# Rasa shell

**URL:** <https://forum.rasa.com/t/rasa-shell/39661>\
**Category:** Rasa Open Source\
**Created:** [January 31, 2021, 12:33pm UTC](https://forum.rasa.com/t/rasa-shell/39661 "2021-01-31T12:33:35Z")\
**Posts on this page:** 10\
**Page:** 1

<div class="post-metadata">

**Author:** ![M\_Salamin](https://avatars.discourse-cdn.com/v4/letter/m/53a042/32.png) [@M\_Salamin](https://forum.rasa.com/u/M_Salamin)\
**Post date:** [January 31, 2021, 12:33pm UTC](https://forum.rasa.com/t/rasa-shell/39661/1 "2021-01-31T12:33:35Z")

</div>

# Hi, I was following the steps with the Rasa course on Udemy, that is presented by Mady Mantha. in section 3, lecture 1, she typed _rasa shell_ and then she started to chat with the bot and it responded to her. I tried what she did and got a different results than her.

NLU model loaded. Type a message and press enter to parse it. Next message: hi { “text”: “hi”, “intent”: { “id”: -2175170181951886750, “name”: “greet”, “confidence”: 0.9976823925971985 }, “entities”: [], “intent\_ranking”: [{ “id”: -2175170181951886750, “name”: “greet”, “confidence”: 0.9976823925971985 }, { “id”: -7296413657259518803, “name”: “goodbye”, “confidence”: 0.0005970306810922921 }, { “id”: -6676306855074094835, “name”: “ask\_eat\_healthy”, “confidence”: 0.0005204474437050521 }, { “id”: 5751836863188558239, “name”: “mood\_great”, “confidence”: 0.00048421788960695267 }, { “id”: 5235019163836095929, “name”: “thankyou”, “confidence”: 0.00020467927970457822 }, { “id”: 5823801132937241105, “name”: “affirm”, “confidence”: 0.00019201762916054577 }, { “id”: 292196863936170350, “name”: “out\_of\_scope”, “confidence”: 0.00018072377133648843 }, { “id”: -4263807556109513959, “name”: “bot\_challenge”, “confidence”: 7.941111834952608e-05 }, { “id”: 6952186192965535413, “name”: “inform”, “confidence”: 3.985824150731787e-05 }, { “id”: -4021515175722973480, “name”: “ask\_lower\_stress”, “confidence”: 1.924479875015095e-05 }], “response\_selector”: { “all\_retrieval\_intents”: [], “default”: { “response”: { “id”: null, “response\_templates”: null, “confidence”: 0.0, “intent\_response\_key”: null, “template\_name”: “utter\_None” }, “ranking”: [] } } }

---

<div class="post-metadata">

**Author:** ![lindig](https://avatars.discourse-cdn.com/v4/letter/l/5daacb/32.png) [@lindig](https://forum.rasa.com/u/lindig)\
**Post date:** [February 2, 2021, 3:18pm UTC](https://forum.rasa.com/t/rasa-shell/39661/2 "2021-02-02T15:18:59Z")

</div>

This looks like the output from `rasa shell nlu`.

Which command you are trying to run?

---

<div class="post-metadata">

**Author:** ![erohmensing](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/erohmensing/32/1789_2.png) [@erohmensing](https://forum.rasa.com/u/erohmensing)\
**Post date:** [February 8, 2021, 12:16pm UTC](https://forum.rasa.com/t/rasa-shell/39661/3 "2021-02-08T12:16:42Z")

</div>

Did you get any specific warning messages when you ran `rasa train`? Seems like you accidentally trained an NLU-only model, perhaps because your domain or stories weren’t available.

---

<div class="post-metadata">

**Author:** ![M\_Salamin](https://avatars.discourse-cdn.com/v4/letter/m/53a042/32.png) [@M\_Salamin](https://forum.rasa.com/u/M_Salamin)\
**Post date:** [February 9, 2021, 8:24pm UTC](https://forum.rasa.com/t/rasa-shell/39661/4 "2021-02-09T20:24:03Z")

</div>

I’m using the _rasa shell_ command

---

<div class="post-metadata">

**Author:** ![M\_Salamin](https://avatars.discourse-cdn.com/v4/letter/m/53a042/32.png) [@M\_Salamin](https://forum.rasa.com/u/M_Salamin)\
**Post date:** [February 9, 2021, 8:25pm UTC](https://forum.rasa.com/t/rasa-shell/39661/5 "2021-02-09T20:25:28Z")

</div>

Honestly, I tried as far as possible like the course. and no errors showed up; except the GPU errors which are unharmful.

---

<div class="post-metadata">

**Author:** ![erohmensing](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/erohmensing/32/1789_2.png) [@erohmensing](https://forum.rasa.com/u/erohmensing)\
**Post date:** [February 10, 2021, 10:57am UTC](https://forum.rasa.com/t/rasa-shell/39661/6 "2021-02-10T10:57:42Z")

</div>

Can you post the full output of the train command? Maybe `rasa train --debug` will be more helpful.

---

<div class="post-metadata">

**Author:** ![M\_Salamin](https://avatars.discourse-cdn.com/v4/letter/m/53a042/32.png) [@M\_Salamin](https://forum.rasa.com/u/M_Salamin)\
**Post date:** [February 11, 2021, 9:21am UTC](https://forum.rasa.com/t/rasa-shell/39661/7 "2021-02-11T09:21:36Z")

</div>

I’ll try

> rasa train --debug

C:\Users\maher\rasa\_course\>rasa shell

fatal: bad revision ‘HEAD’ 2021-02-11 11:05:41.743796: W tensorflow/stream\_executor/platform/default/dso\_loader.cc:59] Could not load dynamic library ‘cudart64\_101.dll’; dlerror: cudart64\_101.dll not found 2021-02-11 11:05:41.744043: I tensorflow/stream\_executor/cuda/cudart\_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine. 2021-02-11 11:05:45.875032: W tensorflow/stream\_executor/platform/default/dso\_loader.cc:59] Could not load dynamic library ‘nvcuda.dll’; dlerror: nvcuda.dll not found 2021-02-11 11:05:45.875183: W tensorflow/stream\_executor/cuda/cuda\_driver.cc:312] failed call to cuInit: UNKNOWN ERROR (303) 2021-02-11 11:05:45.879180: I tensorflow/stream\_executor/cuda/cuda\_diagnostics.cc:169] retrieving CUDA diagnostic information for host: MaherSalamin 2021-02-11 11:05:45.879480: I tensorflow/stream\_executor/cuda/cuda\_diagnostics.cc:176] hostname: MaherSalamin 2021-02-11 11:05:45 INFO rasa.model - Loading model models\nlu-20210211-110125.tar.gz… 2021-02-11 11:05:49.240832: I tensorflow/core/platform/cpu\_feature\_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN)to use the following CPU instructions in performance-critical operations: AVX2 To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags. 2021-02-11 11:05:49.293483: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x1f6af3eabc0 initialized for platform Host (this does not guarantee that XLA will be used). Devices: 2021-02-11 11:05:49.293955: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version NLU model loaded. Type a message and press enter to parse it. Next message:

---

<div class="post-metadata">

**Author:** ![erohmensing](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/erohmensing/32/1789_2.png) [@erohmensing](https://forum.rasa.com/u/erohmensing)\
**Post date:** [February 11, 2021, 11:25am UTC](https://forum.rasa.com/t/rasa-shell/39661/8 "2021-02-11T11:25:08Z")

</div>

This looks like the output of the rasa shell command, not the rasa train command, can you try again?

---

<div class="post-metadata">

**Author:** ![M\_Salamin](https://avatars.discourse-cdn.com/v4/letter/m/53a042/32.png) [@M\_Salamin](https://forum.rasa.com/u/M_Salamin)\
**Post date:** [February 11, 2021, 4:00pm UTC](https://forum.rasa.com/t/rasa-shell/39661/9 "2021-02-11T16:00:20Z")

</div>

I’m sorry, my bad …

here it is… …

Training NLU model… 2021-02-11 10:59:22 INFO rasa.shared.nlu.training\_data.training\_data - Training data stats: 2021-02-11 10:59:22 INFO rasa.shared.nlu.training\_data.training\_data - Number of intent examples: 137 (13 distinct intents)

2021-02-11 10:59:22 INFO rasa.shared.nlu.training\_data.training\_data - Found intents: ‘affirm’, ‘inform’, ‘greet’, ‘out\_of\_scope’, ‘goodbye’, ‘ask\_lower\_stress’, ‘ask\_exercise’, ‘bot\_challenge’, ‘mood\_great’, ‘mood\_unhappy’, ‘thankyou’, ‘ask\_eat\_healthy’, ‘deny’ 2021-02-11 10:59:22 INFO rasa.shared.nlu.training\_data.training\_data - Number of response examples: 0 (0 distinct responses) 2021-02-11 10:59:22 INFO rasa.shared.nlu.training\_data.training\_data - Number of entity examples: 10 (3 distinct entities) 2021-02-11 10:59:22 INFO rasa.shared.nlu.training\_data.training\_data - Found entity types: ‘stress’, ‘sleep’, ‘exercise’ 2021-02-11 10:59:22 INFO rasa.nlu.model - Starting to train component WhitespaceTokenizer 2021-02-11 10:59:22 INFO rasa.nlu.model - Finished training component. 2021-02-11 10:59:22 INFO rasa.nlu.model - Starting to train component RegexFeaturizer 2021-02-11 10:59:22 INFO rasa.nlu.model - Finished training component. 2021-02-11 10:59:22 INFO rasa.nlu.model - Starting to train component LexicalSyntacticFeaturizer 2021-02-11 10:59:22 INFO rasa.nlu.model - Finished training component. 2021-02-11 10:59:22 INFO rasa.nlu.model - Starting to train component CountVectorsFeaturizer 2021-02-11 10:59:22 INFO rasa.nlu.featurizers.sparse\_featurizer.count\_vectors\_featurizer - 200 vocabulary slots consumed out of 1200 slots configured for text attribute. 2021-02-11 10:59:22 INFO rasa.nlu.model - Finished training component. 2021-02-11 10:59:22 INFO rasa.nlu.model - Starting to train component CountVectorsFeaturizer 2021-02-11 10:59:22 INFO rasa.nlu.featurizers.sparse\_featurizer.count\_vectors\_featurizer - 1481 vocabulary slots consumed out of 2481 slots configured for text attribute. 2021-02-11 10:59:23 INFO rasa.nlu.model - Finished training component. 2021-02-11 10:59:23 INFO rasa.nlu.model - Starting to train component DIETClassifier 2021-02-11 10:59:23.908075: I tensorflow/core/platform/cpu\_feature\_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN)to use the following CPU instructions in performance-critical operations: AVX2 To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags. 2021-02-11 10:59:24.623700: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x2c418571860 initialized for platform Host (this does not guarantee that XLA will be used). Devices: 2021-02-11 10:59:24.624184: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version Epochs: 100%|████████████| 100/100 [00:46\<00:00, 2.14it/s, t\_loss=1.924, i\_acc=1.000, e\_f1=1.000] 2021-02-11 11:01:15 INFO rasa.utils.tensorflow.models - Finished training. 2021-02-11 11:01:16 INFO rasa.nlu.model - Finished training component. 2021-02-11 11:01:16 INFO rasa.nlu.model - Starting to train component EntitySynonymMapper 2021-02-11 11:01:16 INFO rasa.nlu.model - Finished training component. 2021-02-11 11:01:16 INFO rasa.nlu.model - Starting to train component ResponseSelector 2021-02-11 11:01:16 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. 2021-02-11 11:01:16 INFO rasa.nlu.model - Finished training component. 2021-02-11 11:01:16 INFO rasa.nlu.model - Starting to train component FallbackClassifier 2021-02-11 11:01:16 INFO rasa.nlu.model - Finished training component. 2021-02-11 11:01:23 INFO rasa.nlu.model - Successfully saved model into ‘C:\Users\maher\AppData\Local\Temp\tmpcjdp06o9\nlu’ NLU model training completed. Your Rasa model is trained and saved at ‘C:\Users\maher\rasa\_course\models\nlu-20210211-110125.tar.gz’. Core training was skipped because no valid domain file was found. Only an NLU-model was created. Please specify a valid domain using the ‘–domain’ argument or check if the provided domain file exists.

---

<div class="post-metadata">

**Author:** ![erohmensing](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/erohmensing/32/1789_2.png) [@erohmensing](https://forum.rasa.com/u/erohmensing)\
**Post date:** [February 11, 2021, 7:37pm UTC](https://forum.rasa.com/t/rasa-shell/39661/10 "2021-02-11T19:37:11Z")

</div>

Here is why there is no core model being trained:

> 20210211-110125.tar.gz’. Core training was skipped because no valid domain file was found. Only an NLU-model was created. Please specify a valid domain using the ‘–domain’ argument or check if the provided domain file exists.

Maybe try running `rasa data validate` to check your domain file? Or does it live in a different location than just `./domain.yml`?