SEBIS/code_trans_t5_base_code_documentation_generation_ruby_multitask

The CodeTrans model is based on the t5-base model . It is trained on tokenized ruby code functions . It can be used on unparsed and untokenized Ruby code . The model could be used to generate the description for the ruby function or be fine-tuned on other ruby code tasks . It has a total of approximately 220M parameters and was trained using the encoder-decoder architecture. The model was trained on a single TPU Pod V3-8 for 160,000 steps in total, using sequence length 512 (batch size 4096) The model has its own SentencePiece vocabulary model. It was trained with the optimizer used is AdaFactor,