:
|
93 |
-
|
94 |
-
hindi = True
|
95 |
if other_label in record["choices"]:
|
96 |
other = True
|
97 |
if ner_label in record["choices"]:
|
98 |
ner = True
|
99 |
-
return
|
100 |
|
101 |
def read_entities(field):
|
102 |
try:
|
@@ -130,7 +127,7 @@ class HashSetManual(datasets.GeneratorBasedBuilder):
|
|
130 |
|
131 |
records = pd.read_csv(filepath).to_dict("records")
|
132 |
for idx, row in enumerate(records):
|
133 |
-
|
134 |
yield idx, {
|
135 |
"index": row["Unnamed: 0"],
|
136 |
"hashtag": row["Hashtag"],
|
@@ -138,8 +135,7 @@ class HashSetManual(datasets.GeneratorBasedBuilder):
|
|
138 |
"spans": read_entities(row["charner"]),
|
139 |
"source": row["Source"],
|
140 |
"gold_position": get_gold_position(row["topk"]),
|
141 |
-
"
|
142 |
-
"hindi": hindi,
|
143 |
"other": other,
|
144 |
"ner": ner,
|
145 |
"annotator_id": int(row["annotator"]),
|
|
|
46 |
),
|
47 |
"source": datasets.Value("string"),
|
48 |
"gold_position": datasets.Value("int32"),
|
49 |
+
"mix": datasets.Value("bool"),
|
|
|
50 |
"other": datasets.Value("bool"),
|
51 |
"ner": datasets.Value("bool"),
|
52 |
"annotator_id": datasets.Value("int32"),
|
|
|
84 |
except json.decoder.JSONDecodeError:
|
85 |
record = {"choices": [field]}
|
86 |
|
87 |
+
mix = False
|
|
|
88 |
other = False
|
89 |
ner = False
|
90 |
if mix_label in record["choices"]:
|
91 |
+
mix = True
|
|
|
92 |
if other_label in record["choices"]:
|
93 |
other = True
|
94 |
if ner_label in record["choices"]:
|
95 |
ner = True
|
96 |
+
return mix, other, ner
|
97 |
|
98 |
def read_entities(field):
|
99 |
try:
|
|
|
127 |
|
128 |
records = pd.read_csv(filepath).to_dict("records")
|
129 |
for idx, row in enumerate(records):
|
130 |
+
mix, other, ner = read_language_labels(row["mutlitoken"])
|
131 |
yield idx, {
|
132 |
"index": row["Unnamed: 0"],
|
133 |
"hashtag": row["Hashtag"],
|
|
|
135 |
"spans": read_entities(row["charner"]),
|
136 |
"source": row["Source"],
|
137 |
"gold_position": get_gold_position(row["topk"]),
|
138 |
+
"mix": mix,
|
|
|
139 |
"other": other,
|
140 |
"ner": ner,
|
141 |
"annotator_id": int(row["annotator"]),
|