Modelpublic

Shijia/furina_seed42_eng_amh_esp_latin_2e-05

sourceHugging Faceupdated 3y agoView on Hugging Face

0likes10downloads

Model Card

furinaseed42engamhesplatin2e-05

This model is a fine-tuned version of yihongLiu/furina on the None dataset. It achieves the following results on the evaluation set:

—Loss: 0.0179
—Spearman Corr: 0.7569

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: 42
—gradientaccumulationsteps: 2
—totaltrainbatch_size: 64
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 30
—mixedprecisiontraining: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Spearman Corr
No log	1.59	200	0.0178	0.7021
0.069	3.17	400	0.0168	0.7397
0.0224	4.76	600	0.0164	0.7463
0.0172	6.35	800	0.0162	0.7618
0.0172	7.94	1000	0.0156	0.7704
0.0133	9.52	1200	0.0156	0.7684
0.011	11.11	1400	0.0157	0.7674
0.0094	12.7	1600	0.0166	0.7665
0.0079	14.29	1800	0.0181	0.7536
0.0079	15.87	2000	0.0185	0.7513
0.007	17.46	2200	0.0182	0.7550
0.0061	19.05	2400	0.0178	0.7520
0.0056	20.63	2600	0.0179	0.7569

Framework versions

—Transformers 4.37.2
—Pytorch 2.1.0+cu121
—Datasets 2.17.0
—Tokenizers 0.15.2