bart-abs-2409-0144-lr-3e-05-bs-4-maxep-6

roequitz/bart-abs-2409-0144-lr-3e-05-bs-4-maxep-6

bart-abs-2409-0144-lr-3e-05-bs-4-maxep-6 is a machine learning model from roequitz. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

17% of all-time downloads

All-time downloads

Public

Parameters

306M

1.2 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors1.2 GB · 100%

At a glance

License: apache-2.0
Model type: bart
Access: Public
Created: Sep 24, 2024
Updated: Sep 24, 2024
SHA: 08ed05d5

Base models

sshleifer/distilbart-xsum-12-6

Model card

From the Hugging Face model README

bart-abs-2409-0144-lr-3e-05-bs-4-maxep-6

This model is a fine-tuned version of sshleifer/distilbart-xsum-12-6 on the None dataset. It achieves the following results on the evaluation set:

Loss: 2.3498
Rouge/rouge1: 0.4687
Rouge/rouge2: 0.2167
Rouge/rougel: 0.3984
Rouge/rougelsum: 0.3997
Bertscore/bertscore-precision: 0.8968
Bertscore/bertscore-recall: 0.8932
Bertscore/bertscore-f1: 0.8948
Meteor: 0.4183
Gen Len: 37.6273

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: 3e-05
train_batch_size: 4
eval_batch_size: 4
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 6
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
2.3602	1.0	217	2.0806	0.4443	0.2067	0.3739	0.3745	0.8966	0.8891	0.8927	0.3937	38.2364
1.6559	2.0	434	1.9951	0.4581	0.22	0.3968	0.3973	0.8994	0.8919	0.8955	0.4032	35.1273
1.2943	3.0	651	2.0581	0.4592	0.2156	0.3932	0.3942	0.8995	0.8924	0.8958	0.4071	35.2364
1.0197	4.0	868	2.1312	0.478	0.2304	0.409	0.4111	0.9003	0.8935	0.8967	0.4209	36.1091
0.8117	5.0	1085	2.2532	0.4878	0.2376	0.4181	0.4194	0.9009	0.8945	0.8976	0.4293	35.2636
0.6733	6.0	1302	2.3498	0.4687	0.2167	0.3984	0.3997	0.8968	0.8932	0.8948	0.4183	37.6273

Framework versions

Transformers 4.44.0
Pytorch 2.4.0
Datasets 2.21.0
Tokenizers 0.19.1

Type: bart
License: apache-2.0
Created: Sep 24, 2024
Updated: Sep 24, 2024