Create prepare.py

prepare.py

@@ -0,0 +1,33 @@
+import re
+from datasets import load_dataset, Dataset, DatasetDict
+
+
+def main():
+    datasets = load_dataset('openlifescienceai/medmcqa')
+
+    _datasets = {}
+    for split in ['train','validation']:
+        data = []
+        for example in datasets[split]:
+            question = str(example['question'])
+            choices = [example['opa'], example['opb'], example['opc'], example['opd']]
+            answer_index = example['cop']
+            answer = chr(ord('A') + answer_index)
+            data.append({
+                'question': question.strip(), 
+                'choices': choices, 
+                'answer': answer.strip(), 
+                'answer_index': answer_index
+            })
+        dataset = Dataset.from_list(data)
+        _datasets[split] = dataset
+
+    datasets = DatasetDict({
+        'train': _datasets['train'], 
+        'validation': _datasets['validation'], 
+    })
+    datasets.push_to_hub('extraordinarylab/medmcqa')
+
+
+if __name__ == '__main__':
+    main()