Hi there. I’m trying to use a pretrained sentiment component nltk mention in thi tutorial How to Enhance Rasa NLU Models with Custom Components | Rasa Blog | The Rasa Blog | Rasa but when i run rasa train I have this error Traceback (most recent call last): File "C:\Users\FARES\anaconda3\envs\rasa2.8\lib\runpy.py", line 192, in _run_module_as_main return _run_code(code, main_globals, None, File "C:\Users\FARES\anaconda3\envs\rasa2.8\lib\runpy.py", line 85, in _run_code exec(code, run_globals) File "C:\Users\FARES\anaconda3\envs\rasa2.8\Scripts\rasa.exe\__main__.py", line 7, in <module> File "C:\Users\FARES\anaconda3\envs\rasa2.8\lib\site-packages\rasa\__main__.py", line 117, in main cmdline_arguments.func(cmdline_arguments) File "C:\Users\FARES\anaconda3\envs\rasa2.8\lib\site-packages\rasa\cli\train.py", line 59, in <lambda> train_parser.set_defaults(func=lambda args: run_training(args, can_exit=True)) File "C:\Users\FARES\anaconda3\envs\rasa2.8\lib\site-packages\rasa\cli\train.py", line 91, in run_training training_result = train_all( File "C:\Users\FARES\anaconda3\envs\rasa2.8\lib\site-packages\rasa\api.py", line 109, in train return rasa.utils.common.run_in_loop( File "C:\Users\FARES\anaconda3\envs\rasa2.8\lib\site-packages\rasa\utils\common.py", line 296, in run_in_loop result = loop.run_until_complete(f) File "C:\Users\FARES\anaconda3\envs\rasa2.8\lib\asyncio\base_events.py", line 608, in run_until_complete return future.result() File "C:\Users\FARES\anaconda3\envs\rasa2.8\lib\site-packages\rasa\model_training.py", line 108, in train_async return await _train_async_internal( File "C:\Users\FARES\anaconda3\envs\rasa2.8\lib\site-packages\rasa\model_training.py", line 288, in _train_async_internal await _do_training( File "C:\Users\FARES\anaconda3\envs\rasa2.8\lib\site-packages\rasa\model_training.py", line 352, in _do_training await _train_core_with_validated_data( File "C:\Users\FARES\anaconda3\envs\rasa2.8\lib\site-packages\rasa\model_training.py", line 549, in _train_core_with_validated_data await rasa.core.train.train( File "C:\Users\FARES\anaconda3\envs\rasa2.8\lib\site-packages\rasa\core\train.py", line 70, in train agent.train(training_data, **additional_arguments) File "C:\Users\FARES\anaconda3\envs\rasa2.8\lib\site-packages\rasa\core\agent.py", line 753, in train self.policy_ensemble.train( File "C:\Users\FARES\anaconda3\envs\rasa2.8\lib\site-packages\rasa\core\policies\ensemble.py", line 206, in train policy.train( File "C:\Users\FARES\anaconda3\envs\rasa2.8\lib\site-packages\rasa\core\policies\ted_policy.py", line 676, in train model_data, label_ids = self._prepare_for_training( File "C:\Users\FARES\anaconda3\envs\rasa2.8\lib\site-packages\rasa\core\policies\ted_policy.py", line 570, in _prepare_for_training tracker_state_features, label_ids, entity_tags = self._featurize_for_training( File "C:\Users\FARES\anaconda3\envs\rasa2.…

Finnaly, I solved this problem. I used this: res = sid.polarity_scores(str(message.get("text"))) instead of res = sid.polarity_scores(message.text) and every thing is working now :muscle: :muscle: @nik202 thank you very much for your responses and encouragement :muscle:.

and this is my sentiment.py file from typing import Text from rasa.nlu.components import Component from rasa.nlu import utils from rasa.nlu.model import Metadata import nltk from nltk.sentiment.vader import SentimentIntensityAnalyzer import os class SentimentAnalyzer(Component): """A pre-trained sentiment component""" name = "sentiment" provides = ["entities"] requires = [] defaults = {} language_list = ["en"] def __init__(self, component_config=None): super(SentimentAnalyzer, self).__init__(component_config) def train(self, training_data, cfg, **kwargs): """Not needed, because the the model is pretrained""" pass def convert_to_rasa(self, value, confidence): """Convert model output into the Rasa NLU compatible output format.""" entity = {"value": value, "confidence": confidence, "entity": "sentiment", "extractor": "sentiment_extractor"} return entity def process(self, message, **kwargs): """Retrieve the text message, pass it to the classifier and append the prediction results to the message class.""" sid = SentimentIntensityAnalyzer() res = sid.polarity_scores(message.text) key, value = max(res.items(), key=lambda x: x[1]) entity = self.convert_to_rasa(key, value) message.set("entities", [entity], add_to_output=True) def persist(self, file_name, model_dir): """Pass because a pre-trained model is already persisted""" pass

Now, I realised that this code (sentiment.py) works without problems with rasa 1.9 !!! but for me i use rasa 2.8 can someone tell me how to fix this problem?

thanks, @nik202 I changed message.text by message message.data[‘text’] and this is the error File "E:\act\sentiment.py", line 44, in process res = sid.polarity_scores(message.data['text']) KeyError: 'text'

@Fares yes, its obvious, but seriously I not implemented this code :frowning: and I am trying to help you only dear!

thanks a lot @nik202 :slightly_smiling_face:

@Fares I know and trust you that you will crack this issue :slight_smile:

@Juste can you please help me in this problem?

@Juste It looks like you are not here :roll_eyes: It wasn’t the first time I mention you in other problem but you never respond :smirk: thank you any way

@Fares I told you and trust you, many congratulations :clap::clap:

Pretrained sentiment component nltk

Rasa Open Source

nik202 (NiK202) September 22, 2021, 1:33pm 4

@Fares For you I have seen this issue it related to message.text, can you please follow this thread and try sort your code? Message object does not recognize text attribute in Custom Component for RASA 2.0 - #11 by fabrice-toussaint

Topic		Replies	Views
Trying implement custom component with sentiment nltk only Rasa Open Source	5	1531	August 14, 2020
How can i implement a sentiment analysis in RASA 2.x Rasa Open Source	3	1134	November 28, 2020
How to use own model in pipeline? Rasa Open Source	24	4491	March 1, 2021
Getting Custom Component to Work Rasa Open Source	8	1620	September 7, 2021
Custom sentiment analysis components issue with Core Rasa Open Source	1	1186	January 27, 2022

Pretrained sentiment component nltk

Related topics