make gensim optional

.github/workflows/ci.yml

@@ -20,7 +20,7 @@ jobs:
       - name: Install Torch cpu
         run: pip install torch --index-url https://download.pytorch.org/whl/cpu
       - name: Install Flair dependencies
-        run: pip install -e .
+        run: pip install -e .[word-embeddings]
       - name: Install unittest dependencies
         run: pip install -r requirements-dev.txt
       - name: Show installed dependencies

flair/class_utils.py

@@ -1,5 +1,7 @@
+import importlib
 import inspect
-from typing import Iterable, Optional, Type, TypeVar
+from types import ModuleType
+from typing import Any, Iterable, List, Optional, Type, TypeVar, Union, overload
 
 T = TypeVar("T")
 
@@ -17,3 +19,27 @@ def get_state_subclass_by_name(cls: Type[T], cls_name: Optional[str]) -> Type[T]
         if sub_cls.__name__ == cls_name:
             return sub_cls
     raise ValueError(f"Could not find any class with name '{cls_name}'")
+
+
+@overload
+def lazy_import(group: str, module: str, first_symbol: None) -> ModuleType: ...
+
+
+@overload
+def lazy_import(group: str, module: str, first_symbol: str, *symbols: str) -> List[Any]: ...
+
+
+def lazy_import(
+    group: str, module: str, first_symbol: Optional[str] = None, *symbols: str
+) -> Union[List[Any], ModuleType]:
+    try:
+        imported_module = importlib.import_module(module)
+    except ImportError:
+        raise ImportError(
+            f"Could not import {module}. Please install the optional '{group}' dependency. Via 'pip install flair[{group}]'"
+        )
+    if first_symbol is None:
+        return imported_module
+    symbols = (first_symbol, *symbols)
+
+    return [getattr(imported_module, symbol) for symbol in symbols]

flair/embeddings/__init__.py

@@ -40,7 +40,6 @@
 
 # Expose token embedding classes
 from .token import (
-    BPEmbSerializable,
     BytePairEmbeddings,
     CharacterEmbeddings,
     FastTextEmbeddings,