Modelpublic

Kuongan/xlm-roberta-base-tir-finetuned

sourceHugging Facemitupdated 2y agoView on Hugging Face

0likes5downloads

Model Card

xlm-roberta-base-tir-finetuned

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: 0.3553
—F1: 0.5424
—Roc Auc: 0.7266
—Accuracy: 0.4965

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: 32
—evalbatchsize: 32
—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	181	0.4097	0.0386	0.5095	0.2061
0.374	2.0	362	0.3871	0.1646	0.5518	0.2600
0.3627	3.0	543	0.3582	0.3093	0.6028	0.3762
0.3245	4.0	724	0.3214	0.4631	0.6751	0.4336
0.2892	5.0	905	0.3198	0.4708	0.6904	0.4426
0.2832	6.0	1086	0.3191	0.4916	0.6949	0.4647
0.2425	7.0	1267	0.3018	0.4955	0.6976	0.4862
0.2426	8.0	1448	0.3007	0.5166	0.7080	0.4834
0.22	9.0	1629	0.3114	0.5113	0.7096	0.4882
0.1964	10.0	1810	0.3267	0.5110	0.7165	0.4938
0.1813	11.0	1991	0.3154	0.5181	0.7129	0.4910
0.1649	12.0	2172	0.3378	0.5144	0.7131	0.4965
0.1602	13.0	2353	0.3440	0.5204	0.7190	0.4876
0.1363	14.0	2534	0.3413	0.5267	0.7240	0.4959
0.1326	15.0	2715	0.3469	0.5266	0.7165	0.4903
0.1383	16.0	2896	0.3476	0.5321	0.7203	0.4931
0.1222	17.0	3077	0.3486	0.5258	0.7173	0.4903
0.1242	18.0	3258	0.3553	0.5424	0.7266	0.4965
0.1179	19.0	3439	0.3542	0.5413	0.7245	0.4938
0.1187	20.0	3620	0.3548	0.5419	0.7254	0.4952

Framework versions

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