xlnet-large-cased-airlines-news-multi-label

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: 0.2329
• F1: 0.9001
• Roc Auc: 0.6501
• Hamming: 0.9145

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: 9e-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
• lr_scheduler_warmup_steps: 150
• num_epochs: 12

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Hamming
No log	1.0	128	0.2866	0.8594	0.5	0.9041
No log	2.0	256	0.2520	0.8955	0.5943	0.9130
No log	3.0	384	0.2431	0.8984	0.6493	0.9130
0.3656	4.0	512	0.2384	0.8984	0.6622	0.9115
0.3656	5.0	640	0.2329	0.9001	0.6501	0.9145
0.3656	6.0	768	0.2353	0.9000	0.6699	0.9130
0.3656	7.0	896	0.2336	0.8959	0.6735	0.9071
0.2788	8.0	1024	0.2318	0.8957	0.6606	0.9086
0.2788	9.0	1152	0.2327	0.8961	0.6606	0.9086
0.2788	10.0	1280	0.2317	0.8975	0.6545	0.9100
0.2788	11.0	1408	0.2311	0.8975	0.6545	0.9100
0.2659	12.0	1536	0.2309	0.8982	0.6554	0.9115

Framework versions

• Transformers 4.41.1
• Pytorch 2.3.0+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1