You can deploy this model and start earning money today!

roequitz/
t5-abs-2309-1054-lr-0.001-bs-10-maxep-20

Name: t5-abs-2309-1054-lr-0.001-bs-10-maxep-20
Availability: InStock

safetensors

generated_from_trainer

google-t5/t5-base

finetune

apache-2.0

Views

No views yet

t5-abs-2309-1054-lr-0.001-bs-10-maxep-20

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

Loss: 2.8637
Rouge/rouge1: 0.4646
Rouge/rouge2: 0.2164
Rouge/rougel: 0.3909
Rouge/rougelsum: 0.3918
Bertscore/bertscore-precision: 0.8938
Bertscore/bertscore-recall: 0.8967
Bertscore/bertscore-f1: 0.8951
Meteor: 0.428
Gen Len: 42.1091

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: 0.001
train_batch_size: 10
eval_batch_size: 10
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 20
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 20
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge/rouge1	Rouge/rouge2	Rouge/rougel	Rouge/rougelsum	Bertscore/bertscore-precision	Bertscore/bertscore-recall	Bertscore/bertscore-f1	Meteor	Gen Len
0.3887	0.9885	43	2.6672	0.4542	0.2002	0.3796	0.3796	0.891	0.8957	0.8932	0.4204	43.5818
0.2721	2.0	87	2.7485	0.4354	0.1974	0.3693	0.3693	0.8915	0.8925	0.8918	0.4012	41.1818
0.2408	2.9885	130	2.8631	0.4364	0.1874	0.3679	0.3694	0.8935	0.8896	0.8914	0.3822	38.0455
0.2241	4.0	174	2.8368	0.4655	0.2128	0.395	0.3965	0.8978	0.8973	0.8975	0.4191	40.3364
0.1864	4.9885	217	2.9424	0.4515	0.2049	0.3805	0.3809	0.8914	0.8972	0.8942	0.4153	45.0909
0.1389	6.0	261	2.9015	0.4604	0.2021	0.3835	0.3839	0.8944	0.8952	0.8947	0.4148	41.1273
0.1076	6.9885	304	2.9469	0.4608	0.2096	0.3901	0.3898	0.8934	0.8964	0.8947	0.4271	42.8364
0.0887	8.0	348	2.7522	0.4573	0.2137	0.381	0.3811	0.8931	0.896	0.8944	0.4246	42.6818
0.0775	8.9885	391	2.7045	0.4555	0.2091	0.3843	0.3846	0.8928	0.8946	0.8936	0.4172	41.9273
0.0673	10.0	435	2.7049	0.4762	0.2229	0.4017	0.4027	0.8961	0.8979	0.8969	0.4369	41.7636
0.0618	10.9885	478	2.7250	0.4716	0.2218	0.3976	0.398	0.894	0.8977	0.8957	0.4389	42.8182
0.0524	12.0	522	2.7454	0.4596	0.2096	0.3886	0.3895	0.8945	0.8962	0.8952	0.4205	41.3545
0.0503	12.9885	565	2.7736	0.4556	0.2084	0.3826	0.3843	0.8942	0.8937	0.8938	0.4077	40.2455
0.0477	14.0	609	2.8055	0.4635	0.2157	0.3918	0.393	0.8959	0.8967	0.8962	0.4259	40.7727
0.0457	14.9885	652	2.8190	0.4657	0.2136	0.3933	0.3943	0.8954	0.8955	0.8953	0.4204	40.3818
0.0425	16.0	696	2.8323	0.4645	0.2134	0.3874	0.389	0.8946	0.8951	0.8948	0.4216	40.5545
0.0431	16.9885	739	2.8401	0.4639	0.2136	0.3893	0.39	0.8947	0.8962	0.8953	0.4237	41.4727
0.0413	18.0	783	2.8531	0.459	0.2092	0.3852	0.3862	0.8939	0.8956	0.8946	0.419	41.5636
0.0419	18.9885	826	2.8607	0.462	0.2116	0.3868	0.3881	0.8937	0.8961	0.8948	0.4239	41.9455
0.0403	19.7701	860	2.8637	0.4646	0.2164	0.3909	0.3918	0.8938	0.8967	0.8951	0.428	42.1091

Framework versions

Transformers 4.44.0
Pytorch 2.4.0
Datasets 2.21.0
Tokenizers 0.19.1

roequitz/t5-abs-2309-1054-lr-0.001-bs-10-maxep-20

t5-abs-2309-1054-lr-0.001-bs-10-maxep-20

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

roequitz/
t5-abs-2309-1054-lr-0.001-bs-10-maxep-20