Update parameters for Urdu

Author: HugoLaurencon

ac_dc/parameters_filtering.py

@@ -736,32 +736,32 @@
     "cond_remove_words_with_incorrect_substrings": True,
     "incorrect_word_substrings": ["http", "www", ".com", "href", "//"],
     "cond_remove_long_words": True,
-    "length_word_max_cutoff": 30,
+    "length_word_max_cutoff": 20,
     "cond_check_number_words": True,
     "tokenization": False,
     "strip_characters": special_characters_default,
-    "number_words_min_cutoff": 1,
+    "number_words_min_cutoff": 25,
     "number_words_max_cutoff": 100000,
     "cond_check_character_repetition_removal": True,
     "character_repetition_length": 10,
-    "character_repetition_max_cutoff": 0.106,
+    "character_repetition_max_cutoff": 0.19,
     "cond_check_word_repetition_removal": True,
-    "word_repetition_length": 5,
-    "word_repetition_max_cutoff": 0.19,
+    "word_repetition_length": 7,
+    "word_repetition_max_cutoff": 0.5,
     "cond_check_special_characters": True,
     "special_characters": special_characters_default,
-    "special_characters_max_cutoff": 0.4,
+    "special_characters_max_cutoff": 0.25,
     "cond_words_augmentation": False,
     "words_augmentation_group_sizes": [],
     "words_augmentation_join_char": "",
     "cond_check_stopwords": True,
-    "stopwords_min_cutoff": 0,
+    "stopwords_min_cutoff": 0.01,
     "cond_check_flagged_words": True,
-    "flagged_words_max_cutoff": 0.2,
+    "flagged_words_max_cutoff": 0.025,
     "cond_check_lang_id": True,
-    "lang_id_min_cutoff": 0.75,
+    "lang_id_min_cutoff": 0.90,
     "cond_check_perplexity": True,
-    "perplexity_max_cutoff": 3000000,
+    "perplexity_max_cutoff": 1495,
 }
 
 parameters_filtering_vi = {