kaz_legal_distilbert_full_corpus_10.0_16

This model is a fine-tuned version of kaisar-barlybay-sse/kaz_legal_distilbert_full_corpus_10.0 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 4.0806
• Accuracy: 0.4351

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: 5e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 16

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.4013	1.0	501	1.3849	0.3473
1.3954	2.0	1002	1.3853	0.3234
1.3874	3.0	1503	1.3864	0.2216
1.3995	4.0	2004	1.3863	0.3513
1.3944	5.0	2505	1.3862	0.3473
1.3829	6.0	3006	1.3858	0.3713
1.2411	7.0	3507	1.4126	0.3972
0.9506	8.0	4008	1.7895	0.3992
0.7435	9.0	4509	1.5649	0.4212
0.5988	10.0	5010	2.6088	0.4371
0.4972	11.0	5511	2.8238	0.4371
0.4111	12.0	6012	3.2902	0.4431
0.3471	13.0	6513	3.5568	0.4212
0.3065	14.0	7014	3.9031	0.4351
0.3015	15.0	7515	4.0874	0.4311
0.2873	16.0	8016	4.0806	0.4351

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.0+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3