schlevik
commited on
Commit
·
9192d01
1
Parent(s):
6dad2a1
change the bigbio text format to correct text
Browse files
psytar.py
CHANGED
|
@@ -41,6 +41,7 @@ This dataset can be used for:
|
|
| 41 |
In the source schema, systematic annotation with UMLS and SNOMED-CT concepts are provided.
|
| 42 |
"""
|
| 43 |
|
|
|
|
| 44 |
import re
|
| 45 |
from dataclasses import dataclass
|
| 46 |
from pathlib import Path
|
|
@@ -486,14 +487,14 @@ class PsyTARDataset(datasets.GeneratorBasedBuilder):
|
|
| 486 |
example = {
|
| 487 |
"id": idx,
|
| 488 |
"document_id": f"{row['drug_id']}_{row['sentence_index']}",
|
| 489 |
-
"text": row
|
| 490 |
-
"labels": row["
|
| 491 |
}
|
| 492 |
yield example
|
| 493 |
|
| 494 |
def _generate_examples(self, filepath) -> Tuple[int, Dict]:
|
| 495 |
"""Yields examples as (key, example) tuples."""
|
| 496 |
-
|
| 497 |
if self.config.schema == "source":
|
| 498 |
examples = self._convert_xlsx_to_source(filepath)
|
| 499 |
|
|
|
|
| 41 |
In the source schema, systematic annotation with UMLS and SNOMED-CT concepts are provided.
|
| 42 |
"""
|
| 43 |
|
| 44 |
+
import os
|
| 45 |
import re
|
| 46 |
from dataclasses import dataclass
|
| 47 |
from pathlib import Path
|
|
|
|
| 487 |
example = {
|
| 488 |
"id": idx,
|
| 489 |
"document_id": f"{row['drug_id']}_{row['sentence_index']}",
|
| 490 |
+
"text": row.sentences,
|
| 491 |
+
"labels": row["label"],
|
| 492 |
}
|
| 493 |
yield example
|
| 494 |
|
| 495 |
def _generate_examples(self, filepath) -> Tuple[int, Dict]:
|
| 496 |
"""Yields examples as (key, example) tuples."""
|
| 497 |
+
filepath = os.path.join(filepath, "PsyTAR_dataset.xlsx")
|
| 498 |
if self.config.schema == "source":
|
| 499 |
examples = self._convert_xlsx_to_source(filepath)
|
| 500 |
|