-
-
Save nmstoker/04355dc113608fe88b999ea1b95fb355 to your computer and use it in GitHub Desktop.
from rasa_nlu.converters import load_data | |
# This re-uses the Rasa NLU converters code to turn a JSON Rasa NLU training | |
# file into MD format and save it | |
# Assumes you have Rasa NLU installed :-) | |
# If you want other options, look at the NLU code to work out how to handle them | |
# USE AT YOUR OWN RISK | |
input_training_file = '/home/username/Projects/your-project/data/training.json' | |
# ******************************************************* | |
# TAKE CARE: output_md_file is overwritten automatically | |
# ******************************************************* | |
output_md_file = '/home/username/Projects/your-project/data/training.md' | |
with open(output_md_file,'w') as f: | |
f.write(load_data(input_training_file).as_markdown()) |
Another method.
from rasa.nlu.convert import convert_training_data
convert_training_data(data_file="ip_file.json", out_file="out_file.md", output_format="md", language="")
What are the changes that I need to make to make this work for Rasa 1.1.4?
Another method.
from rasa.nlu.convert import convert_training_data
convert_training_data(data_file="ip_file.json", out_file="out_file.md", output_format="md", language="")
@KarthiAru I am using Rasa1.x and this worked fine for me.
@sivu1, I'm getting the below error
ImportError: No module named rasa.nlu.convert
I just installed rasa using the below command as mentioned in https://rasa.com/docs/getting-started/
pip install rasa-x --extra-index-url https://pypi.rasa.com/simple
Should I install any additional packages?
@KarthiAru In a fresh environment, just try pip install rasa
That will install the latest version 1.1.4. You will also get rasa x. The docs is not yet updated.
@KarthiAru In a fresh environment, just try
pip install rasa
That will install the latest version 1.1.4. You will also get rasa x. The docs is not yet updated.
Thanks, @sivu1. This worked!
Another method.
from rasa.nlu.convert import convert_training_data
convert_training_data(data_file="ip_file.json", out_file="out_file.md", output_format="md", language="")
ya its working. Thanks
from rasa_nlu.convert import convert_training_data
convert_training_data(data_file="nlu.json", out_file="out_file.md", output_format="md", language="")
I think it's also wise to explicitly set the encoding to utf-8 when creating the file. Thanks for the gist!