{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"2022-01-16-rasa-colab.ipynb","provenance":[{"file_id":"https://github.com/recohut/nbs/blob/main/raw/RASA_Main.ipynb","timestamp":1644643228679}],"collapsed_sections":[],"authorship_tag":"ABX9TyOc/1OPJf9nTWlne0QciDt5"},"kernelspec":{"name":"python3","display_name":"Python 3"}},"cells":[{"cell_type":"markdown","source":["# Running RASA Chatbot in Colab"],"metadata":{"id":"fBP36hRuXl76"}},{"cell_type":"markdown","source":["## RASA Main App"],"metadata":{"id":"svM2GDQSXo_L"}},{"cell_type":"code","metadata":{"id":"sIgLEkDmJdNm"},"source":["!pip install rasa"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"666GQ_ygcHfV","executionInfo":{"status":"ok","timestamp":1601356535200,"user_tz":-330,"elapsed":3263,"user":{"displayName":"Sparsh Agarwal","photoUrl":"","userId":"13037694610922482904"}},"outputId":"4664a4c9-6166-4c02-f0ba-e1ac00f6c9af","colab":{"base_uri":"https://localhost:8080/","height":138}},"source":["!git clone https://github.com/RasaHQ/rasa-for-beginners.git"],"execution_count":null,"outputs":[{"output_type":"stream","text":["Cloning into 'rasa-for-beginners'...\n","remote: Enumerating objects: 117, done.\u001b[K\n","remote: Counting objects: 100% (117/117), done.\u001b[K\n","remote: Compressing objects: 100% (57/57), done.\u001b[K\n","remote: Total 117 (delta 57), reused 117 (delta 57), pack-reused 0\u001b[K\n","Receiving objects: 100% (117/117), 376.04 KiB | 690.00 KiB/s, done.\n","Resolving deltas: 100% (57/57), done.\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"2CEXelioLMDh","executionInfo":{"status":"ok","timestamp":1601356574862,"user_tz":-330,"elapsed":1201,"user":{"displayName":"Sparsh Agarwal","photoUrl":"","userId":"13037694610922482904"}},"outputId":"b72617f6-5007-4cfd-fbc0-b2a2ff830ddb","colab":{"base_uri":"https://localhost:8080/","height":34}},"source":["cd rasa-for-beginners"],"execution_count":null,"outputs":[{"output_type":"stream","text":["/content/rasa-for-beginners\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"8EzYfNENSisg","executionInfo":{"status":"ok","timestamp":1601356730327,"user_tz":-330,"elapsed":145945,"user":{"displayName":"Sparsh Agarwal","photoUrl":"","userId":"13037694610922482904"}},"outputId":"d43e45c7-f37b-424e-9188-a8be20676a49","colab":{"base_uri":"https://localhost:8080/","height":817}},"source":["!rasa train"],"execution_count":null,"outputs":[{"output_type":"stream","text":["\u001b[94mTraining Core model...\u001b[0m\n","2020-09-29 05:16:27 \u001b[1;30mINFO    \u001b[0m \u001b[34mroot\u001b[0m  - Generating grammar tables from /usr/lib/python3.6/lib2to3/Grammar.txt\n","2020-09-29 05:16:27 \u001b[1;30mINFO    \u001b[0m \u001b[34mroot\u001b[0m  - Generating grammar tables from /usr/lib/python3.6/lib2to3/PatternGrammar.txt\n","Processed Story Blocks: 100% 10/10 [00:00<00:00, 3630.17it/s, # trackers=1]\n","Processed Story Blocks: 100% 10/10 [00:00<00:00, 515.19it/s, # trackers=10]\n","Processed Story Blocks: 100% 10/10 [00:00<00:00, 146.74it/s, # trackers=35]\n","Processed Story Blocks: 100% 10/10 [00:00<00:00, 150.66it/s, # trackers=30]\n","Processed trackers: 100% 10/10 [00:00<00:00, 4573.44it/s, # actions=34]\n","Processed actions: 34it [00:00, 12878.74it/s, # examples=34]\n","Processed trackers: 100% 510/510 [00:01<00:00, 422.11it/s, # actions=518]\n","2020-09-29 05:16:30.260075: E tensorflow/stream_executor/cuda/cuda_driver.cc:351] failed call to cuInit: CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected\n","Epochs: 100% 100/100 [00:54<00:00,  1.82it/s, t_loss=0.086, loss=0.002, acc=1.000]\n","2020-09-29 05:17:32 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.utils.tensorflow.models\u001b[0m  - Finished training.\n","Processed trackers: 100% 10/10 [00:00<00:00, 4390.56it/s, # actions=33]\n","2020-09-29 05:17:32 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.core.agent\u001b[0m  - Persisted model to '/tmp/tmp196xx3c1/core'\n","\u001b[94mCore model training completed.\u001b[0m\n","\u001b[94mTraining NLU model...\u001b[0m\n","2020-09-29 05:17:33 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.training_data.training_data\u001b[0m  - Training data stats:\n","2020-09-29 05:17:33 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.training_data.training_data\u001b[0m  - Number of intent examples: 120 (11 distinct intents)\n","2020-09-29 05:17:33 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.training_data.training_data\u001b[0m  -   Found intents: 'ask_lower_stress', 'inform', 'greet', 'deny', 'affirm', 'ask_exercise', 'goodbye', 'thankyou', 'bot_challenge', 'out_of_scope', 'ask_eat_healthy'\n","2020-09-29 05:17:33 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.training_data.training_data\u001b[0m  - Number of response examples: 0 (0 distinct responses)\n","2020-09-29 05:17:33 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.training_data.training_data\u001b[0m  - Number of entity examples: 18 (3 distinct entities)\n","2020-09-29 05:17:33 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.training_data.training_data\u001b[0m  -   Found entity types: 'sleep', 'exercise', 'stress'\n","2020-09-29 05:17:33 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.model\u001b[0m  - Starting to train component WhitespaceTokenizer\n","2020-09-29 05:17:33 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.model\u001b[0m  - Finished training component.\n","2020-09-29 05:17:33 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.model\u001b[0m  - Starting to train component RegexFeaturizer\n","2020-09-29 05:17:33 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.model\u001b[0m  - Finished training component.\n","2020-09-29 05:17:33 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.model\u001b[0m  - Starting to train component LexicalSyntacticFeaturizer\n","2020-09-29 05:17:33 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.model\u001b[0m  - Finished training component.\n","2020-09-29 05:17:33 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.model\u001b[0m  - Starting to train component CountVectorsFeaturizer\n","2020-09-29 05:17:33 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.model\u001b[0m  - Finished training component.\n","2020-09-29 05:17:33 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.model\u001b[0m  - Starting to train component CountVectorsFeaturizer\n","2020-09-29 05:17:33 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.model\u001b[0m  - Finished training component.\n","2020-09-29 05:17:33 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.model\u001b[0m  - Starting to train component DIETClassifier\n","Epochs: 100% 100/100 [00:56<00:00,  1.78it/s, t_loss=2.196, i_loss=0.301, entity_loss=0.038, i_acc=0.983, entity_f1=0.866]\n","2020-09-29 05:18:42 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.utils.tensorflow.models\u001b[0m  - Finished training.\n","2020-09-29 05:18:42 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.model\u001b[0m  - Finished training component.\n","2020-09-29 05:18:42 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.model\u001b[0m  - Starting to train component EntitySynonymMapper\n","2020-09-29 05:18:42 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.model\u001b[0m  - Finished training component.\n","2020-09-29 05:18:42 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.model\u001b[0m  - Starting to train component ResponseSelector\n","2020-09-29 05:18:42 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.selectors.response_selector\u001b[0m  - Retrieval intent parameter was left to its default value. This response selector will be trained on training examples combining all retrieval intents.\n","2020-09-29 05:18:42 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.model\u001b[0m  - Finished training component.\n","2020-09-29 05:18:43 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa.nlu.model\u001b[0m  - Successfully saved model into '/tmp/tmp196xx3c1/nlu'\n","\u001b[94mNLU model training completed.\u001b[0m\n","\u001b[92mYour Rasa model is trained and saved at '/content/rasa-for-beginners/models/20200929-051843.tar.gz'.\u001b[0m\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"CJirF9xSTJ5Z","executionInfo":{"status":"ok","timestamp":1601357081304,"user_tz":-330,"elapsed":22369,"user":{"displayName":"Sparsh Agarwal","photoUrl":"","userId":"13037694610922482904"}},"outputId":"486b79f6-1ca5-4ae5-d678-b8068647d533","colab":{"base_uri":"https://localhost:8080/","height":156}},"source":["# !pip install colab_ssh\n","from colab_ssh import launch_ssh\n","launch_ssh('<token>', '<pass>')"],"execution_count":null,"outputs":[{"output_type":"stream","text":["Successfully running 2.tcp.ngrok.io:19039\n","[Optional] You can also connect with VSCode SSH Remote extension using this configuration:\n","\n","\tHost google_colab_ssh\n","\t\tHostName 2.tcp.ngrok.io\n","\t\tUser root\n","\t\tPort 19039\n","\t  \n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"CCHqKj0VTXSf","executionInfo":{"status":"ok","timestamp":1601365699519,"user_tz":-330,"elapsed":8944,"user":{"displayName":"Sparsh Agarwal","photoUrl":"","userId":"13037694610922482904"}},"outputId":"6a7150ab-aa64-47c0-f979-20888b19462b","colab":{"base_uri":"https://localhost:8080/","height":52}},"source":["!pip install -q pyngrok\n","from pyngrok import ngrok\n","!ngrok authtoken <token>"],"execution_count":null,"outputs":[{"output_type":"stream","text":["  Building wheel for pyngrok (setup.py) ... \u001b[?25l\u001b[?25hdone\n","Authtoken saved to configuration file: /root/.ngrok2/ngrok.yml\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"y2qhNPDT1EXe","executionInfo":{"status":"ok","timestamp":1601371988402,"user_tz":-330,"elapsed":2455,"user":{"displayName":"Sparsh Agarwal","photoUrl":"","userId":"13037694610922482904"}},"outputId":"614a8644-02bb-48a9-a30d-7ddad315dc96","colab":{"base_uri":"https://localhost:8080/","height":34}},"source":["ngrok.kill()\n","public_url = ngrok.connect(port='80'); public_url"],"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"application/vnd.google.colaboratory.intrinsic+json":{"type":"string"},"text/plain":["'http://c41fc2bf36a0.ngrok.io'"]},"metadata":{"tags":[]},"execution_count":31}]},{"cell_type":"code","metadata":{"id":"qdBxCylu1ZEY"},"source":["!rasa run -p 80"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"yLzYBzWf31-n"},"source":["# Twilio Endpoint\n","# http://c41fc2bf36a0.ngrok.io/webhooks/twilio/webhook"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"UlUzJhhGPZTI"},"source":["!mv /content/*.py /content/rasa-for-beginners/files\n","!mv /content/*.ipynb /content/rasa-for-beginners/files"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"h1HbDOtLRFms","executionInfo":{"status":"ok","timestamp":1601372977683,"user_tz":-330,"elapsed":1626,"user":{"displayName":"Sparsh Agarwal","photoUrl":"","userId":"13037694610922482904"}},"outputId":"5c0fad2d-e924-418a-896d-3bb1da273eb4","colab":{"base_uri":"https://localhost:8080/","height":34}},"source":["%cd /content"],"execution_count":null,"outputs":[{"output_type":"stream","text":["/content\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"hAn-zB4yRKbY"},"source":["# !mkdir -p /content/chatbots/wellnessTracker\n","# !mv  -v /content/rasa-for-beginners/* /content/chatbots/wellnessTracker/\n","# !rm -r /content/chatbots/wellnessTracker/.git"],"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["## RASA Action server"],"metadata":{"id":"uuQPV-sfX5lA"}},{"cell_type":"code","metadata":{"id":"PEVk_s57RAZf"},"source":["# !pip install rasa\n","# !pip3 install python-dotenv"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"kGHY6Do_RXC0"},"source":["!git clone https://github.com/RasaHQ/rasa-for-beginners.git"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"yVo-1GLmROoC","executionInfo":{"status":"ok","timestamp":1601363760012,"user_tz":-330,"elapsed":3749,"user":{"displayName":"Sparsh Agarwal","photoUrl":"","userId":"13037694610922482904"}},"outputId":"d8829330-8621-4fc9-b009-78a07811b1b0","colab":{"base_uri":"https://localhost:8080/","height":34}},"source":["%%writefile actions.py\n","\n","from typing import Any, Text, Dict, List, Union\n","from dotenv import load_dotenv\n","\n","from rasa_sdk import Action, Tracker\n","from rasa_sdk.executor import CollectingDispatcher\n","from rasa_sdk.forms import FormAction\n","\n","import requests\n","import json\n","import os\n","\n","load_dotenv()\n","\n","airtable_api_key='<api-key>' #os.getenv(\"AIRTABLE_API_KEY\")\n","base_id='<id>' #os.getenv(\"BASE_ID\")\n","table_name='TABLEX' #os.getenv(\"TABLE_NAME\")\n","\n","def create_health_log(confirm_exercise, exercise, sleep, diet, stress, goal):\n","    request_url=f\"https://api.airtable.com/v0/{base_id}/{table_name}\"\n","\n","    headers = {\n","        \"Content-Type\": \"application/json\",\n","        \"Accept\": \"application/json\",\n","        \"Authorization\": f\"Bearer {airtable_api_key}\",\n","    }  \n","    data = {\n","        \"fields\": {\n","            \"Exercised?\": confirm_exercise,\n","            \"Type of exercise\": exercise,\n","            \"Amount of sleep\": sleep,\n","            \"Stress\": stress,\n","            \"Diet\": diet,\n","            \"Goal\": goal,\n","        }\n","    }\n","    try:\n","        response = requests.post(\n","            request_url, headers=headers, data=json.dumps(data)\n","        )\n","        response.raise_for_status()\n","    except requests.exceptions.HTTPError as err:\n","        raise SystemExit(err)\n","    \n","    return response\n","    print(response.status_code)\n","\n","class HealthForm(FormAction):\n","\n","    def name(self):\n","        return \"health_form\"\n","\n","    @staticmethod\n","    def required_slots(tracker):\n","\n","        if tracker.get_slot('confirm_exercise') == True:\n","            return [\"confirm_exercise\", \"exercise\", \"sleep\",\n","             \"diet\", \"stress\", \"goal\"]\n","        else:\n","            return [\"confirm_exercise\", \"sleep\",\n","             \"diet\", \"stress\", \"goal\"]\n","\n","    def slot_mappings(self) -> Dict[Text, Union[Dict, List[Dict]]]:\n","        \"\"\"A dictionary to map required slots to\n","            - an extracted entity\n","            - intent: value pairs\n","            - a whole message\n","            or a list of them, where a first match will be picked\"\"\"\n","\n","        return {\n","            \"confirm_exercise\": [\n","                self.from_intent(intent=\"affirm\", value=True),\n","                self.from_intent(intent=\"deny\", value=False),\n","                self.from_intent(intent=\"inform\", value=True),\n","            ],\n","            \"sleep\": [\n","                self.from_entity(entity=\"sleep\"),\n","                self.from_intent(intent=\"deny\", value=\"None\"),\n","            ],\n","            \"diet\": [\n","                self.from_text(intent=\"inform\"),\n","                self.from_text(intent=\"affirm\"),\n","                self.from_text(intent=\"deny\"),\n","            ],\n","            \"goal\": [\n","                self.from_text(intent=\"inform\"),\n","            ],\n","        }\n","\n","    def submit(\n","        self,\n","        dispatcher: CollectingDispatcher,\n","        tracker: Tracker,\n","        domain: Dict[Text, Any],\n","    ) -> List[Dict]:\n","\n","        confirm_exercise = tracker.get_slot(\"confirm_exercise\")\n","        exercise = tracker.get_slot(\"exercise\")\n","        sleep = tracker.get_slot(\"sleep\")\n","        stress = tracker.get_slot(\"stress\")\n","        diet = tracker.get_slot(\"diet\")\n","        goal = tracker.get_slot(\"goal\")\n","\n","        response = create_health_log(\n","                confirm_exercise=confirm_exercise,\n","                exercise=exercise,\n","                sleep=sleep,\n","                stress=stress,\n","                diet=diet,\n","                goal=goal\n","            )\n","\n","        dispatcher.utter_message(\"Thanks, your answers have been recorded!\")\n","        return []"],"execution_count":null,"outputs":[{"output_type":"stream","text":["Overwriting actions.py\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"l8HpM34cQ94Q","executionInfo":{"status":"ok","timestamp":1601363765937,"user_tz":-330,"elapsed":2654,"user":{"displayName":"Sparsh Agarwal","photoUrl":"","userId":"13037694610922482904"}},"outputId":"29a0ff19-65c4-4da9-c0f9-35d587e03f96","colab":{"base_uri":"https://localhost:8080/","height":34}},"source":["# !pip install -q pyngrok\n","# from pyngrok import ngrok\n","# !ngrok authtoken <token>\n","ngrok.kill()\n","public_url = ngrok.connect(port='5055'); public_url"],"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"application/vnd.google.colaboratory.intrinsic+json":{"type":"string"},"text/plain":["'http://d40440ffde45.ngrok.io'"]},"metadata":{"tags":[]},"execution_count":18}]},{"cell_type":"code","metadata":{"id":"CJC5-67zRw2l","executionInfo":{"status":"ok","timestamp":1601373399992,"user_tz":-330,"elapsed":7284826,"user":{"displayName":"Sparsh Agarwal","photoUrl":"","userId":"13037694610922482904"}},"outputId":"9cb32b23-f095-4a9e-88a3-5e4422d92e55","colab":{"base_uri":"https://localhost:8080/","height":138}},"source":["!rasa run actions"],"execution_count":null,"outputs":[{"output_type":"stream","text":["2020-09-29 07:16:10 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa_sdk.endpoint\u001b[0m  - Starting action endpoint server...\n","2020-09-29 07:16:10 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa_sdk.executor\u001b[0m  - Registered function for 'health_form'.\n","2020-09-29 07:16:10 \u001b[1;30mINFO    \u001b[0m \u001b[34mrasa_sdk.endpoint\u001b[0m  - Action endpoint is up and running on http://localhost:5055\n","2020-09-29 09:13:26 \u001b[1;30mERROR   \u001b[0m \u001b[34mrasa_sdk.endpoint\u001b[0m  - \u001b[31mFailed to extract slot confirm_exercise with action health_form\u001b[0m\n","2020-09-29 09:15:14 \u001b[1;30mERROR   \u001b[0m \u001b[34mrasa_sdk.endpoint\u001b[0m  - \u001b[31mFailed to extract slot confirm_exercise with action health_form\u001b[0m\n","2020-09-29 09:32:13 \u001b[1;30mERROR   \u001b[0m \u001b[34mrasa_sdk.endpoint\u001b[0m  - \u001b[31mFailed to extract slot stress with action health_form\u001b[0m\n","2020-09-29 09:34:28 \u001b[1;30mERROR   \u001b[0m \u001b[34mrasa_sdk.endpoint\u001b[0m  - \u001b[31mFailed to extract slot exercise with action health_form\u001b[0m\n"],"name":"stdout"}]}]}