aces-roberta-13

gijs/aces-roberta-13

aces-roberta-13 is a token classification model from gijs. Use it when you need labels on individual words, such as names. It is set up for transformers. The card lists the license as mit.

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

Downloads · 30 days

0% of all-time downloads

All-time downloads

447K

Public

Repo size

2.8 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin1.4 GB · 100%

At a glance

Task: Token Classification
Library: transformers
License: mit
Model type: roberta
Access: Public
Created: Mar 9, 2023
Updated: Mar 9, 2023
SHA: b62e0364

Try a prompt

Model card

From the Hugging Face model README

aces-roberta-13

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

Loss: 0.4600
Precision: 0.8364
Recall: 0.8452
F1: 0.8383
Accuracy: 0.8452
F1 Who: 0.9189
F1 What: 0.8621
F1 Where: 0.9231
F1 How: 0.9141

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: 1e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy	F1 Who	F1 What	F1 Where	F1 How
1.9849	0.35	20	1.4123	0.5426	0.6351	0.5494	0.6351	0.1026	0.6222	0.3232	0.7857
1.2159	0.7	40	0.9450	0.6559	0.7188	0.6592	0.7188	0.6780	0.7539	0.7071	0.7882
0.8634	1.05	60	0.6885	0.7652	0.7994	0.7725	0.7994	0.9067	0.8152	0.8070	0.8940
0.6777	1.4	80	0.6144	0.7650	0.7946	0.7711	0.7946	0.9189	0.7876	0.8039	0.9085
0.6051	1.75	100	0.5485	0.8126	0.8278	0.8150	0.8278	0.9315	0.8362	0.8148	0.9241
0.5511	2.11	120	0.5264	0.8113	0.8167	0.8036	0.8167	0.9315	0.8444	0.8257	0.9199
0.486	2.46	140	0.4867	0.8230	0.8357	0.8248	0.8357	0.9315	0.8539	0.9091	0.9048
0.4813	2.81	160	0.4767	0.8285	0.8278	0.8213	0.8278	0.9189	0.8701	0.9076	0.9135
0.4494	3.16	180	0.5042	0.8152	0.8199	0.8126	0.8199	0.9315	0.8427	0.8333	0.8956
0.4018	3.51	200	0.4802	0.8248	0.8357	0.8249	0.8357	0.9189	0.8736	0.8780	0.9357
0.4205	3.86	220	0.4723	0.8340	0.8389	0.8346	0.8389	0.9189	0.8636	0.9138	0.8986
0.3535	4.21	240	0.4669	0.8324	0.8452	0.8364	0.8452	0.9189	0.8571	0.9138	0.9167
0.3808	4.56	260	0.4585	0.8349	0.8452	0.8383	0.8452	0.9189	0.8621	0.9231	0.9141
0.3491	4.91	280	0.4600	0.8364	0.8452	0.8383	0.8452	0.9189	0.8621	0.9231	0.9141

Framework versions

Transformers 4.26.0
Pytorch 1.13.1+cu117
Datasets 2.8.0
Tokenizers 0.13.2

Task: Token Classification
Library: transformers
Type: roberta
License: mit
Created: Mar 9, 2023
Updated: Mar 9, 2023