Merge pull request #2236 from enricoboos/master

added kazakh as new language for pos

flair/datasets/__init__.py

@@ -98,6 +98,7 @@
 from .treebanks import UniversalDependenciesDataset
 from .treebanks import UD_ARMENIAN
 from .treebanks import UD_ENGLISH
+from .treebanks import UD_KAZAKH
 from .treebanks import UD_ESTONIAN
 from .treebanks import UD_GERMAN
 from .treebanks import UD_GERMAN_HDT

flair/datasets/treebanks.py

@@ -251,6 +251,33 @@ def __init__(self, base_path: Union[str, Path] = None, in_memory: bool = True, s
 
         super(UD_ENGLISH, self).__init__(data_folder, in_memory=in_memory, split_multiwords=split_multiwords)
 
+
+class UD_KAZAKH(UniversalDependenciesCorpus):
+    def __init__(self, base_path: Union[str, Path] = None, in_memory: bool = True, split_multiwords: bool = True):
+
+        if type(base_path) == str:
+            base_path: Path = Path(base_path)
+
+        # this dataset name
+        dataset_name = self.__class__.__name__.lower()
+
+        # default dataset folder is the cache root
+        if not base_path:
+            base_path = Path(flair.cache_root) / "datasets"
+        data_folder = base_path / dataset_name
+
+        # download data if necessary
+        web_path = "https://raw.githubusercontent.com/UniversalDependencies/UD_Kazakh-KTB/master"
+        cached_path(
+            f"{web_path}/kk_ktb-ud-test.conllu", Path("datasets") / dataset_name
+        )
+        cached_path(
+            f"{web_path}/kk_ktb-ud-train.conllu", Path("datasets") / dataset_name
+        )
+
+        super(UD_KAZAKH, self).__init__(data_folder, in_memory=in_memory, split_multiwords=split_multiwords)
+
+
 class UD_OLD_CHURCH_SLAVONIC(UniversalDependenciesCorpus):
     def __init__(self, base_path: Union[str, Path] = None, in_memory: bool = True, split_multiwords: bool = True):