Modelpublic

Image-Captioning-ML/Vit-GPT2-COCO2017Flickr-115k-nlpconnect

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face

0likes13downloads

Model Card

Vit-GPT2-COCO2017Flickr-115k-nlpconnect

This model is a fine-tuned version of NourFakih/Vit-GPT2-COCO2017Flickr-115k-nlpconnect on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 0.4978
—Rouge1: 40.7656
—Rouge2: 15.3367
—Rougel: 36.9775
—Rougelsum: 36.9808
—Gen Len: 11.8545

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: 4
—evalbatchsize: 4
—seed: 42
—gradientaccumulationsteps: 4
—totaltrainbatch_size: 16
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Gen Len	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
0.4218	0.2783	2000	11.7001	0.4402	40.8374	15.0971	36.9258	36.9122
0.4071	0.5565	4000	11.2742	0.4353	41.1785	15.7543	37.5183	37.5112
0.3914	0.8348	6000	11.7367	0.4345	41.4425	15.991	37.5387	37.5422
0.3367	1.1130	8000	11.8434	0.4481	41.0677	15.8609	37.3991	37.3749
0.3265	1.3913	10000	11.6695	0.4518	41.3243	16.1077	37.5478	37.5367
0.3158	1.6696	12000	11.7170	0.4556	41.3022	15.9154	37.3943	37.3924
0.3094	1.9479	14000	11.5969	0.4602	41.411	16.1338	37.6313	37.6262
0.2639	2.2262	16000	11.8349	0.4871	40.8402	15.4182	36.9302	36.9355
0.2515	2.5045	18000	0.4940	41.0116	15.4407	37.0026	37.0188	11.8727
0.253	2.7827	20000	0.4978	40.7656	15.3367	36.9775	36.9808	11.8545

Framework versions

—Transformers 4.42.3
—Pytorch 2.1.2
—Datasets 2.20.0
—Tokenizers 0.19.1