Modelpublic

sercetexam9/cs221-afro-xlmr-large-76L-pcm-finetuned-10-epochs

sourceHugging Facemitupdated 2y agoView on Hugging Face

0likes10downloads

Model Card

cs221-afro-xlmr-large-76L-pcm-finetuned-10-epochs

This model is a fine-tuned version of Davlan/afro-xlmr-large-76L on the None dataset. It achieves the following results on the evaluation set:

—Loss: 0.3586
—F1: 0.6422
—Roc Auc: 0.7578
—Accuracy: 0.3861

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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy
0.4949	1.0	94	0.4847	0.4327	0.6336	0.2185
0.4129	2.0	188	0.3764	0.5659	0.7068	0.2949
0.3874	3.0	282	0.3892	0.6153	0.7428	0.3257
0.3357	4.0	376	0.3524	0.6229	0.7445	0.3512
0.2843	5.0	470	0.3586	0.6422	0.7578	0.3861
0.2229	6.0	564	0.3846	0.6287	0.7544	0.3445
0.2028	7.0	658	0.3966	0.6240	0.7484	0.3592
0.1728	8.0	752	0.4041	0.6316	0.7539	0.3525

Framework versions

—Transformers 4.48.0
—Pytorch 2.5.1+cu121
—Datasets 3.2.0
—Tokenizers 0.21.0