microsoft-deberta-v3-large_ner_conll2003

Gladiator/microsoft-deberta-v3-large_ner_conll2003

microsoft-deberta-v3-large_ner_conll2003 is a token classification model from Gladiator. 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

144

0% of all-time downloads

All-time downloads

48.9K

Public

Parameters

434M

12.2 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin1.7 GB · 50%

Parameter types

How the weights are stored.

F32434M · 100%

Try a prompt

Datasets

conll2003

Model card

From the Hugging Face model README

microsoft-deberta-v3-large_ner_conll2003

This model is a fine-tuned version of microsoft/deberta-v3-large on the conll2003 dataset. It achieves the following results on the evaluation set:

Loss: 0.0293
Precision: 0.9667
Recall: 0.9724
F1: 0.9695
Accuracy: 0.9945

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: 16
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.0986	1.0	878	0.0323	0.9453	0.9596	0.9524	0.9921
0.0212	2.0	1756	0.0270	0.9571	0.9675	0.9623	0.9932
0.009	3.0	2634	0.0280	0.9638	0.9714	0.9676	0.9940
0.0035	4.0	3512	0.0290	0.9657	0.9712	0.9685	0.9943
0.0022	5.0	4390	0.0293	0.9667	0.9724	0.9695	0.9945

Framework versions

Transformers 4.20.1
Pytorch 1.11.0
Datasets 2.1.0
Tokenizers 0.12.1

Task: Token Classification
Library: transformers
Type: deberta-v2
License: mit
Created: Dec 9, 2022
Updated: Jun 18, 2023