Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Training fails on windows 10 , python 3.6 #604

Open
geekDecker opened this issue Jan 8, 2021 · 1 comment
Open

Training fails on windows 10 , python 3.6 #604

geekDecker opened this issue Jan 8, 2021 · 1 comment

Comments

@geekDecker
Copy link

I was trying to train rasa on windows 10 , python 3.6 with rasa 2.0.8 training stops after starting DIET Classifier
Please help me resolve issue . I am stuck since two days

here are the logs .
2021-01-08 19:59:04 INFO rasa.nlu.model - Starting to train component DIETClassifier
Traceback (most recent call last):
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\runpy.py", line 193, in run_module_as_main
"main", mod_spec)
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\runpy.py", line 85, in run_code
exec(code, run_globals)
File "C:\Users\KNI9KOR\Anaconda3\envs\rasabot\Scripts\rasa.exe_main
.py", line 7, in
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\rasa_main
.py", line 116, in main
cmdline_arguments.func(cmdline_arguments)
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\rasa\cli\train.py", line 90, in train
nlu_additional_arguments=extract_nlu_additional_arguments(args),
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\rasa\train.py", line 55, in train
loop,
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\rasa\utils\common.py", line 308, in run_in_loop
result = loop.run_until_complete(f)
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\asyncio\base_events.py", line 488, in run_until_complete
return future.result()
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\rasa\train.py", line 110, in train_async
nlu_additional_arguments=nlu_additional_arguments,
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\rasa\train.py", line 207, in _train_async_internal
old_model_zip_path=old_model,
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\rasa\train.py", line 246, in _do_training
additional_arguments=nlu_additional_arguments,
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\rasa\train.py", line 547, in _train_nlu_with_validated_data
**additional_arguments,
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\rasa\nlu\train.py", line 114, in train
interpreter = trainer.train(training_data, **kwargs)
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\rasa\nlu\model.py", line 206, in train
updates = component.train(working_data, self.config, **context)
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\rasa\nlu\classifiers\diet_classifier.py", line 760, in train
self.model = self._instantiate_model_class(model_data)
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\rasa\nlu\classifiers\diet_classifier.py", line 1086, in _instantiate_model_class
config=self.component_config,
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\rasa\nlu\classifiers\diet_classifier.py", line 1104, in init
super().init("DIET", config, data_signature, label_data)
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\rasa\utils\tensorflow\models.py", line 654, in init
checkpoint_model=config[CHECKPOINT_MODEL],
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\rasa\utils\tensorflow\models.py", line 83, in init
super().init(**kwargs)
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\tensorflow\python\training\tracking\base.py", line 457, in _method_wrapper
result = method(self, *args, **kwargs)
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\tensorflow\python\keras\engine\training.py", line 308, in init
self._init_batch_counters()
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\tensorflow\python\training\tracking\base.py", line 457, in _method_wrapper
result = method(self, *args, **kwargs)
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\tensorflow\python\keras\engine\training.py", line 317, in _init_batch_counters
self._train_counter = variables.Variable(0, dtype='int64', aggregation=agg)
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\tensorflow\python\ops\variables.py", line 262, in call
return cls._variable_v2_call(*args, **kwargs)
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\tensorflow\python\ops\variables.py", line 256, in _variable_v2_call
shape=shape)
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\tensorflow\python\ops\variables.py", line 237, in
previous_getter = lambda **kws: default_variable_creator_v2(None, **kws)
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\tensorflow\python\ops\variable_scope.py", line 2646, in default_variable_creator_v2
shape=shape)
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\tensorflow\python\ops\variables.py", line 264, in call
return super(VariableMetaclass, cls).call(*args, **kwargs)
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\tensorflow\python\ops\resource_variable_ops.py", line 1518, in init
distribute_strategy=distribute_strategy)
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\tensorflow\python\ops\resource_variable_ops.py", line 1666, in _init_from_args
graph_mode=self._in_graph_mode)
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\tensorflow\python\ops\resource_variable_ops.py", line 243, in eager_safe_variable_handle
shape, dtype, shared_name, name, graph_mode, initial_value)
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\tensorflow\python\ops\resource_variable_ops.py", line 175, in _variable_handle_from_shape_and_dtype
math_ops.logical_not(exists), [exists], name="EagerVariableNameReuse")
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\tensorflow\python\ops\gen_logging_ops.py", line 49, in _assert
_ops.raise_from_not_ok_status(e, name)
File "c:\users\kni9kor\anaconda3\envs\rasabot\lib\site-packages\tensorflow\python\framework\ops.py", line 6843, in raise_from_not_ok_status
six.raise_from(core._status_to_exception(e.code, message), None)
File "", line 3, in raise_from
tensorflow.python.framework.errors_impl.InvalidArgumentError: assertion failed: [0] [Op:Assert] name: EagerVariableNameReuse

@sara-tagger
Copy link

Thanks for raising this issue, @b-quachtran will get back to you about it soon✨

Please also check out the docs and the forum in case your issue was raised there too 🤗

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants