Merge pull request #2223 from IrynaRepinetska-dev/feature/UD_Belarusian

Add support for UD_BELARUSIAN-HSE dataset (POS)
flairNLP · Apr 19, 2021 · 77e6ad6 · 77e6ad6
2 parents 217e3e6 + 9dac167
commit 77e6ad6
Show file tree

Hide file tree

Showing 2 changed files with 29 additions and 2 deletions.
diff --git a/flair/datasets/__init__.py b/flair/datasets/__init__.py
@@ -135,6 +135,7 @@
 from .treebanks import UD_OLD_FRENCH
 from .treebanks import UD_GOTHIC
 from .treebanks import UD_WOLOF
+from .treebanks import UD_BELARUSIAN
 from .treebanks import UD_OLD_CHURCH_SLAVONIC
 from .treebanks import UD_COPTIC
 

diff --git a/flair/datasets/treebanks.py b/flair/datasets/treebanks.py
@@ -1510,6 +1510,33 @@ def __init__(self, base_path: Union[str, Path] = None, in_memory: bool = True, s
         super(UD_WOLOF, self).__init__(data_folder, in_memory=in_memory, split_multiwords=split_multiwords)
 
 
+class UD_BELARUSIAN(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://github.com/UniversalDependencies/UD_Belarusian-HSE/master"
+        cached_path(f"{web_path}/be_hse-ud-dev.conllu", Path("datasets") / dataset_name)
+        cached_path(
+            f"{web_path}/be_hse-ud-test.conllu", Path("datasets") / dataset_name
+        )
+        cached_path(
+            f"{web_path}/be_hse-ud-train.conllu", Path("datasets") / dataset_name
+        )
+
+        super(UD_BELARUSIAN, self).__init__(data_folder, in_memory=in_memory, split_multiwords=split_multiwords)
+
+
 class UD_COPTIC(UniversalDependenciesCorpus):
     def __init__(self, base_path: Union[str, Path] = None, in_memory: bool = True, split_multiwords: bool = True):
 
@@ -1534,5 +1561,4 @@ def __init__(self, base_path: Union[str, Path] = None, in_memory: bool = True, s
             f"{web_path}/cop_scriptorium-ud-train.conllu", Path("datasets") / dataset_name
         )
 
-        super(UD_COPTIC, self).__init__(data_folder, in_memory=in_memory, split_multiwords=split_multiwords)
-
+        super(UD_COPTIC, self).__init__(data_folder, in_memory=in_memory, split_multiwords=split_multiwords)