t-5-base-extractive-500

jgibb/t-5-base-extractive-500

t-5-base-extractive-500 is a machine learning model from jgibb. 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

18% of all-time downloads

All-time downloads

Public

Parameters

223M

5.4 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors892 MB · 100%

At a glance

Library: transformers
License: apache-2.0
Model type: t5
Access: Public
Created: Mar 30, 2024
Updated: Apr 1, 2024
SHA: 4da2ea30

Base models

Model card

From the Hugging Face model README

t-5-base-extractive-500

This model is a fine-tuned version of t5-base on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 1.2324
Rouge1: 0.6638
Rouge2: 0.3892
Rougel: 0.6005
Rougelsum: 0.6005
Wer: 0.5044
Bleurt: 0.3568

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: 2e-05
train_batch_size: 6
eval_batch_size: 6
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 2
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Wer	Bleurt
No log	0.13	250	1.3949	0.6375	0.3523	0.5692	0.5692	0.5415	0.216
1.8708	0.27	500	1.3361	0.6456	0.3636	0.5796	0.5796	0.5304	0.3009
1.8708	0.4	750	1.3101	0.6495	0.3699	0.5843	0.5842	0.5241	0.3009
1.411	0.53	1000	1.2892	0.6542	0.3755	0.5889	0.5889	0.5197	0.3881
1.411	0.66	1250	1.2786	0.6552	0.3774	0.5909	0.591	0.5165	0.4109
1.3604	0.8	1500	1.2670	0.6573	0.3802	0.5933	0.5934	0.5132	0.3568
1.3604	0.93	1750	1.2573	0.6583	0.3816	0.5948	0.5949	0.5117	0.3881
1.3566	1.06	2000	1.2507	0.6597	0.3837	0.596	0.5961	0.5094	0.3881
1.3566	1.2	2250	1.2462	0.6615	0.3858	0.5977	0.5977	0.5084	0.3568
1.3167	1.33	2500	1.2423	0.6623	0.3868	0.5986	0.5987	0.5073	0.3779
1.3167	1.46	2750	1.2382	0.6627	0.3874	0.5991	0.5992	0.5062	0.3779
1.329	1.6	3000	1.2362	0.6636	0.3882	0.5998	0.5998	0.506	0.3779
1.329	1.73	3250	1.2343	0.6634	0.3886	0.6001	0.6002	0.5054	0.3779
1.2989	1.86	3500	1.2325	0.6634	0.3889	0.6001	0.6001	0.5046	0.3568
1.2989	1.99	3750	1.2324	0.6638	0.3892	0.6005	0.6005	0.5044	0.3568

Framework versions

Transformers 4.38.2
Pytorch 2.2.1+cu121
Datasets 2.18.0
Tokenizers 0.15.2

Library: transformers
Type: t5
License: apache-2.0
Created: Mar 30, 2024
Updated: Apr 1, 2024