Hoax0930

This model is a fine-tuned version of Helsinki-NLP/opus-tatoeba-en-ja on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 5.2891 - Bleu: 0.3128 The following hyperparameters were used during training: - learningrate: 2e-05 - trainbatchsize: 32 - evalbatchsize: 64 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lrschedulertype: linear - numepochs: 3 - mixedprecisiontraining: Native AMP - Transformers 4.22.2 - Pytorch 1.12.1+cu113 - Datasets 2.5.1 - Tokenizers 0.12.1

This model is a fine-tuned version of Helsinki-NLP/opus-tatoeba-en-ja on the kde4 dataset. It achieves the following results on the evaluation set: - Loss: 0.9825 - Bleu: 37.1098 The following hyperparameters were used during training: - learningrate: 2e-05 - trainbatchsize: 32 - evalbatchsize: 64 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lrschedulertype: linear - numepochs: 3 - mixedprecisiontraining: Native AMP - Transformers 4.21.2 - Pytorch 1.12.1+cu113 - Datasets 2.4.0 - Tokenizers 0.12.1