electra-finetuned-ner-S800

This model is a fine-tuned version of google/electra-base-discriminator on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0697
• Precision: 0.6146
• Recall: 0.7181
• F1: 0.6624
• Accuracy: 0.9758

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	55	0.1115	0.4736	0.5161	0.4940	0.9552
No log	2.0	110	0.0765	0.5789	0.6690	0.6207	0.9721
No log	3.0	165	0.0711	0.5671	0.7055	0.6288	0.9730
No log	4.0	220	0.0698	0.6266	0.7083	0.6649	0.9753
No log	5.0	275	0.0697	0.6146	0.7181	0.6624	0.9758

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3