XLnet-cased-AS-HU-f1-score

This model is a fine-tuned version of xlnet-base-cased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.6569
• F1-score: 0.8203

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: 5e-05
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	F1-score
0.7105	1.0	64	0.6670	0.3697
0.654	2.0	128	0.6053	0.6119
0.509	3.0	192	0.4952	0.7471
0.3514	4.0	256	0.5171	0.7849
0.2233	5.0	320	0.6229	0.8059
0.1441	6.0	384	0.9291	0.8112
0.1376	7.0	448	1.4415	0.7421
0.0624	8.0	512	1.3527	0.8128
0.0224	9.0	576	1.4643	0.8103
0.0136	10.0	640	1.3725	0.8268
0.0177	11.0	704	1.5826	0.8142
0.0101	12.0	768	1.6241	0.8150
0.0003	13.0	832	1.6149	0.8165
0.0002	14.0	896	1.6481	0.8203
0.0001	15.0	960	1.6569	0.8203

Framework versions

• Transformers 4.41.1
• Pytorch 2.0.1+cu117
• Datasets 2.19.1
• Tokenizers 0.19.1