Pretrained models

Dear NLP heros
I’m trying to implement a machine translation on my own dataset. using this week’s assignment
I did it step by step everything ok but in the model instantiate when we used

initialize weights from a pre-trained model

model.init_from_file(“model.pkl.gz”, weights_only=True)

I don’t have any pre-trained model. what should I do instead?
best regards

You can download the pre trained model from the Jupyter instance. If you go to file → open on the assignment page, you can find the weights.

Dear david
my source translation lang is persian
does this appropriate for my dataset?

I think it should have Persian. It supports 100 languages.