bert-small-finetuned-finetuned

This model is a fine-tuned version of ryantaw/bert-small-finetuned on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.0767
• Accuracy: 0.6119
• F1 Score: 0.6156

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: 86
• eval_batch_size: 86
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Score
0.7125	1.0	18	1.0136	0.6011	0.5997
0.604	2.0	36	1.0198	0.6038	0.6058
0.5421	3.0	54	1.0517	0.6065	0.6068
0.4724	4.0	72	1.0767	0.6119	0.6156
0.42	5.0	90	1.1184	0.5768	0.5751
0.3823	6.0	108	1.1217	0.5876	0.5881
0.3312	7.0	126	1.1425	0.6065	0.6053
0.3045	8.0	144	1.1760	0.6065	0.6095
0.2662	9.0	162	1.2044	0.6065	0.6090
0.2403	10.0	180	1.2143	0.6011	0.6011
0.2308	11.0	198	1.2394	0.5903	0.5927
0.2053	12.0	216	1.2589	0.6038	0.6068
0.1808	13.0	234	1.2895	0.6065	0.6071
0.1599	14.0	252	1.3144	0.6065	0.6086
0.1497	15.0	270	1.3386	0.5930	0.5951
0.1383	16.0	288	1.3608	0.5903	0.5931
0.1321	17.0	306	1.3624	0.5876	0.5888
0.1183	18.0	324	1.3810	0.5930	0.5945
0.1196	19.0	342	1.3827	0.5903	0.5927
0.1181	20.0	360	1.3805	0.5903	0.5920

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1