my_awesome_wnut_model_

This model is a fine-tuned version of cointegrated/rubert-tiny2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1997
• Precision: 0.3076
• Recall: 0.4690
• F1: 0.3716
• Accuracy: 0.9322

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: 7
• eval_batch_size: 7
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 24

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.621	2.41	140	0.4025	0.0	0.0	0.0	0.9050
0.3224	4.83	280	0.2750	0.2036	0.2074	0.2055	0.9118
0.2421	7.24	420	0.2326	0.2706	0.3406	0.3016	0.9220
0.2061	9.66	560	0.2146	0.2968	0.4102	0.3444	0.9269
0.1779	12.07	700	0.2037	0.3125	0.4257	0.3604	0.9306
0.1606	14.48	840	0.2042	0.3044	0.4613	0.3668	0.9298
0.1544	16.9	980	0.2001	0.3101	0.4690	0.3734	0.9310
0.1402	19.31	1120	0.1991	0.3130	0.4690	0.3755	0.9316
0.139	21.72	1260	0.1997	0.3076	0.4690	0.3716	0.9322

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.19.0
• Tokenizers 0.15.2