SEBIS/code_trans_t5_large_program_synthese_multitask

The CodeTrans model is based on the t5-large model architecture . It has its own SentencePiece vocabulary model . It could be used to generate lisp inspired DSL code given the human language description tasks . The model was trained on a single TPU Pod V3-8 for 220,000 steps in total, using sequence length 512 (batch size 4096) It has a total of approximately 220M parameters and was trained using the encoder-decoder architecture . The optimizer used is AdaFactor with inverse square root learning rate schedule for pre-training . The code documentation tasks, different models achieves the following results on different programming languages (in BLEU score): test results,