- I am using rasa with the pipeline spacy_sklearn.
- ner is the main motto of my project everything is going good but training for only two son files its taking around 1hr
- The size of two json files are 15mb each
- How to reduce the training time pls anyone help me out
and how to increase parallel jobs = 1 to 10
What version of rasa are you using, and how are you training your bot – via the command line, python API, etc?
I am using rasa-nlu==0.15.1
I am training the bot using rasa with pipeline space_sklearn
through python I am training the bot
import spacy
#nlp=spacy.load(‘en’)
import subprocess
from rasa_nlu.training_data import load_data
from rasa_nlu.config import RasaNLUModelConfig
from rasa_nlu.model import Trainer
from rasa_nlu import config
from rasa_nlu.model import Metadata, Interpreter
import time
training_data=load_data(“data/dt/geo_train.json”)
#config backend using spacy and sklearn
trainer= Trainer(config.load(“config_spacy.yml”))
print (“time.ctime() : %s” % time.ctime())
trainer.train(training_data)
print (“time.ctime() : %s” % time.ctime())
#directory to store our model
model_directory= trainer.persist("./projects/")
#making predictions owth the model
interpreter= Interpreter.load(model_directory)
see my code