AugWordNet_BERT_FPB_finetuned_v1

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: 0.4204
• Accuracy: 0.9376
• F1: 0.9376
• Precision: 0.9377
• Recall: 0.9376

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: 0.0001
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1000
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.8417	1.0	91	0.7510	0.7410	0.7376	0.7449	0.7410
0.4682	2.0	182	0.3860	0.8534	0.8540	0.8585	0.8534
0.268	3.0	273	0.3011	0.8861	0.8855	0.8873	0.8861
0.1877	4.0	364	0.3048	0.8877	0.8877	0.8883	0.8877
0.1499	5.0	455	0.3375	0.8877	0.8876	0.8945	0.8877
0.1133	6.0	546	0.4436	0.8736	0.8723	0.8806	0.8736
0.1088	7.0	637	0.3466	0.8924	0.8924	0.8927	0.8924
0.0837	8.0	728	0.3562	0.9048	0.9050	0.9055	0.9048
0.0972	9.0	819	0.4039	0.9017	0.9012	0.9049	0.9017
0.0921	10.0	910	0.3287	0.9048	0.9047	0.9047	0.9048
0.0943	11.0	1001	0.4174	0.9095	0.9093	0.9128	0.9095
0.05	12.0	1092	0.3632	0.9095	0.9088	0.9126	0.9095
0.0562	13.0	1183	0.3304	0.9345	0.9343	0.9346	0.9345
0.0174	14.0	1274	0.3997	0.9173	0.9173	0.9173	0.9173
0.0224	15.0	1365	0.3719	0.9329	0.9330	0.9331	0.9329
0.0004	16.0	1456	0.4295	0.9345	0.9344	0.9345	0.9345
0.0103	17.0	1547	0.4351	0.9267	0.9266	0.9269	0.9267
0.0101	18.0	1638	0.3973	0.9360	0.9360	0.9361	0.9360
0.0003	19.0	1729	0.4204	0.9376	0.9376	0.9377	0.9376
0.0008	20.0	1820	0.4218	0.9329	0.9330	0.9331	0.9329

Framework versions

• Transformers 4.37.0
• Pytorch 2.1.2
• Datasets 2.1.0
• Tokenizers 0.15.1