bert-base-uncased-finetuned-classification_TokenNew

This model is a fine-tuned version of bert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 41.4553
• Mse: 41.4553
• Mae: 4.7280
• R2: 0.7658
• Accuracy: 0.1600
• Msev: 0.0241

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: 1e-05
• train_batch_size: 16
• eval_batch_size: 16
• 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	Mse	Mae	R2	Accuracy	Msev
10.3178	1.0	5215	41.8852	41.8852	4.6705	0.7634	0.1827	0.0239
3.6731	2.0	10430	45.6101	45.6101	4.9092	0.7423	0.1809	0.0219
2.0891	3.0	15645	42.1319	42.1319	4.7640	0.7620	0.1525	0.0237
1.5213	4.0	20860	42.0646	42.0646	4.7562	0.7623	0.1588	0.0238
1.1904	5.0	26075	42.0155	42.0155	4.7778	0.7626	0.1563	0.0238
1.0127	6.0	31290	41.6389	41.6389	4.7342	0.7647	0.1660	0.0240
0.9218	7.0	36505	40.9860	40.9860	4.7009	0.7684	0.1589	0.0244
0.7466	8.0	41720	40.1809	40.1809	4.6686	0.7730	0.1629	0.0249
0.7264	9.0	46935	40.9795	40.9795	4.7043	0.7685	0.1616	0.0244
0.6968	10.0	52150	41.4553	41.4553	4.7280	0.7658	0.1600	0.0241

Framework versions

• Transformers 4.21.3
• Pytorch 1.12.1+cu113
• Datasets 2.4.0
• Tokenizers 0.12.1