Naveen-k/KanBERTo

1M data samples are used to train this model from OSCAR page(https://traces1.inria.fr/oscar) The data set is of 1.7 GB due to resource constraint to train model . If you are interested in collaboration and have computational resources to train on you are most welcome to do so . The model is for anyone who wants to make use of kannada language models for various tasks like language generation, translation and many more use cases . It is a small language model for Kannada . It has been trained for 12 epochs and save model for every 10k steps is set for 2,000 steps . The training parameters are set to,