NER 实体标注转换

BIO

B-begin，I-inside，O-outside

BMES

B-begin，M-middle，E-end，S-single

BIOES

B-begin，I-inside，O-outside，E-end，S-single

​ 以下转换之前默认处理好文件了(如无空行等)。只做了简单的测试，如果还有 bug 自己改一下就好了。因为我用的改版 Yeeda，可以直接拿到 BMES 标注的 anns 文件。

BMES 2 BIOES

import os

cur_dir = os.getcwd()

file_bmes = open('%s\\1_BMES.txt' % cur_dir, 'r', encoding='utf-8')
file_bioes = open('%s\\1_BIOES.txt' % cur_dir, 'w+', encoding='utf-8')
str1 = []

for line in file_bmes.readlines():
    if line != "\n":
        line1 = line.split()
        str2 = line1[0]
        for i in range(1, len(line1)):
            line2 = list(line1[i])
            if line2[0] == "M":
                line2[0] = "I"
            str3 = ''
            for i in line2:
                str3 = str3 + i
            str2 = str2 + ' ' + str3
        str1.append(str2)
    else:
        str1.append(line)
str = '\n'.join(str1)
file_bioes.write(str)

# 这个最后会多个空行
# for j in str1:
    # file_bioes.write(j)
    # file_bioes.write("\n")
file_bioes.close()
file_bmes.close()

BMES 2 BIO

import os

cur_dir = os.getcwd()

file_bmes = open('%s\\1_BMES.txt' % cur_dir, 'r', encoding='utf-8')
file_bio = open('%s\\1_BIO.txt' % cur_dir, 'w+', encoding='utf-8')
str1 = []
for line in file_bmes.readlines():
    line1 = line.split()
    if len(line1) > 1:
        line2 = list(line1[1])
        if line2[0] == "M" or line2[0] == "E":
            line2[0] = "I"
        elif line2[0] == "S":
            line2[0] = "B"
        str2 = ''
        for i in line2:
            str2 = str2 + i
        line = line1[0] + ' ' + str2
    str1.append(line)

str = '\n'.join(str1)
file_bio.write(str)
# for j in str1:
#     file_bio.write(j)
#     file_bio.write("\n")
file_bio.close()
file_bmes.close()