BERT-finetuned-ner-pablo-classifier-then-full

This model is a fine-tuned version of pabRomero/BERT-finetuned-ner-pablo-just-classifier on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1753
• Precision: 0.7975
• Recall: 0.8083
• F1: 0.8028
• Accuracy: 0.9743

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: 0.0002
• train_batch_size: 512
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.05
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	72	0.1020	0.7241	0.7862	0.7539	0.9711
No log	2.0	144	0.0966	0.7538	0.7947	0.7737	0.9739
No log	3.0	216	0.0890	0.7975	0.7882	0.7928	0.9755
No log	4.0	288	0.1002	0.7868	0.7950	0.7909	0.9744
No log	5.0	360	0.1099	0.7936	0.8034	0.7985	0.9750
No log	6.0	432	0.1380	0.8018	0.8021	0.8019	0.9740
0.0554	7.0	504	0.1397	0.7974	0.7983	0.7978	0.9743
0.0554	8.0	576	0.1558	0.7944	0.8061	0.8002	0.9740
0.0554	9.0	648	0.1681	0.7987	0.8076	0.8031	0.9744
0.0554	10.0	720	0.1753	0.7975	0.8083	0.8028	0.9743

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1