Update T2Ranking.py
Browse files- T2Ranking.py +9 -3
T2Ranking.py
CHANGED
|
@@ -17,6 +17,7 @@ import datasets
|
|
| 17 |
import json
|
| 18 |
from typing import List
|
| 19 |
import pandas as pd
|
|
|
|
| 20 |
|
| 21 |
_LICENSE = "http://www.apache.org/licenses/LICENSE-2.0"
|
| 22 |
_HOMEPAGE='https://huggingface.co/datasets/THUIR/T2Ranking'
|
|
@@ -178,7 +179,12 @@ class T2Ranking(datasets.GeneratorBasedBuilder):
|
|
| 178 |
return split_things
|
| 179 |
|
| 180 |
def _generate_examples(self, filepath):
|
| 181 |
-
data = pd.read_csv(filepath, sep='\t', quoting=3)
|
|
|
|
| 182 |
keys = _FEATURES_DICT[self.config.name].keys()
|
| 183 |
-
|
| 184 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
import json
|
| 18 |
from typing import List
|
| 19 |
import pandas as pd
|
| 20 |
+
import csv
|
| 21 |
|
| 22 |
_LICENSE = "http://www.apache.org/licenses/LICENSE-2.0"
|
| 23 |
_HOMEPAGE='https://huggingface.co/datasets/THUIR/T2Ranking'
|
|
|
|
| 179 |
return split_things
|
| 180 |
|
| 181 |
def _generate_examples(self, filepath):
|
| 182 |
+
# data = pd.read_csv(filepath, sep='\t', quoting=3)
|
| 183 |
+
reader = csv.DictReader(filepath, delimiter='\t')
|
| 184 |
keys = _FEATURES_DICT[self.config.name].keys()
|
| 185 |
+
idx = -1
|
| 186 |
+
for data in reader:
|
| 187 |
+
idx+=1
|
| 188 |
+
yield idx, {key: data[key] for key in keys}
|
| 189 |
+
# for idx in range(data.shape[0]):
|
| 190 |
+
# yield idx, {key: data[key][idx] for key in keys}
|