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: 1.2312

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: 128
• eval_batch_size: 32
• 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
2.3399	1.0	73	1.7462
1.799	2.0	146	1.4703
1.6353	3.0	219	1.4796
1.5464	4.0	292	1.3851
1.4697	5.0	365	1.3032
1.4146	6.0	438	1.3339
1.3677	7.0	511	1.3349
1.3345	8.0	584	1.2818
1.3053	9.0	657	1.2646
1.2886	10.0	730	1.2355
1.278	11.0	803	1.3037
1.2568	12.0	876	1.1511
1.2399	13.0	949	1.2578
1.2369	14.0	1022	1.2487
1.2165	15.0	1095	1.2581
1.2289	16.0	1168	1.2312

Framework versions

• Transformers 4.21.3
• Pytorch 1.11.0+cu113
• Datasets 2.7.1
• Tokenizers 0.12.1