--- a
+++ b/benchmark/description2icd10.py
@@ -0,0 +1,72 @@
+'''
+icd code maps to description 
+input: "icdcode/ICD-10-Medical-Diagnosis-Codes.pdf"
+output:  "icdcode/description2icd10.pkl" 
+
+'''
+
+
+
+import icd10, PyPDF2, pickle, os, csv 
+
+# def code2description(code):
+# 	# code = "A1803"
+# 	code = icd10.find(code)
+# 	description = code.description 
+# 	return description 
+
+# def extract_icdcode(file, pkl_file):
+# 	f = open(file, 'rb')
+# 	description2icd = dict()
+# 	fileReader = PyPDF2.PdfFileReader(file)
+# 	for pagenum in range(fileReader.numPages):
+# 		pageObj = fileReader.getPage(pagenum)
+# 		text = pageObj.extractText()
+# 		text = text.split('\n')
+# 		text = list(filter(lambda x:len(x.strip())>0, text))
+# 		if pagenum == 0:
+# 			text = list(filter(lambda x:x[0]=='A', text))
+# 			print(text)
+# 		for i in text:
+# 			icd = i.split()[0]
+# 			description = ' '.join(i.strip().split()[1:]).lower()
+# 			description2icd[description] = icd 
+# 			# print(description, icd)
+# 	print("code number is", str(len(description2icd)))
+# 	# code number is 76331
+# 	pickle.dump(description2icd, open(pkl_file, 'wb'))
+
+
+
+###  csv_file = 'icdcode/icd_10_direct_mapping.csv'
+def extract_icdcode(csv_file, pkl_file):
+	with open(csv_file, 'r') as csvfile:
+		rows = list(csv.reader(csvfile, delimiter=','))
+	description2icd = dict() 
+	for row in rows:
+		icd = row[0]
+		description = row[1]
+		print(icd)
+		description2icd[description] = icd
+	print("code number is", str(len(description2icd)))	
+	pickle.dump(description2icd, open(pkl_file, 'wb'))
+
+
+
+
+if __name__ == "__main__":
+	# file = "icdcode/ICD-10-Medical-Diagnosis-Codes.pdf"
+	csv_file = 'icdcode/icd_10_direct_mapping.csv'
+	pkl_file = "icdcode/description2icd10.pkl"
+	if not os.path.exists(pkl_file):
+		extract_icdcode(csv_file, pkl_file)
+	description2icd10 = pickle.load(open(pkl_file, 'rb'))
+	# for description, icd in description2icd10.items():
+	# 	#if len(description.split())==1:
+	# 	print(description, "  -->  ", icd)
+
+
+
+
+
+