bert-base-uncased-issues-128

This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.0940

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: 32
• eval_batch_size: 8
• 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
2.1003	1.0	291	1.6578
1.6211	2.0	582	1.4140
1.4964	3.0	873	1.3040
1.41	4.0	1164	1.3011
1.336	5.0	1455	1.3095
1.2862	6.0	1746	1.3739
1.2271	7.0	2037	1.2743
1.2043	8.0	2328	1.2019
1.1701	9.0	2619	1.2696
1.1498	10.0	2910	1.2507
1.1194	11.0	3201	1.1398
1.1094	12.0	3492	1.1309
1.0913	13.0	3783	1.0740
1.0683	14.0	4074	1.1201
1.0607	15.0	4365	1.1690
1.0558	16.0	4656	1.0940

Framework versions

• Transformers 4.21.2
• Pytorch 1.13.0+cu117
• Datasets 2.7.1
• Tokenizers 0.12.1