cs221-xlnet-large-cased-eng-pt

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: 0.5460
• F1: 0.7566
• Roc Auc: 0.8089
• Accuracy: 0.4828

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
0.5889	1.0	87	0.5873	0.1408	0.5	0.1207
0.4707	2.0	174	0.4159	0.6512	0.7304	0.3966
0.3648	3.0	261	0.3671	0.6550	0.7572	0.4483
0.2675	4.0	348	0.3692	0.7085	0.7779	0.4397
0.1929	5.0	435	0.3821	0.7077	0.7781	0.4483
0.1407	6.0	522	0.4573	0.7087	0.7753	0.4224
0.097	7.0	609	0.4498	0.7392	0.8005	0.4569
0.0603	8.0	696	0.4655	0.7396	0.8002	0.4483
0.0455	9.0	783	0.4833	0.7472	0.8075	0.4483
0.0277	10.0	870	0.5366	0.7338	0.7972	0.4655
0.0254	11.0	957	0.5452	0.7429	0.8051	0.4569
0.0138	12.0	1044	0.5668	0.7460	0.8062	0.4655
0.0128	13.0	1131	0.5460	0.7566	0.8089	0.4828
0.0072	14.0	1218	0.5875	0.7551	0.8117	0.4828
0.0058	15.0	1305	0.6071	0.7474	0.8038	0.4655
0.0064	16.0	1392	0.5952	0.7531	0.8120	0.4828
0.005	17.0	1479	0.5976	0.7468	0.8041	0.4655

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0