Modelpublic

Kuongan/CS221-xlm-roberta-base-finetuned-tir

sourceHugging Facemitupdated 2y agoView on Hugging Face

0likes5downloads

Model Card

CS221-xlm-roberta-base-finetuned-tir

This model is a fine-tuned version of xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:

—Loss: 0.3832
—F1: 0.4195
—Roc Auc: 0.6710
—Accuracy: 0.4739

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
—trainbatchsize: 8
—evalbatchsize: 8
—seed: 42
—optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
—lrschedulertype: cosine
—lrschedulerwarmup_steps: 100
—num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy
0.4043	1.0	461	0.3929	0.0	0.5	0.2182
0.3534	2.0	922	0.3772	0.0706	0.5158	0.2899
0.3301	3.0	1383	0.3425	0.2158	0.5857	0.3811
0.3369	4.0	1844	0.3353	0.2500	0.5934	0.4088
0.2964	5.0	2305	0.3293	0.3116	0.6149	0.4267
0.28	6.0	2766	0.3518	0.2809	0.6173	0.3844
0.302	7.0	3227	0.3366	0.3366	0.6327	0.4300
0.2706	8.0	3688	0.3400	0.3423	0.6323	0.4414
0.2359	9.0	4149	0.3594	0.3998	0.6546	0.4544
0.2739	10.0	4610	0.3811	0.3900	0.6573	0.4463
0.2442	11.0	5071	0.3832	0.4195	0.6710	0.4739
0.1307	12.0	5532	0.3986	0.3945	0.6582	0.4723
0.1754	13.0	5993	0.3965	0.4135	0.6673	0.4642
0.1576	14.0	6454	0.4098	0.4182	0.6655	0.4446

Framework versions

—Transformers 4.47.1
—Pytorch 2.5.1+cu121
—Datasets 3.2.0
—Tokenizers 0.21.0