言語処理100本ノック 2015をやってみた（第１章）
スポンサーリンク
言語処理100本ノック2015の第1章をやってみたので記録。大学の時のこういう授業受けてみたかったなぁ。www.cl.ecei.tohoku.ac.jp
#一部ちゃんと前提を読んでおらず、pythonではなく、unixのコマンドでやってしまいました。
00. 文字列の逆順
【実行&結果】
$ echo stressed | rev desserts
01. 「パタトクカシーー」
【プログラム】
string=$1 for n in 1,3,5,7 do echo $string | cut -c $n done
【実行&結果】
./01.sh "パタトクカシーー" パトカー
02. 「パトカー」＋「タクシー」＝「パタトクカシーー」
【プログラム】
#!/usr/bin/env python
# coding:utf-8
import sys
import random
str1=list(u"パトカー")
str2=list(u"タクシー")
temp=""
i=0
if len(str1) > len(str2):
max_val = len(str1)
else:
max_val = len(str2)
while i < max_val:
if i < len(str1):
temp += str1[i]
if i < len(str2):
temp += str2[i]
i += 1
print temp【実行&結果】
$ sudo python 02.py パタトクカシーー
03. 円周率
【プログラム】
#!/usr/bin/env python
# coding:utf-8
import sys
import random
argvs = sys.argv
argc = len(argvs)
org_str=str(argvs[1])
ngram_char = org_str.replace(',', '')
ngram_char = ngram_char.replace('.', '')
ngram_char = ngram_char.split(" ")
temp = ""
for val in ngram_char:
temp += str(len(val))
print temp【実行&結果】
$ sudo python 03.py "Now I need a drink, alcoholic of course, after the heavy lectures involving quantum mechanics." 314159265358979
04. 元素記号
【プログラム】
str=$1
strings=`echo $str | sed -e 's/,//g' -e 's/\.//g'`
array=($strings)
last_array_num=`expr ${#array[*]} - 1`
for n in `seq 0 $last_array_num`
do
array[$n]=`echo ${array[$n]} | cut -c 1-2`
done
for n in 1 5 6 7 8 9 15 16 19
do
array_no=`expr $n - 1`
temp2=`echo ${array[$array_no]} | cut -c 1`
array[$array_no]=$temp2
done
echo ${array[@]}【実行&結果】
$ ./04.sh "Hi He Lied Because Boron Could Not Oxidize Fluorine. New Nations Might Also Sign Peace Security Clause. Arthur King Can." H He Li Be B C N O F Ne Na Mi Al Si P S Cl Ar K Ca
05. n-gram
【プログラム】
#!/usr/bin/env python
# coding:utf-8
import sys
argvs = sys.argv
argc = len(argvs)
num=int(argvs[1])
org_str=str(argvs[2])
flag=str(argvs[3])
ngram_char = org_str.replace(',', '')
ngram_char = ngram_char.replace('.', '')
if flag=='char':
ngram_char = ngram_char.replace(' ', '')
else:
ngram_char = ngram_char.split(" ")
def n_gram(ngram_char,num):
counter = 0
while counter < (len(ngram_char)-1):
if len(ngram_char[counter:(counter+num)]) == num:
print ngram_char[counter:(counter+num)]
counter = counter+1
if __name__ == "__main__":
n_gram(ngram_char,num)
【実行&結果】
単語bi-gram
$ $ sudo python 05.py 2 "I am an NLPer" word ['I', 'am'] ['am', 'an'] ['an', 'NLPer']
文字bi-gram
$ sudo python 05.py 2 "I am an NLPer" char Ia am ma an nN NL LP Pe er
06. 集合
【プログラム】
#!/usr/bin/env python
# coding:utf-8
import sys
argvs = sys.argv
argc = len(argvs)
num=int(argvs[1])
arr_data=[]
arr_data.append(str(argvs[2]))
arr_data.append(str(argvs[3]))
counter = 0
while counter < len(arr_data):
temp = arr_data[counter].replace(',', '')
temp = temp.replace('.', '')
temp = temp.replace(' ', '')
arr_data[counter]=temp
counter = counter + 1
def n_gram(ngram_char,num):
counter = 0
temp_arr=[]
while counter < (len(ngram_char)-1):
if len(ngram_char[counter:(counter+num)]) == num:
temp_arr.append(ngram_char[counter:(counter+num)])
counter = counter+1
return temp_arr
if __name__ == "__main__":
rs_data=[]
for temp in arr_data:
rs_data.append(n_gram(temp,num))
print "\nA"
print rs_data[0]
print "\nB:"
print rs_data[1]
print ""
s = set(rs_data[0])
print "\n和集合"
print s.union(rs_data[1])
print "\n差集合"
print s.difference(rs_data[1])
print "\n積集合"
print s.intersection(rs_data[1])
【実行&結果】
$ sudo python 06.py 2 "paraparaparadise" "paragraph" A ['pa', 'ar', 'ra', 'ap', 'pa', 'ar', 'ra', 'ap', 'pa', 'ar', 'ra', 'ad', 'di', 'is', 'se'] B: ['pa', 'ar', 'ra', 'ag', 'gr', 'ra', 'ap', 'ph'] 和集合 set(['ad', 'ag', 'di', 'is', 'ap', 'pa', 'ra', 'ph', 'ar', 'se', 'gr']) 差集合 set(['ad', 'di', 'is', 'se']) 積集合 set(['ap', 'pa', 'ar', 'ra'])
07. テンプレートによる文生成
【プログラム】
#!/usr/bin/env python # coding:utf-8 import sys argvs = sys.argv argc = len(argvs) x=str(argvs[1]) y=str(argvs[2]) z=str(argvs[3]) print x + "時の" + y + "は" + z print "x=" + x + ",y=" + y +",z=" + z
【実行&結果】
$ sudo python 07.py 12 気温 22.4 12時の気温は22.4 x=12,y=気温,z=22.4
08. 暗号文
【プログラム】
#!/usr/bin/env python
# coding:utf-8
import re
import sys
def cipher(arr_str):
arr_encrypt=[]
temp_str = ""
for temp in arr_str:
matchOB = re.search("[a-z]",temp)
if matchOB is None:
arr_encrypt.append(temp)
temp_str+=temp
else:
val_temp = 219-int(ord(temp))
arr_encrypt.append(chr(val_temp))
temp_str+=chr(val_temp)
return arr_encrypt
def decrypt(arr_encrypt):
arr_decrypt=[]
for temp in arr_encrypt:
matchOB = re.search("[a-z]",temp)
if matchOB is None:
arr_decrypt.append(temp)
else:
val_temp = 219 - int(ord(temp))
arr_decrypt.append(chr(val_temp))
return arr_decrypt
def arr_to_str(arr_str):
temp_str=""
for temp in arr_str:
temp_str += temp
return temp_str
if __name__ == "__main__":
argvs = sys.argv
argc = len(argvs)
arr_str = list(argvs[1])
print "encrpt : " + arr_to_str(cipher(arr_str))
print "decrypt : " + arr_to_str(decrypt(cipher(arr_str)))
【実行&結果】
$ sudo python 08.py A1abcdE1F3 encrpt : A1zyxwE1F3 decrypt : A1abcdE1F3
09. Typoglycemia
【プログラム】
#!/usr/bin/env python
# coding:utf-8
import sys
import random
argvs = sys.argv
argc = len(argvs)
org_str=str(argvs[1])
ngram_char = org_str.replace(',', '')
ngram_char = ngram_char.replace('.', '')
ngram_char = ngram_char.split(" ")
def switch_val(arr_str,i):
flag=0
while flag==0:
rand_no = random.randint(1,(len(arr_str)-2))
if len(arr_str[rand_no]) > 4:
arr_str[i],arr_str[rand_no]=arr_str[rand_no],arr_str[i]
flag=1
return arr_str
if __name__ == "__main__":
ngram_char.remove("")
i = 1
while i < (len(ngram_char) - 1 ):
if len(ngram_char[i]) > 4:
ngram_char = switch_val(ngram_char,i)
i+=1
for temp in ngram_char:
print temp
【実行&結果】
$ sudo python 09.py "I couldn't believe that I could actually understand what I was reading : the phenomenal power of the human mind ." I power actually that I understand reading phenomenal what I was believe : the could couldn't of the human mind
