indo-t5-base-v2-nusax

This model is a fine-tuned version of lazarus-project/indo-t5-base-v2 on the lazarus-project/NusaX-MT dataset. It achieves the following results on the evaluation set:

• Loss: 2.5238
• Bleu: 15.7908
• Gen Len: 48.9244

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0002
• train_batch_size: 32
• eval_batch_size: 16
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• total_train_batch_size: 256
• total_eval_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
1.8317	2.33	500	2.4804	3.1525	18.6295
1.4375	4.65	1000	2.5238	3.2043	18.62

Framework versions

• Transformers 4.27.1
• Pytorch 2.0.0+cu117
• Datasets 2.9.0
• Tokenizers 0.13.2