Changeset - 5ab4acb6f293
[Not reviewed]
default
0 1 0
Laman - 3 years ago 2022-09-28 22:15:07

implemented Sample.compare method
1 file changed with 15 insertions and 2 deletions:
0 comments (0 inline, 0 general)
languedoc.py
Show inline comments
 
import os
 
import re
 
import random
 
import itertools
 

	
 
random.seed(19181028)
 

	
 
@@ -14,7 +15,7 @@ def extract_ngram_freqs(text, k):
 
	n = len(text)
 
	d = dict()
 

	
 
	for i in range(0, n-k):
 
	for i in range(0, n-k+1):
 
		key = text[i:i+k]
 
		if key.isspace():
 
			continue
 
@@ -63,7 +64,19 @@ class Sample:
 
		return res
 

	
 
	def compare(self, other):
 
		pass
 
		"""take k most common
 
		use frequencies x order
 
		use letter, digrams, trigrams
 
		use absolute x square"""
 
		ordered_own_trigrams = sorted(self.frequencies[2].items(), key=lambda kv: -kv[1])[:400]
 
		ordered_other_trigrams = sorted(other.frequencies[2].items(), key=lambda kv: -kv[1])[:400]
 
		ranked_own_trigrams = dict(zip([key for (key, freq) in ordered_own_trigrams], itertools.count(0)))
 
		ranked_other_trigrams = dict(zip([key for (key, freq) in ordered_other_trigrams], itertools.count(0)))
 

	
 
		res = sum(abs(v-ranked_other_trigrams.get(k, 400)) for (k, v) in ranked_own_trigrams.items()) + \
 
					sum(abs(v-ranked_own_trigrams.get(k, 400)) for (k, v) in ranked_other_trigrams.items())
 
		print(">", self.language, res)
 
		return res
 

	
 
	def print_overview(self):
 
		print(f"Sample({self.language}):")
0 comments (0 inline, 0 general)