new_camembert_jb

This model is a fine-tuned version of Jean-Baptiste/camembert-ner on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0752
• Overall Precision: 0.8202
• Overall Recall: 0.8595
• Overall F1: 0.8394
• Overall Accuracy: 0.9814
• Er F1: 0.8430
• Oc F1: 0.8418
• Umanprod F1: 0.6933

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: linear
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	Overall Precision	Overall Recall	Overall F1	Overall Accuracy	Er F1	Oc F1	Umanprod F1
0.2682	1.0	613	0.0813	0.7550	0.8071	0.7802	0.9749	0.7920	0.7709	0.6667
0.0717	2.0	1226	0.0706	0.8139	0.8411	0.8273	0.9808	0.8446	0.8126	0.6857
0.0524	3.0	1839	0.0723	0.8215	0.8567	0.8387	0.9812	0.8462	0.8346	0.7368
0.0372	4.0	2452	0.0752	0.8202	0.8595	0.8394	0.9814	0.8430	0.8418	0.6933

Framework versions

• Transformers 4.25.1
• Pytorch 1.13.0+cpu
• Datasets 2.7.1
• Tokenizers 0.13.2