Changeset - 4efb46769b28
[Not reviewed]
default
0 1 0
Laman - 3 years ago 2022-09-28 22:57:22

more tests in the crossvalidation
1 file changed with 9 insertions and 7 deletions:
0 comments (0 inline, 0 general)
languedoc.py
Show inline comments
 
@@ -123,13 +123,15 @@ def cross_validate(sample_sets):
 
			test_models = [partial_model] + [m for m in models if m.language != real_lang]
 

	
 
			for k in TEST_LENS:
 
				j = random.randrange(0, len(test_text)-k)
 
				t = test_text[j:j+k]
 
				predicted_lang = identify(t, test_models)
 
				print(real_lang, predicted_lang, t)
 
				if predicted_lang == real_lang:
 
					score += 1
 
				max_score += 1
 
				for i in range(10):
 
					j = random.randrange(0, len(test_text)-k)
 
					t = test_text[j:j+k]
 
					predicted_lang = identify(t, test_models)
 
					if predicted_lang == real_lang:
 
						score += 1
 
					else:
 
						print(real_lang, predicted_lang, t)
 
					max_score += 1
 

	
 
	return score / max_score, (score, max_score)
 

	
0 comments (0 inline, 0 general)