byt5-small-wikipron-eng-latn-ca-broad

This model is a fine-tuned version of google/byt5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1882
• Per: 0.33
• Gen Len: 16.2671

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: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_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
• num_epochs: 10.0

Training results

Training Loss	Epoch	Step	Validation Loss	Per	Gen Len
2.2268	1.0	238	0.3982	0.5115	16.2763
0.3822	2.0	477	0.2639	0.4174	16.0942
0.2862	3.0	715	0.2263	0.3669	16.1344
0.245	4.0	954	0.2080	0.3507	16.2382
0.2224	5.0	1192	0.2005	0.3414	16.2146
0.2067	6.0	1431	0.1960	0.3361	16.241
0.1964	7.0	1669	0.1917	0.3352	16.2499
0.1874	8.0	1908	0.1882	0.3319	16.2605
0.1828	9.0	2146	0.1889	0.3318	16.2561
0.1777	9.98	2380	0.1882	0.33	16.2671

Framework versions

• Transformers 4.29.2
• Pytorch 2.0.1+cu117
• Datasets 2.12.0
• Tokenizers 0.13.3