Hi, so I’ve tried installing rasa on a GCP Instance ( n1-standard-4 (4 vCPUs, 15 GB Memory) ). So far so good. Followed the installation guide step by step and the initial setup works. But the problem is that only the initial setup is without problems.
Core model training completed.
Training NLU model...
2020-02-12 11:33:29 INFO rasa.nlu.training_data.training_data - Training data stats:
- intent examples: 43 (7 distinct intents)
- Found intents: 'bot_challenge', 'mood_great', 'mood_unhappy', 'goodbye', 'affirm', 'greet', 'deny'
- Number of response examples: 0 (0 distinct response)
- entity examples: 0 (0 distinct entities)
- found entities:
2020-02-12 11:33:29 INFO rasa.nlu.model - Starting to train component WhitespaceTokenizer
2020-02-12 11:33:29 INFO rasa.nlu.model - Finished training component.
2020-02-12 11:33:29 INFO rasa.nlu.model - Starting to train component RegexFeaturizer
2020-02-12 11:33:29 INFO rasa.nlu.model - Finished training component.
2020-02-12 11:33:29 INFO rasa.nlu.model - Starting to train component CRFEntityExtractor
2020-02-12 11:33:29 INFO rasa.nlu.model - Finished training component.
2020-02-12 11:33:29 INFO rasa.nlu.model - Starting to train component EntitySynonymMapper
2020-02-12 11:33:29 INFO rasa.nlu.model - Finished training component.
2020-02-12 11:33:29 INFO rasa.nlu.model - Starting to train component CountVectorsFeaturizer
2020-02-12 11:33:29 INFO rasa.nlu.model - Finished training component.
2020-02-12 11:33:29 INFO rasa.nlu.model - Starting to train component CountVectorsFeaturizer
2020-02-12 11:33:29 INFO rasa.nlu.model - Finished training component.
2020-02-12 11:33:29 INFO rasa.nlu.model - Starting to train component EmbeddingIntentClassifier
Epochs: 100%|█████████████████████████████████████████████| 300/300 [00:05<00:00, 54.67it/s, loss=1.017, acc=1.000]
2020-02-12 11:33:35 INFO rasa.utils.train_utils - Finished training embedding policy, train loss=1.017, train accuracy=1.000
2020-02-12 11:33:36 INFO rasa.nlu.model - Finished training component.
2020-02-12 11:33:36 INFO rasa.nlu.model - Successfully saved model into '/tmp/tmpahuuaaqz/nlu'
NLU model training completed. ######### INIT TRAINING FINISHED HERE ##########
Your Rasa model is trained and saved at '/home/rasa/MinBot/models/20200212-113310.tar.gz'.
? Do you want to speak to the trained assistant on the command line? 🤖 Yes
2020-02-12 11:33:37 INFO root - Connecting to channel 'cmdline' which was specified by the '--connector' argument. Any other channels will be ignored. To connect to all given channels, omit the '--connector' argument.
2020-02-12 11:33:37 INFO root - Starting Rasa server on http://localhost:5005
Bot loaded. Type a message and press enter (use '/stop' to exit):
Your input -> Hey ########## RASA SHELL WORKS FINE IN RASA INIT ###############
Hey! How are you?
Your input -> Not good
Here is something to cheer you up:
Image: https://i.imgur.com/nGF1K8f.jpg
Did that help you?
Your input -> Yes!
Great, carry on!
Your input -> Bye
Bye
Your input -> /stop
2020-02-12 11:34:00 INFO root - Killing Sanic server now.
(venv) root@mentor-chatbot:/home/rasa/MinBot# rasa shell ######## TRYING TO RESTART SHELL DOESNT WORK ########
2020-02-12 11:34:06 INFO root - Connecting to channel 'cmdline' which was specified by the '--connector' argument. Any other channels will be ignored. To connect to all given channels, omit the '--connector' argument.
2020-02-12 11:34:06 INFO root - Starting Rasa server on http://localhost:5005
2020-02-12 11:34:09 INFO absl - Entry Point [tensor2tensor.envs.tic_tac_toe_env:TicTacToeEnv] registered with id [T2TEnv-TicTacToeEnv-v0]
Bot loaded. Type a message and press enter (use '/stop' to exit):
Your input -> Segmentation fault (core dumped) ###### ERROR HERE ########
(venv) root@mentor-chatbot:/home/rasa/MinBot# rasa run
2020-02-12 11:40:15 INFO root - Starting Rasa server on http://localhost:5005
2020-02-12 11:40:17 INFO absl - Entry Point [tensor2tensor.envs.tic_tac_toe_env:TicTacToeEnv] registered with id [T2TEnv-TicTacToeEnv-v0] #### RASA RUN WORKS? ##########
^C(venv) root@mentor-chatbot:/home/rasa/MinBot# rasa train
Nothing changed. You can use the old model stored at '/home/rasa/MinBot/models/20200212-113310.tar.gz'.
(venv) root@mentor-chatbot:/home/rasa/MinBot# rm models/20200212-113310.tar.gz
(venv) root@mentor-chatbot:/home/rasa/MinBot# rasa train ####### BUT I CANT TRAIN AGAIN #####
Training Core model...
2020-02-12 11:42:17 INFO absl - Entry Point [tensor2tensor.envs.tic_tac_toe_env:TicTacToeEnv] registered with id [T2TEnv-TicTacToeEnv-v0]
Processed Story Blocks: 0%| | 0/5 [00:00<?, ?it/s, # trackers=1]Segmentation fault (core dumped)
(venv) root@mentor-chatbot:/home/rasa/MinBot#
After running rasa init, rasa shell and rasa train both run into Segmentation fault (core dumped). Which is weird because it did work while inside rasa init. Am i missing some kind of basic compatibility issue here? I have run everything as root and as a normal user. Did setup a venv and upgraded pip and installed rasa inside the venv. Did try placing the bot in a few different directories just in case there was some weird stuff with permissions.
Rasa Version : 1.7.1
VM OS : Ubuntu 18
Machine Type : n1-standard-4 (4 vCPUs, 15 GB Memory) Google Cloud Compute Enginge Instance