Modelpublic

haryoaw/scenario-TCR-data-AmazonScience-massive-all_1.1-model-xlm-roberta-base

sourceHugging Facemitupdated 3y agoView on Hugging Face

0likes12downloads

Model Card

scenario-TCR-data-AmazonScience-massive-all_1.1-model-xlm-roberta-base

This model is a fine-tuned version of xlm-roberta-base on the massive dataset. It achieves the following results on the evaluation set:

—Loss: 0.8546
—Accuracy: 0.8443
—F1: 0.8187

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: 5e-05
—trainbatchsize: 32
—evalbatchsize: 32
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 500

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.6138	0.27	5000	0.6955	0.8197	0.7723
0.4705	0.53	10000	0.6878	0.8280	0.7855
0.4019	0.8	15000	0.6850	0.8368	0.8005
0.2929	1.07	20000	0.7296	0.8316	0.7965
0.3011	1.34	25000	0.7140	0.8426	0.8139
0.2921	1.6	30000	0.7252	0.8418	0.8152
0.2799	1.87	35000	0.7186	0.8431	0.8152
0.2103	2.14	40000	0.7557	0.8462	0.8172
0.2251	2.41	45000	0.7926	0.8411	0.8095
0.2118	2.67	50000	0.7915	0.8427	0.8126
0.2239	2.94	55000	0.7813	0.8416	0.8076
0.1727	3.21	60000	0.8273	0.8471	0.8224
0.1785	3.47	65000	0.8192	0.8447	0.8149
0.2008	3.74	70000	0.8043	0.8464	0.8213
0.1773	4.01	75000	0.8555	0.8407	0.8121
0.165	4.28	80000	0.8556	0.8456	0.8218
0.1658	4.54	85000	0.8546	0.8443	0.8187

Framework versions

—Transformers 4.36.2
—Pytorch 2.1.2
—Datasets 2.16.0
—Tokenizers 0.15.0