tokipona_to_eng_model_v1

This model is a fine-tuned version of t5-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.2414
• Bleu: 25.378
• Gen Len: 8.8964

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: 0.0001
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
1.8666	1.0	1430	1.4045	17.3719	8.7792
1.4468	2.0	2860	1.2761	20.4928	8.9122
1.2337	3.0	4290	1.2259	22.5629	8.8094
1.1073	4.0	5720	1.2052	23.2202	8.9101
1.0195	5.0	7150	1.2041	24.1567	8.9248
0.9152	6.0	8580	1.2082	24.1598	8.9977
0.8753	7.0	10010	1.2128	24.7257	8.8864
0.818	8.0	11440	1.2249	25.0503	8.9076
0.7854	9.0	12870	1.2341	25.0301	8.9097
0.7655	10.0	14300	1.2414	25.378	8.8964

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0