byt5-base-es_kbh

This model is a fine-tuned version of google/byt5-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7315
• Bleu: 5.1736
• Gen Len: 100.0266

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: 5e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 65
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10.0

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
No log	1.0	395	1.0572	0.0232	19.0
1.5347	2.0	790	0.9249	0.1035	19.0
1.0539	3.0	1185	0.8549	0.1313	19.0
0.9461	4.0	1580	0.8109	0.1392	19.0
0.9461	5.0	1975	0.7860	0.1545	19.0
0.8824	6.0	2370	0.7620	0.1756	19.0
0.8447	7.0	2765	0.7495	0.1705	19.0
0.814	8.0	3160	0.7384	0.14	19.0
0.8031	9.0	3555	0.7331	0.1807	19.0
0.8031	10.0	3950	0.7315	0.1841	19.0

Framework versions

• Transformers 4.27.4
• Pytorch 2.0.0+cu118
• Datasets 2.11.0
• Tokenizers 0.13.3