Modelpublic

kidusabe/distilbert-base-multilingual-cased-language_detection

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face

0likes6downloads

Model Card

distilbert-base-multilingual-cased-language_detection

This model is a fine-tuned version of distilbert-base-multilingual-cased on the None dataset. It achieves the following results on the evaluation set:

—Loss: 0.0544
—Accuracy: 0.9912
—Weighted f1: 0.9912
—Micro f1: 0.9912
—Macro f1: 0.9900
—Weighted recall: 0.9912
—Micro recall: 0.9912
—Macro recall: 0.9904
—Weighted precision: 0.9914
—Micro precision: 0.9912
—Macro precision: 0.9899

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: 64
—evalbatchsize: 64
—seed: 42
—optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
—lrschedulertype: linear
—num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Weighted f1	Micro f1	Macro f1	Weighted recall	Micro recall	Macro recall	Weighted precision	Micro precision	Macro precision
0.943	1.0	128	0.1361	0.9863	0.9863	0.9863	0.9845	0.9863	0.9863	0.9847	0.9868	0.9863	0.9849
0.0991	2.0	256	0.0640	0.9892	0.9893	0.9892	0.9880	0.9892	0.9892	0.9889	0.9896	0.9892	0.9874
0.056	3.0	384	0.0544	0.9912	0.9912	0.9912	0.9900	0.9912	0.9912	0.9904	0.9914	0.9912	0.9899

Framework versions

—Transformers 4.48.3
—Pytorch 2.5.1+cu124
—Datasets 3.3.2
—Tokenizers 0.21.0