Modelpublic

marcussunderl0040/mt5-small-finetuned-persian-news-summarization-with-additional-hyperparameters

sourceHugging Faceapache-2.0updated 3d agoView on Hugging Face

0likes98downloads

Model Card

mt5-small-finetuned-persian-news-summarization-with-additional-hyperparameters

This model is a fine-tuned version of google/mt5-small on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 1.1164
—Rouge1 F1: 56.7956
—Rouge1 Precision: 57.5473
—Rouge1 Recall: 58.6205
—Rouge2 F1: 40.1594
—Rouge2 Precision: 40.4966
—Rouge2 Recall: 41.7103
—Rougel F1: 49.1004
—Rougel Precision: 49.6461
—Rougel Recall: 50.8065

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: 4e-05
—trainbatchsize: 8
—evalbatchsize: 8
—seed: 42
—optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
—lrschedulertype: cosine
—lrschedulerwarmup_ratio: 0.1
—num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1 F1	Rouge1 Precision	Rouge1 Recall	Rouge2 F1	Rouge2 Precision	Rouge2 Recall	Rougel F1	Rougel Precision	Rougel Recall
1.7072	1.0	1205	1.2617	51.8527	61.6615	48.1083	37.2937	44.4226	34.639	45.3777	53.9129	42.1657
1.4721	2.0	2410	1.1533	55.6793	58.3728	56.0708	39.5451	41.3271	40.0375	48.2426	50.4432	48.7276
1.4029	3.0	3615	1.1242	56.6985	56.856	59.0696	39.961	39.8776	41.9066	48.8833	48.9108	51.0644
1.4023	4.0	4820	1.1164	56.7956	57.5473	58.6205	40.1594	40.4966	41.7103	49.1004	49.6461	50.8065

Framework versions

—Transformers 4.57.6
—Pytorch 2.11.0+cu128
—Datasets 4.8.5
—Tokenizers 0.22.2