xlnet-large-cased-finetuned-augmentation-LUNAR-TAPT-DAIR-finetuned-augmentation-LUNAR-TAPT-macro

This model is a fine-tuned version of sercetexam9/xlnet-large-cased-finetuned-augmentation-LUNAR-TAPT-DAIR on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2222
• F1: 0.8939
• Roc Auc: 0.9254
• Accuracy: 0.7922

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: cosine
• lr_scheduler_warmup_steps: 100
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy
0.2521	1.0	421	0.2089	0.8618	0.8955	0.7536
0.2487	2.0	842	0.2215	0.8522	0.8907	0.7280
0.2449	3.0	1263	0.2067	0.8608	0.8946	0.7488
0.1682	4.0	1684	0.2217	0.8673	0.8975	0.7607
0.1869	5.0	2105	0.2144	0.8664	0.8976	0.7601
0.1113	6.0	2526	0.2268	0.8638	0.8940	0.7637
0.1099	7.0	2947	0.2123	0.8865	0.9179	0.7743
0.1322	8.0	3368	0.2178	0.8757	0.9088	0.7660
0.1067	9.0	3789	0.2225	0.8854	0.9228	0.7690
0.0845	10.0	4210	0.2129	0.8893	0.9172	0.7862
0.0792	11.0	4631	0.2190	0.8850	0.9170	0.7732
0.0762	12.0	5052	0.2164	0.8932	0.9238	0.7856
0.0805	13.0	5473	0.2230	0.8860	0.9162	0.7850
0.0718	14.0	5894	0.2222	0.8939	0.9254	0.7922
0.0696	15.0	6315	0.2280	0.8885	0.9202	0.7821
0.057	16.0	6736	0.2252	0.8900	0.9191	0.7922
0.0694	17.0	7157	0.2235	0.8935	0.9221	0.7945
0.0492	18.0	7578	0.2269	0.8891	0.9208	0.7904

Framework versions

• Transformers 4.45.1
• Pytorch 2.4.0
• Datasets 3.0.1
• Tokenizers 0.20.0