finetuned_bert-base-uncased

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: 1.1947
• Accuracy: 0.6793

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
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.2761	1.0	102	1.3225	0.3375
0.9847	2.0	204	1.0792	0.5509
0.6882	3.0	306	0.9260	0.6382
0.5099	4.0	408	0.9072	0.6634
0.4614	5.0	510	0.9115	0.6867
0.3406	6.0	612	1.0022	0.6751
0.189	7.0	714	1.0881	0.6751
0.2179	8.0	816	1.1520	0.6712
0.2085	9.0	918	1.2567	0.6896
0.1914	10.0	1020	1.2074	0.6828
0.1271	11.0	1122	1.3389	0.6887
0.1236	12.0	1224	1.3539	0.6790
0.0946	13.0	1326	1.4042	0.6838
0.0968	14.0	1428	1.4079	0.6877
0.1095	15.0	1530	1.4884	0.6799
0.1102	16.0	1632	1.5244	0.6790
0.1159	17.0	1734	1.5238	0.6799
0.1448	18.0	1836	1.5568	0.6780
0.1105	19.0	1938	1.5629	0.6780
0.092	20.0	2040	1.5588	0.6809

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2