Modelpublic

toreleon/synthetic-40-vsfc-xlm-r

sourceHugging Facemitupdated 4y agoView on Hugging Face

0likes5downloads

Model Card

synthetic-40-vsfc-xlm-r

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

—Loss: 1.3921
—Precision: 0.8929
—Recall: 0.4782
—F1 Weighted: 0.5759
—F1 Macro: 0.4410

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: 64
—evalbatchsize: 128
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—lrschedulerwarmup_steps: 100
—num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1 Weighted	F1 Macro
1.0989	0.39	25	1.0971	0.6042	0.3430	0.3150	0.2438
0.8219	0.78	50	0.8525	0.7859	0.7682	0.7577	0.5881
0.5421	1.17	75	1.2589	0.8433	0.5047	0.6097	0.4640
0.456	1.56	100	1.2742	0.8534	0.5111	0.6086	0.4656
0.4956	1.95	125	1.4465	0.8210	0.4195	0.4998	0.3903
0.3392	2.34	150	1.1332	0.8779	0.5111	0.6126	0.4699
0.3776	2.73	175	1.3973	0.8858	0.5199	0.6248	0.4729
0.3287	3.12	200	1.4185	0.8428	0.5022	0.5984	0.4556
0.3325	3.52	225	1.3921	0.8929	0.4782	0.5759	0.4410

Framework versions

—Transformers 4.27.1
—Pytorch 1.13.1+cu116
—Datasets 2.10.1
—Tokenizers 0.13.2