Modelpublic

vsamuel/scifive_seven_epoch

sourceHugging Faceupdated 3y agoView on Hugging Face

0likes10downloads

Model Card

scifivesevenepoch

This model is a fine-tuned version of razent/SciFive-base-Pubmed_PMC on the None dataset. It achieves the following results on the evaluation set:

—Loss: 1.8379
—Rouge1: 0.3666
—Rouge2: 0.2139
—Rougel: 0.307
—Rougelsum: 0.3084
—Gen Len: 17.53

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: 4
—evalbatchsize: 2
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 7

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	275	2.2065	0.2714	0.1462	0.2336	0.2327	17.09
2.4022	2.0	550	1.9771	0.339	0.1958	0.2907	0.2911	17.76
2.4022	3.0	825	1.9156	0.3652	0.2128	0.3035	0.3046	17.73
1.8204	4.0	1100	1.8698	0.37	0.2197	0.3103	0.3102	17.37
1.8204	5.0	1375	1.8525	0.3638	0.2103	0.3042	0.3049	17.65
1.694	6.0	1650	1.8384	0.3713	0.2175	0.3104	0.3113	17.55
1.694	7.0	1925	1.8379	0.3666	0.2139	0.307	0.3084	17.53

Framework versions

—Transformers 4.35.2
—Pytorch 2.1.0+cu121
—Datasets 2.16.1
—Tokenizers 0.15.1