camembert-base-NER-favsbot

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

• Loss: 1.7433
• Precision: 0.6
• Recall: 0.0121
• F1: 0.0238
• Accuracy: 0.4208

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: 1.5e-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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	4	2.2915	0.1364	0.1215	0.1285	0.3475
No log	2.0	8	2.2230	0.2909	0.0648	0.1060	0.4395
No log	3.0	12	2.1573	0.4545	0.0202	0.0388	0.4225
No log	4.0	16	2.0961	0.0	0.0	0.0	0.4123
No log	5.0	20	2.0426	0.0	0.0	0.0	0.4123
No log	6.0	24	1.9965	0.0	0.0	0.0	0.4123
No log	7.0	28	1.9575	0.0	0.0	0.0	0.4123
No log	8.0	32	1.9233	0.0	0.0	0.0	0.4123
No log	9.0	36	1.8933	0.0	0.0	0.0	0.4123
No log	10.0	40	1.8674	0.0	0.0	0.0	0.4123
No log	11.0	44	1.8441	0.0	0.0	0.0	0.4123
No log	12.0	48	1.8240	0.0	0.0	0.0	0.4123
No log	13.0	52	1.8060	1.0	0.0040	0.0081	0.4140
No log	14.0	56	1.7899	1.0	0.0040	0.0081	0.4140
No log	15.0	60	1.7762	1.0	0.0040	0.0081	0.4140
No log	16.0	64	1.7647	0.5	0.0040	0.0080	0.4157
No log	17.0	68	1.7556	0.5	0.0040	0.0080	0.4157
No log	18.0	72	1.7490	0.6667	0.0081	0.016	0.4174
No log	19.0	76	1.7449	0.75	0.0121	0.0239	0.4191
No log	20.0	80	1.7433	0.6	0.0121	0.0238	0.4208

Framework versions

• Transformers 4.21.1
• Pytorch 1.12.1
• Datasets 2.4.0
• Tokenizers 0.12.1