Modelpublic

Kankanaghosh/Scientific_data_summarisation_model

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face

0likes4downloads

Model Card

Scientificdatasummarisation_model

This model is a fine-tuned version of sambydlo/bart-large-scientific-lay-summarisation on the scientificlaysummarisation dataset. It achieves the following results on the evaluation set:

—Loss: 2.0270
—Rouge1: 0.1355
—Rouge2: 0.0486
—Rougel: 0.1082
—Rougelsum: 0.1082
—Gen Len: 21.0

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: 16
—evalbatchsize: 16
—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: 5
—mixedprecisiontraining: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
2.1526	1.0	1549	2.0667	0.1336	0.0483	0.1075	0.1075	21.0
2.0567	2.0	3098	2.0417	0.136	0.0486	0.1082	0.1082	21.0
1.975	3.0	4647	2.0308	0.1352	0.0488	0.1083	0.1082	21.0
1.9154	4.0	6196	2.0268	0.1356	0.048	0.1079	0.1078	21.0
1.8691	5.0	7745	2.0270	0.1355	0.0486	0.1082	0.1082	21.0

Framework versions

—Transformers 4.49.0.dev0
—Pytorch 2.5.1+cu121
—Datasets 3.2.0
—Tokenizers 0.21.0