Update files from the datasets library (from 1.7.0)

Release notes: https://github.com/huggingface/datasets/releases/tag/1.7.0

README.md

@@ -17,6 +17,7 @@ task_categories:
 - text-classification
 task_ids:
 - multi-label-classification
+paperswithcode_id: null
 ---
 
 # Dataset Card for ohsumed
@@ -24,7 +25,7 @@ task_ids:
 ## Table of Contents
 - [Dataset Description](#dataset-description)
   - [Dataset Summary](#dataset-summary)
-  - [Supported Tasks](#supported-tasks-and-leaderboards)
+  - [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
   - [Languages](#languages)
 - [Dataset Structure](#dataset-structure)
   - [Data Instances](#data-instances)

ohsumed.py

@@ -188,7 +188,7 @@ class Ohsumed(datasets.GeneratorBasedBuilder):
                 if line.startswith(".I"):
                     tag = ".I"
                     if data["medline_ui"] != -1:
-                        id_ = data["seq_id"]
+                        id_ = data["seq_id"] + "_" + data["medline_ui"]
                         yield id_, {
                             "seq_id": data["seq_id"],
                             "medline_ui": data["medline_ui"],