Datasets:
Add category to all samples
Browse files- legal_lama.py +12 -0
legal_lama.py
CHANGED
@@ -122,6 +122,17 @@ class LegalLAMA(datasets.GeneratorBasedBuilder):
|
|
122 |
),
|
123 |
]
|
124 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
125 |
def _generate_examples(self, filepath, split):
|
126 |
"""This function returns the examples in the raw (text) form."""
|
127 |
with open(filepath, encoding="utf-8") as f:
|
@@ -130,5 +141,6 @@ class LegalLAMA(datasets.GeneratorBasedBuilder):
|
|
130 |
example = {
|
131 |
"text": data["masked_sentences"][0],
|
132 |
"label": data["obj_label"],
|
|
|
133 |
}
|
134 |
yield id_, example
|
|
|
122 |
),
|
123 |
]
|
124 |
|
125 |
+
def _get_category(self, sample):
|
126 |
+
if 'canadian_article' in sample:
|
127 |
+
category = sample['canadian_article']
|
128 |
+
elif 'legal_topic' in sample:
|
129 |
+
category = sample['legal_topic']
|
130 |
+
elif 'echr_article' in sample:
|
131 |
+
category = sample['echr_article']
|
132 |
+
else:
|
133 |
+
category = sample['obj_label']
|
134 |
+
return category
|
135 |
+
|
136 |
def _generate_examples(self, filepath, split):
|
137 |
"""This function returns the examples in the raw (text) form."""
|
138 |
with open(filepath, encoding="utf-8") as f:
|
|
|
141 |
example = {
|
142 |
"text": data["masked_sentences"][0],
|
143 |
"label": data["obj_label"],
|
144 |
+
"category": self._get_category(data)
|
145 |
}
|
146 |
yield id_, example
|