Datasets:
Upload folder using huggingface_hub
Browse files- README.md +16 -8
- medqa_corpus_en.py +0 -10
- shelves.png +3 -0
README.md
CHANGED
@@ -51,8 +51,11 @@ task_ids:
|
|
51 |
|
52 |
### Dataset Summary
|
53 |
|
54 |
-
MedQA
|
55 |
-
|
|
|
|
|
|
|
56 |
|
57 |
### Languages
|
58 |
|
@@ -80,9 +83,14 @@ valuable for pre-training of medical LLMS.
|
|
80 |
#### Initial Data Collection and Normalization
|
81 |
|
82 |
Langchain's RecursiveCharacterTextSplitter is used for chunking and the most commonly-appearing non-ASCII characters
|
83 |
-
are replaced with readable equivalents. The
|
84 |
-
|
85 |
-
|
86 |
-
-
|
87 |
-
-
|
88 |
-
-
|
|
|
|
|
|
|
|
|
|
|
|
51 |
|
52 |
### Dataset Summary
|
53 |
|
54 |
+
[MedQA](https://github.com/jind11/MedQA) includes "prepared text materials from a total of 18 English medical textbooks
|
55 |
+
that have been widely used by medical students and USMLE takers" [Jin, Di, et al. 2020].
|
56 |
+
|
57 |
+
This dataset is derived from these medical textbooks (those in English), providing subsets that coincide with Medical
|
58 |
+
subspecialties for use in pre-training medical LLMs with gold standard domain text.
|
59 |
|
60 |
### Languages
|
61 |
|
|
|
83 |
#### Initial Data Collection and Normalization
|
84 |
|
85 |
Langchain's RecursiveCharacterTextSplitter is used for chunking and the most commonly-appearing non-ASCII characters
|
86 |
+
are replaced with readable equivalents. The textbooks are then broken into separate subsets, indicated below along with
|
87 |
+
the textbooks they comprise:
|
88 |
+
|
89 |
+
- _*Core Clinical Medicine*_
|
90 |
+
- Anatomy_Gray.txt, First_Aid_Step1.txt, First_Aid_Step2.txt, Immunology_Janeway.txt, InternalMed_Harrison.txt, Neurology_Adams.txt, Obstentrics_Williams.txt, Pathoma_Husain.txt, Pediatrics_Nelson.txt, and Surgery_Schwartz.txt
|
91 |
+
- _*Basic Biology*_
|
92 |
+
- Biochemistry_Lippincott.txt, Cell_Biology_Alberts.txt, Histology_Ross.txt, Pathology_Robbins.txt, and Physiology_Levy.txt
|
93 |
+
- _*Pharmacology*_
|
94 |
+
- Pharmacology_Katzung.txt
|
95 |
+
- _*Psychiatry*_
|
96 |
+
- Psichiatry_DSM-5.txt
|
medqa_corpus_en.py
CHANGED
@@ -156,15 +156,5 @@ class MedQACorpus(datasets.GeneratorBasedBuilder):
|
|
156 |
]
|
157 |
|
158 |
def _generate_examples(self):
|
159 |
-
print(f"Generating examples in {self.base_dir} under {self.config.name}")
|
160 |
for key, record in enumerate(get_med_qa_textbooks(self.base_dir, self.config.name)):
|
161 |
yield key, record
|
162 |
-
|
163 |
-
# if __name__ == '__main__':
|
164 |
-
# c = MedQACorpus()
|
165 |
-
# c.base_dir = Path('/home/chimezie/Projects/Oori/medical_llm/MedQACorpus/textbooks/en')
|
166 |
-
# c.config.name = "basic_biology"
|
167 |
-
# for k, r in c._generate_examples():
|
168 |
-
# print(k, r)
|
169 |
-
# break
|
170 |
-
#
|
|
|
156 |
]
|
157 |
|
158 |
def _generate_examples(self):
|
|
|
159 |
for key, record in enumerate(get_med_qa_textbooks(self.base_dir, self.config.name)):
|
160 |
yield key, record
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
shelves.png
ADDED
Git LFS Details
|