metadata

library_name: transformers
license: apache-2.0
base_model: google-t5/t5-small
tags:
  - translation
  - generated_from_trainer
metrics:
  - bleu
model-index:
  - name: t5-small-finetuned-english-to-hausa
    results: []

t5-small-finetuned-english-to-hausa

This model is a fine-tuned version of google-t5/t5-small on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.6851
Bleu: 71.9442
Gen Len: 14.3679

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: 0.0008
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: 20
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
1.2594	1.0	1497	0.8236	59.41	14.2172
0.7848	2.0	2994	0.6581	64.4839	14.219
0.6172	3.0	4491	0.5897	66.4564	14.2357
0.5151	4.0	5988	0.5619	68.0986	14.4905
0.4457	5.0	7485	0.5477	69.2175	14.4141
0.3938	6.0	8982	0.5413	70.0663	14.4059
0.3555	7.0	10479	0.5338	70.1734	14.4734
0.3154	8.0	11976	0.5485	70.3692	14.3035
0.2837	9.0	13473	0.5454	70.7837	14.4556
0.2507	10.0	14970	0.5616	70.976	14.3807
0.2265	11.0	16467	0.5728	71.2008	14.3692
0.2041	12.0	17964	0.5808	71.4766	14.362
0.1848	13.0	19461	0.5981	71.3804	14.3114
0.1715	14.0	20958	0.6122	71.43	14.4295
0.1547	15.0	22455	0.6309	71.753	14.351
0.1417	16.0	23952	0.6411	71.7608	14.3513
0.1267	17.0	25449	0.6612	71.93	14.4243
0.1208	18.0	26946	0.6662	71.8591	14.3486
0.1076	19.0	28443	0.6799	72.0417	14.3862
0.1046	20.0	29940	0.6851	71.9442	14.3679

Framework versions

Transformers 4.44.2
Pytorch 2.3.1+cu121
Datasets 2.21.0
Tokenizers 0.19.1