Modelpublic

fashxp/flan-t5-large-cars-descriptions

sourceHugging Faceapache-2.0updated 3y agoView on Hugging Face

0likes7downloads

Model Card

flan-t5-large-cars-descriptions

This model is a fine-tuned version of google/flan-t5-large on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 0.7687
—Rouge1: 19.5278
—Rouge2: 13.4042
—Rougel: 17.2713
—Rougelsum: 18.2289
—Gen Len: 19.0

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: 5e-05
—trainbatchsize: 8
—evalbatchsize: 8
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	28	0.8985	20.106	13.8185	17.5397	19.1007	19.0
No log	2.0	56	0.8444	19.963	13.2848	17.4604	18.8592	19.0
No log	3.0	84	0.8163	19.9916	13.9678	17.8614	19.0133	19.0
No log	4.0	112	0.7918	19.8144	12.9257	17.5265	18.6706	19.0
No log	5.0	140	0.7859	19.6936	13.9027	17.548	18.4906	19.0
No log	6.0	168	0.7796	19.4132	13.4927	17.3465	18.1291	19.0
No log	7.0	196	0.7732	19.7803	13.3453	17.5581	18.4902	19.0
No log	8.0	224	0.7724	19.2412	12.4472	16.8241	17.8911	19.0
No log	9.0	252	0.7687	19.5278	13.4042	17.2713	18.2289	19.0
No log	10.0	280	0.7691	19.5748	13.3162	17.5093	18.4578	19.0

Framework versions

—Transformers 4.35.1
—Pytorch 2.1.0+cu118
—Datasets 2.14.6
—Tokenizers 0.14.1