summarise_v11

debbiesoon/summarise_v11

summarise_v11 is a machine learning model from debbiesoon. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as apache-2.0.

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

Downloads · 30 days

4% of all-time downloads

All-time downloads

383

Public

Repo size

1.9 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin648 MB · 99%

At a glance

Library: transformers
License: apache-2.0
Model type: led
Access: Public
Created: Nov 2, 2022
Updated: Nov 2, 2022
SHA: 9cfbf518

Model card

From the Hugging Face model README

summarise_v11

This model is a fine-tuned version of allenai/led-base-16384 on the None dataset. It achieves the following results on the evaluation set:

Loss: 1.6322
Rouge1 Precision: 0.6059
Rouge1 Recall: 0.6233
Rouge1 Fmeasure: 0.5895
Rouge2 Precision: 0.4192
Rouge2 Recall: 0.4512
Rouge2 Fmeasure: 0.4176
Rougel Precision: 0.4622
Rougel Recall: 0.4946
Rougel Fmeasure: 0.4566
Rougelsum Precision: 0.4622
Rougelsum Recall: 0.4946
Rougelsum Fmeasure: 0.4566

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: 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: 10
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1 Precision	Rouge1 Recall	Rouge1 Fmeasure	Rouge2 Precision	Rouge2 Recall	Rouge2 Fmeasure	Rougel Precision	Rougel Recall	Rougel Fmeasure	Rougelsum Precision	Rougelsum Recall	Rougelsum Fmeasure
1.6201	0.45	10	1.4875	0.3203	0.64	0.3932	0.197	0.3839	0.2385	0.1952	0.4051	0.2454	0.1952	0.4051	0.2454
0.9172	0.91	20	1.4404	0.4917	0.5134	0.4699	0.288	0.3095	0.276	0.3371	0.3594	0.3277	0.3371	0.3594	0.3277
1.0923	1.36	30	1.3575	0.519	0.5505	0.4936	0.3114	0.3237	0.2958	0.3569	0.3702	0.3364	0.3569	0.3702	0.3364
1.1287	1.82	40	1.3269	0.4913	0.5997	0.5068	0.3108	0.3964	0.3269	0.3355	0.427	0.3521	0.3355	0.427	0.3521
0.9938	2.27	50	1.3189	0.5339	0.5781	0.4973	0.3555	0.3883	0.3345	0.3914	0.4289	0.3678	0.3914	0.4289	0.3678
0.8659	2.73	60	1.3241	0.525	0.638	0.5165	0.3556	0.4349	0.3535	0.3914	0.4793	0.3886	0.3914	0.4793	0.3886
0.6187	3.18	70	1.3360	0.5875	0.5864	0.5416	0.4005	0.4045	0.3701	0.4485	0.4556	0.414	0.4485	0.4556	0.414
0.3941	3.64	80	1.4176	0.5373	0.6415	0.5328	0.3576	0.446	0.3642	0.3787	0.4586	0.3781	0.3787	0.4586	0.3781
0.4145	4.09	90	1.3936	0.4127	0.6553	0.4568	0.2568	0.4498	0.2988	0.2918	0.4933	0.328	0.2918	0.4933	0.328
0.4203	4.55	100	1.4703	0.6545	0.601	0.5981	0.4789	0.4373	0.438	0.5251	0.4851	0.4818	0.5251	0.4851	0.4818
0.687	5.0	110	1.4304	0.5566	0.6357	0.5637	0.3734	0.4186	0.3748	0.4251	0.4825	0.4286	0.4251	0.4825	0.4286
0.4006	5.45	120	1.5399	0.5994	0.5794	0.5515	0.4215	0.4218	0.398	0.4359	0.4369	0.4084	0.4359	0.4369	0.4084
0.2536	5.91	130	1.5098	0.5074	0.6254	0.4874	0.3369	0.4189	0.3256	0.3802	0.4738	0.3664	0.3802	0.4738	0.3664
0.2218	6.36	140	1.5278	0.5713	0.6059	0.5688	0.3887	0.4233	0.3916	0.4414	0.4795	0.4457	0.4414	0.4795	0.4457
0.2577	6.82	150	1.5469	0.5148	0.5941	0.5175	0.3284	0.3856	0.3335	0.3616	0.4268	0.3681	0.3616	0.4268	0.3681
0.1548	7.27	160	1.5986	0.5983	0.657	0.5862	0.4322	0.4877	0.4287	0.4466	0.5167	0.4482	0.4466	0.5167	0.4482
0.1535	7.73	170	1.5796	0.5609	0.641	0.5616	0.3856	0.4428	0.3892	0.4238	0.4921	0.4263	0.4238	0.4921	0.4263
0.1568	8.18	180	1.6052	0.5669	0.617	0.5679	0.3911	0.4382	0.3969	0.4363	0.4877	0.4417	0.4363	0.4877	0.4417
0.2038	8.64	190	1.6191	0.5466	0.5973	0.5313	0.3543	0.4114	0.3531	0.4061	0.4666	0.404	0.4061	0.4666	0.404
0.1808	9.09	200	1.6165	0.5751	0.5919	0.5587	0.3831	0.4097	0.3817	0.4482	0.4728	0.4405	0.4482	0.4728	0.4405
0.1021	9.55	210	1.6316	0.5316	0.6315	0.535	0.3588	0.4563	0.3697	0.405	0.502	0.4126	0.405	0.502	0.4126
0.1407	10.0	220	1.6322	0.6059	0.6233	0.5895	0.4192	0.4512	0.4176	0.4622	0.4946	0.4566	0.4622	0.4946	0.4566

Framework versions

Transformers 4.21.3
Pytorch 1.12.1+cu113
Datasets 1.2.1
Tokenizers 0.12.1

Library: transformers
Type: led
License: apache-2.0
Created: Nov 2, 2022
Updated: Nov 2, 2022