Memory Handling about disco HOT 1 CLOSED

So I did some research using Vuex and creating a store for the entire app. This allows having a shared state between components. This might be useful if we want to create functionality across tasks.
However, this does not solve the aforementioned problem. There's still the error that TensorFlow js does not recognize a network saved in a component, even if the component has been stored in the shared store.
I might push the store file and all the modifications added to bring the store to life as it might be useful for later use.
@mmilenkoski are okay with the solution of not having a model variable stored in the component? I will share the model's name with the functions that require the model and using this name the function will be able to load the model from the indexeddb database.
If my explanations are not clear, I'm also available for a zoom. Thanks!

from disco.