ja_classification_brl

This model is a fine-tuned version of dicta-il/BEREL_2.0 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0712
• Precision: 0.9846
• Recall: 0.9846
• F1: 0.9846
• Accuracy: 0.9846

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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	1125	0.0522	0.9819	0.9819	0.9819	0.9819
No log	2.0	2250	0.0490	0.9837	0.9837	0.9837	0.9837
No log	3.0	3375	0.0481	0.9843	0.9843	0.9843	0.9843
No log	4.0	4500	0.0514	0.9844	0.9844	0.9844	0.9844
No log	5.0	5625	0.0548	0.9848	0.9848	0.9848	0.9848
No log	6.0	6750	0.0587	0.9846	0.9846	0.9846	0.9846
No log	7.0	7875	0.0636	0.9844	0.9844	0.9844	0.9844
No log	8.0	9000	0.0669	0.9846	0.9846	0.9846	0.9846
No log	9.0	10125	0.0685	0.9844	0.9844	0.9844	0.9844
No log	10.0	11250	0.0712	0.9846	0.9846	0.9846	0.9846

Framework versions

• Transformers 4.28.1
• Pytorch 1.13.0+cu117
• Datasets 2.11.0
• Tokenizers 0.11.6