Cannot start RASA shell , RASA nlu shell is loaded issue with tensorflow version 2.1.0

Rasa Open Source
1

Error messages are as follow,

2020-03-04 17:57:56.088639: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library ‘libnvinfer.so.6’; dlerror: libnvinfer.so.6: cannot open shared object file: No such file or directory

2020-03-04 17:57:56.088723: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library ‘libnvinfer_plugin.so.6’; dlerror: libnvinfer_plugin.so.6: cannot open shared object file: No such file or directory

2020-03-04 17:57:56.088739: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:30] Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.

2020-03-04 17:57:57.396943: E tensorflow/stream_executor/cuda/cuda_driver.cc:351] failed call to cuInit: UNKNOWN ERROR (303) 2020-03-04 17:58:04 INFO rasa.nlu.selectors.response_selector - Retrieval intent parameter was left to its default value. This response selector will be trained on training examples combining all retrieval intents.

2

hey @wink94, this is an warning which occurs if you are using CPU. for more info read this thread:

1 Like
3

@JiteshGaikwad thanks for the reply man. But my RASA shell does not start local server. “rasa shell” command starts the RASA nlu shell. i think its because of this issue.

4

What is the error you faced while running rasa shell command

5

@JiteshGaikwad errors that i mentioned in the question. “rasa shell” command is redirected to “rasa shell nlu”

6

Running into same issue. When I launch ‘rasa shell’, i am instead redirected to ‘rasa shell nlu’. Has anyone solved this misery so far? Any chance it’s due to conflict in underlying library versions?

7

It’s because only your nlu was trained and core training was failed. You could be seeing below message in yellow color while you were running rasa train command.

“Core training was skipped because no valid domain file was found. Only an nlu-model was created.Please specify a valid domain using ‘–domain’ argument or check if the provided domain file exists.”

if your core training is successful you would get a message like “Core model training completed.” and also your story blocks would have processed.

you can retrain your model by explicitly specifying the domain file or save the domain file and again run rasa train.

Hope this resolves your issue