sdyy's picture
Update README.md
20334b8 verified
I used the codes on this page
Thank you to whoever provided the codes
https://github.com/huggingface/transformers/tree/main/examples/pytorch/translation
I changed the model language and created a dataset file using AI
https://huggingface.co/datasets/sdyy/en-ar
All training is on free CPU in colab
an idea
If the data set could be divided into smaller parts
Choose a free template
Several individuals who are programming enthusiasts participate and coordinate among them
A small, raw, free language model can be trained on a specific type of data
Python programming language or translation from one language to another
Or train the free model on large data sets such as Wikipedia
Instead of working hard individually, participating in training for one goal turns the smallest, raw, free models into something wonderful
And this is just an idea