cs221-xlnet-large-cased-eng-finetuned-20-epochs-tapt

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

• Loss: 0.5355
• F1: 0.7594
• Roc Auc: 0.8228
• Accuracy: 0.4506

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: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 100
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy
0.6009	1.0	73	0.5947	0.436	0.6134	0.1369
0.5678	2.0	146	0.5712	0.4273	0.6124	0.1456
0.5116	3.0	219	0.4825	0.6053	0.7128	0.2877
0.4375	4.0	292	0.4319	0.6749	0.7600	0.3501
0.3723	5.0	365	0.3959	0.7178	0.7919	0.4073
0.2906	6.0	438	0.3896	0.7329	0.8001	0.4385
0.2457	7.0	511	0.4272	0.7291	0.8009	0.4125
0.2035	8.0	584	0.4255	0.7516	0.8198	0.4385
0.1546	9.0	657	0.4514	0.7462	0.8126	0.4402
0.1238	10.0	730	0.4732	0.7518	0.8155	0.4419
0.0896	11.0	803	0.5441	0.7437	0.8104	0.4506
0.0659	12.0	876	0.5355	0.7594	0.8228	0.4506
0.0459	13.0	949	0.5450	0.7582	0.8200	0.4714
0.0436	14.0	1022	0.5919	0.7525	0.8164	0.4558
0.0366	15.0	1095	0.6016	0.7588	0.8226	0.4541

Framework versions

• Transformers 4.48.0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0