handle unknown args passed to Trainer.from_argparse_args

CHANGELOG.md

@@ -32,8 +32,11 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 - Fixed user warning when apex was used together with learning rate schedulers ([#1873](https://github.com/PyTorchLightning/pytorch-lightning/pull/1873))
 
+- Fixed an issue with `Trainer.from_argparse_args` when passing in unknown Trainer args ([#1932](https://github.com/PyTorchLightning/pytorch-lightning/pull/1932))
+
 - Fix bug related to logger not being reset correctly for model after tuner algorithms ([#1933](https://github.com/PyTorchLightning/pytorch-lightning/pull/1933))
 
+
 ## [0.7.6] - 2020-05-16
 
 ### Added

pytorch_lightning/trainer/trainer.py

@@ -730,20 +730,32 @@ def parse_argparser(arg_parser: Union[ArgumentParser, Namespace]) -> Namespace:
 
     @classmethod
     def from_argparse_args(cls, args: Union[Namespace, ArgumentParser], **kwargs) -> 'Trainer':
-        """create an instance from CLI arguments
+        """
+        Create an instance from CLI arguments.
+
+        Args:
+            args: The parser or namespace to take arguments from. Only known arguments will be
+                parsed and passed to the :class:`Trainer`.
+            **kwargs: Additional keyword arguments that may override ones in the parser or namespace.
+                These must be valid Trainer arguments.
 
         Example:
             >>> parser = ArgumentParser(add_help=False)
             >>> parser = Trainer.add_argparse_args(parser)
+            >>> parser.add_argument('--my_custom_arg', default='something')  # doctest: +SKIP
             >>> args = Trainer.parse_argparser(parser.parse_args(""))
-            >>> trainer = Trainer.from_argparse_args(args)
+            >>> trainer = Trainer.from_argparse_args(args, logger=False)
         """
         if isinstance(args, ArgumentParser):
-            args = Trainer.parse_argparser(args)
+            args = cls.parse_argparser(args)
         params = vars(args)
-        params.update(**kwargs)
 
-        return cls(**params)
+        # we only want to pass in valid Trainer args, the rest may be user specific
+        valid_kwargs = inspect.signature(cls.__init__).parameters
+        trainer_kwargs = dict((name, params[name]) for name in valid_kwargs if name in params)
+        trainer_kwargs.update(**kwargs)
+
+        return cls(**trainer_kwargs)
 
     @property
     def num_gpus(self) -> int:

tests/trainer/test_trainer_cli.py

@@ -1,5 +1,6 @@
 import inspect
 import pickle
+import sys
 from argparse import ArgumentParser, Namespace
 from unittest import mock
 
@@ -110,3 +111,28 @@ def test_argparse_args_parsing(cli_args, expected):
     for k, v in expected.items():
         assert getattr(args, k) == v
     assert Trainer.from_argparse_args(args)
+
+
+@pytest.mark.skipif(
+    sys.version_info < (3, 7),
+    reason="signature inspection while mocking is not working in Python < 3.7 despite autospec"
+)
+@pytest.mark.parametrize(['cli_args', 'extra_args'], [
+    pytest.param({}, {}),
+    pytest.param({'logger': False}, {}),
+    pytest.param({'logger': False}, {'logger': True}),
+    pytest.param({'logger': False}, {'checkpoint_callback': True}),
+])
+def test_init_from_argparse_args(cli_args, extra_args):
+    unknown_args = dict(unknown_arg=0)
+
+    # unkown args in the argparser/namespace should be ignored
+    with mock.patch('pytorch_lightning.Trainer.__init__', autospec=True, return_value=None) as init:
+        trainer = Trainer.from_argparse_args(Namespace(**cli_args, **unknown_args), **extra_args)
+        expected = dict(cli_args)
+        expected.update(extra_args)  # extra args should override any cli arg
+        init.assert_called_with(trainer, **expected)
+
+    # passing in unknown manual args should throw an error
+    with pytest.raises(TypeError, match=r"__init__\(\) got an unexpected keyword argument 'unknown_arg'"):
+        Trainer.from_argparse_args(Namespace(**cli_args), **extra_args, **unknown_args)