Modelpublic

shanhy/furina_clir_seed12_cross_translation_augmentation_0.1val

sourceHugging Faceupdated 3y agoView on Hugging Face

0likes7downloads

Model Card

xlmrobertaclirseed12crosstranslationaugmentation0.1val

This model is a fine-tuned version of yihongLiu/furina on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 0.0213
—Spearman Corr: 0.7695

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: 128
—seed: 12
—gradientaccumulationsteps: 2
—totaltrainbatch_size: 64
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 10
—mixedprecisiontraining: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Spearman Corr
No log	0.54	200	0.0414	0.5959
No log	1.08	400	0.0290	0.6668
No log	1.62	600	0.0296	0.6789
0.0482	2.15	800	0.0270	0.6991
0.0482	2.69	1000	0.0254	0.7192
0.0482	3.23	1200	0.0273	0.7202
0.0482	3.77	1400	0.0248	0.7340
0.0237	4.31	1600	0.0239	0.7367
0.0237	4.85	1800	0.0235	0.7400
0.0237	5.38	2000	0.0237	0.7417
0.0237	5.92	2200	0.0231	0.7474
0.0173	6.46	2400	0.0228	0.7547
0.0173	7.0	2600	0.0220	0.7609
0.0173	7.54	2800	0.0218	0.7634
0.0135	8.08	3000	0.0216	0.7640
0.0135	8.61	3200	0.0217	0.7654
0.0135	9.15	3400	0.0213	0.7692
0.0135	9.69	3600	0.0213	0.7695

Framework versions

—Transformers 4.37.2
—Pytorch 2.1.0+cu121
—Datasets 2.16.1
—Tokenizers 0.15.1