Full-8epoch-BERT-base-multilingual-finetuned-CEFR_ner-60000news – AI Model by DioBot2000 | AlphaNeural AI

You can deploy this model and start earning money today!

DioBot2000/
Full-8epoch-BERT-base-multilingual-finetuned-CEFR_ner-60000news

like

0

token-classification

generated_from_trainer

google-bert/bert-base-multilingual-cased

autotrain_compatible

endpoints_compatible

Views

No views yet

Full-8epoch-BERT-base-multilingual-finetuned-CEFR_ner-60000news

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

Loss: 0.0772
Accuracy: 0.3055
Precision: 0.5539
Recall: 0.8444
F1: 0.5463

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: 32
eval_batch_size: 32
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 8
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.1609	1.0	1563	0.1237	0.3003	0.5394	0.8062	0.5257
0.1031	2.0	3126	0.0926	0.3033	0.5628	0.8280	0.5475
0.0805	3.0	4689	0.0831	0.3043	0.5396	0.8360	0.5317
0.0687	4.0	6252	0.0789	0.3048	0.5514	0.8404	0.5428
0.0595	5.0	7815	0.0767	0.3051	0.5386	0.8415	0.5347
0.0529	6.0	9378	0.0770	0.3053	0.5506	0.8425	0.5444
0.048	7.0	10941	0.0765	0.3055	0.5516	0.8444	0.5451
0.0453	8.0	12504	0.0772	0.3055	0.5539	0.8444	0.5463

Framework versions

Transformers 4.41.2
Pytorch 2.2.1
Datasets 2.19.2
Tokenizers 0.19.1