my_awesome_wnut_model

This model is a fine-tuned version of distilbert-base-uncased on the wnut_17 dataset. It achieves the following results on the evaluation set:

• Loss: 0.4068
• Precision: 0.5525
• Recall: 0.4050
• F1: 0.4674
• Accuracy: 0.9465

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: 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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	213	0.2742	0.5216	0.3355	0.4083	0.9421
No log	2.0	426	0.2810	0.6107	0.3503	0.4452	0.9455
0.0744	3.0	639	0.3305	0.6560	0.3411	0.4488	0.9456
0.0744	4.0	852	0.3382	0.5480	0.3596	0.4342	0.9443
0.0295	5.0	1065	0.3461	0.5635	0.3865	0.4585	0.9454
0.0295	6.0	1278	0.3823	0.5744	0.3828	0.4594	0.9454
0.0295	7.0	1491	0.3404	0.5080	0.4096	0.4536	0.9445
0.0128	8.0	1704	0.3926	0.5302	0.3744	0.4389	0.9441
0.0128	9.0	1917	0.3505	0.5033	0.4226	0.4594	0.9449
0.0071	10.0	2130	0.3825	0.5685	0.3846	0.4588	0.9456
0.0071	11.0	2343	0.3806	0.5155	0.4171	0.4611	0.9451
0.0044	12.0	2556	0.4035	0.5422	0.3985	0.4594	0.9454
0.0044	13.0	2769	0.4106	0.5940	0.3865	0.4683	0.9465
0.0044	14.0	2982	0.4069	0.5485	0.4032	0.4647	0.9457
0.0032	15.0	3195	0.4280	0.6029	0.3800	0.4662	0.9466
0.0032	16.0	3408	0.4049	0.5798	0.4208	0.4876	0.9472
0.0026	17.0	3621	0.4129	0.5758	0.4013	0.4730	0.9470
0.0026	18.0	3834	0.4131	0.5731	0.4069	0.4759	0.9469
0.0021	19.0	4047	0.4074	0.5557	0.4022	0.4667	0.9465
0.0021	20.0	4260	0.4068	0.5525	0.4050	0.4674	0.9465

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.4
• Tokenizers 0.13.3