metadata

library_name: transformers
license: mit
base_model: xlnet-large-cased
tags:
  - generated_from_trainer
metrics:
  - f1
  - accuracy
model-index:
  - name: CS221-xlnet-large-cased-finetuned-augmentation
    results: []

CS221-xlnet-large-cased-finetuned-augmentation

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

Loss: 0.2505
F1: 0.9280
Roc Auc: 0.9437
Accuracy: 0.8749

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: 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
0.5536	1.0	360	0.5670	0.1475	0.5	0.1425
0.4339	2.0	720	0.4273	0.5310	0.6840	0.3391
0.3421	3.0	1080	0.3298	0.7397	0.8147	0.4927
0.2113	4.0	1440	0.2764	0.8083	0.8578	0.6338
0.1513	5.0	1800	0.2489	0.8476	0.8869	0.6810
0.1097	6.0	2160	0.2260	0.8777	0.9021	0.7665
0.0735	7.0	2520	0.2448	0.8996	0.9200	0.8026
0.0411	8.0	2880	0.2378	0.9129	0.9315	0.8214
0.0468	9.0	3240	0.2554	0.8990	0.9316	0.8235
0.0253	10.0	3600	0.2470	0.9084	0.9306	0.8464
0.021	11.0	3960	0.2412	0.9183	0.9388	0.8582
0.0125	12.0	4320	0.2440	0.9217	0.9383	0.8631
0.002	13.0	4680	0.2552	0.9232	0.9391	0.8673
0.0038	14.0	5040	0.2489	0.9237	0.9412	0.8666
0.0064	15.0	5400	0.2522	0.9234	0.9407	0.8659
0.0017	16.0	5760	0.2466	0.9254	0.9410	0.8728
0.0009	17.0	6120	0.2505	0.9280	0.9437	0.8749
0.0013	18.0	6480	0.2529	0.9272	0.9420	0.8749
0.002	19.0	6840	0.2530	0.9258	0.9412	0.8721
0.001	20.0	7200	0.2530	0.9258	0.9412	0.8721

Framework versions

Transformers 4.47.0
Pytorch 2.5.1+cu121
Datasets 3.2.0
Tokenizers 0.21.0