ChristopherA08/IndoELECTRA

This repository contains the pre-trained Electra Base model (tensorflow 1.15.0) trained in a Large Indonesian corpus . The training of the model has been performed using Google's original Tensorflow code on eight core Google Cloud TPU v2.0 . The model is based on ELECTRA architecture for the Indonesian Language . The data was used to train a large corpus of 16GB of raw text and 2B Indonesian words . The code was trained using Google Cloud Storage bucket, for persistent storage of training data and models. The training procedure is described as a training procedure. The data is trained using a Google Cloud storage bucket, with the model being trained using an original T,