Temporarily removed memory optimization bc of issues with MWT dependency.

VERSION

-1.10.4
+1.10.5

texta_mlp/mlp.py

 import logging
 import os
 import pathlib
-import regex as re
 import shutil
+from typing import List, Optional
+from urllib.parse import urlparse
+from urllib.request import urlopen
+
+import regex as re
 import stanza
 from bs4 import BeautifulSoup
 from langdetect import detect
 from pelecanus import PelicanJson
-from typing import List, Optional
-from urllib.parse import urlparse
-from urllib.request import urlopen
 
 from texta_mlp.document import Document
 from texta_mlp.entity_mapper import EntityMapper
-from texta_mlp.exceptions import LanguageNotSupported
 from texta_mlp.utils import parse_bool_env
 
 
@@ -269,7 +269,6 @@ class MLP:
             stanza_resource_path = pathlib.Path(self.resource_dir) / "stanza"
             stanza_pipelines[lang] = stanza.Pipeline(
                 lang=lang,
-                processors=self._get_stanza_processors(lang),
                 dir=str(stanza_resource_path),
                 use_gpu=self.use_gpu,
                 logging_level=logging_level
@@ -277,18 +276,6 @@ class MLP:
         return stanza_pipelines
 
 
-    @staticmethod
-    def _get_stanza_processors(lang):
-        """
-        Returns processor options based on language and NER support in Stanza.
-        """
-        print(lang)
-        if lang in STANZA_NER_SUPPORT:
-            return "tokenize,pos,lemma,ner"
-        else:
-            return "tokenize,pos,lemma"
-
-
     def process(self, raw_text: str, analyzers: list = ["all"], lang=None):
         """
         Processes raw text.