# Load two models from one interpreter?

**URL:** <https://forum.rasa.com/t/load-two-models-from-one-interpreter/12405>\
**Category:** Rasa Open Source\
**Created:** [July 3, 2019, 5:05am UTC](https://forum.rasa.com/t/load-two-models-from-one-interpreter/12405 "2019-07-03T05:05:22Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![rajesh2211](https://avatars.discourse-cdn.com/v4/letter/r/b3f665/32.png) [@rajesh2211](https://forum.rasa.com/u/rajesh2211)\
**Post date:** [July 3, 2019, 5:05am UTC](https://forum.rasa.com/t/load-two-models-from-one-interpreter/12405/1 "2019-07-03T05:05:23Z")

</div>

i want to load two separate models with one interpreter . I have already trained models and kept them in two different directories now what i want is to load bith the models at once Thanks in advance

---

<div class="post-metadata">

**Author:** ![akelad](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/akelad/32/69_2.png) [@akelad](https://forum.rasa.com/u/akelad)\
**Post date:** [July 5, 2019, 2:22pm UTC](https://forum.rasa.com/t/load-two-models-from-one-interpreter/12405/2 "2019-07-05T14:22:58Z")

</div>

What kind of models? And which Rasa version are you using?

---

<div class="post-metadata">

**Author:** ![rajesh2211](https://avatars.discourse-cdn.com/v4/letter/r/b3f665/32.png) [@rajesh2211](https://forum.rasa.com/u/rajesh2211)\
**Post date:** [July 6, 2019, 8:26am UTC](https://forum.rasa.com/t/load-two-models-from-one-interpreter/12405/3 "2019-07-06T08:26:30Z")

</div>

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

and what I want is I will train two different models and interpreter should give the prediction from both the models

---

<div class="post-metadata">

**Author:** ![akelad](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/akelad/32/69_2.png) [@akelad](https://forum.rasa.com/u/akelad)\
**Post date:** [July 8, 2019, 3:40pm UTC](https://forum.rasa.com/t/load-two-models-from-one-interpreter/12405/4 "2019-07-08T15:40:42Z")

</div>

that’s not possible. You can load up multiple models into memory with the projects feature, but they won’t make predictions at the same time