Modelpublic

swadhindas324/vit-resnet-mistral-RSICD-with-6-captioning

sourceHugging Faceupdated 3mo agoView on Hugging Face

0likes3downloads

Model Card

vit-resnet-mistral-RSICD-with-6-captioning

This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 1.0749
—Accuracy: 83.69
—Bleu-1: 0.7129
—Bleu-2: 0.5909
—Bleu-3: 0.4955
—Bleu-4: 0.4213
—Meteor: 0.6677
—Rouge-l: 0.6364
—Cider: 2.2411

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: 0.0001
—trainbatchsize: 64
—evalbatchsize: 64
—seed: 50
—optimizer: Use OptimizerNames.ADAMWTORCHFUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
—lrschedulertype: linear
—lrschedulerwarmup_steps: 0.1
—num_epochs: 128
—mixedprecisiontraining: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Bleu-1	Bleu-2	Bleu-3	Bleu-4	Meteor	Rouge-l	Cider
No log	1.0	768	0.6911	80.18	0.4668	0.3027	0.2115	0.1537	0.4085	0.3975	0.4800
0.9634	2.0	1536	0.6668	80.0	0.5394	0.3880	0.2931	0.2275	0.5084	0.4761	0.8994
0.4727	3.0	2304	0.6267	81.01	0.7016	0.5789	0.4859	0.4116	0.6863	0.6354	2.2058
0.3503	4.0	3072	0.6738	82.3	0.7048	0.5808	0.4869	0.4135	0.6742	0.6340	2.2838
0.3503	5.0	3840	0.7228	80.8	0.6741	0.5502	0.4577	0.3871	0.6539	0.6097	2.1323
0.2434	6.0	4608	0.7652	82.97	0.7096	0.5857	0.4930	0.4206	0.6704	0.6376	2.2930
0.1932	7.0	5376	0.8145	82.46	0.6879	0.5653	0.4734	0.4019	0.6713	0.6244	2.1774
0.1692	8.0	6144	0.8359	82.7	0.6868	0.5636	0.4715	0.3996	0.6713	0.6243	2.2719
0.1692	9.0	6912	0.8724	83.21	0.6891	0.5638	0.4692	0.3949	0.6685	0.6272	2.1536
0.1510	10.0	7680	0.8612	83.31	0.7111	0.5896	0.4942	0.4196	0.6844	0.6435	2.3086
0.1461	11.0	8448	0.9065	83.18	0.6983	0.5737	0.4806	0.4097	0.6660	0.6240	2.2408
0.1426	12.0	9216	0.9210	83.13	0.6999	0.5739	0.4788	0.4049	0.6728	0.6297	2.1826
0.1426	13.0	9984	0.9393	83.65	0.7133	0.5897	0.4951	0.4221	0.6770	0.6403	2.2777
0.1358	14.0	10752	0.9753	83.58	0.7189	0.5983	0.5059	0.4324	0.6893	0.6496	2.4369
0.1278	15.0	11520	0.9849	83.49	0.7123	0.5919	0.4995	0.4256	0.6884	0.6468	2.3441
0.1207	16.0	12288	1.0215	83.64	0.7169	0.5932	0.4992	0.4240	0.6777	0.6392	2.3050
0.1207	17.0	13056	1.0412	83.2	0.6969	0.5727	0.4785	0.4060	0.6759	0.6289	2.2425
0.1145	18.0	13824	1.0660	83.19	0.7011	0.5772	0.4834	0.4109	0.6777	0.6316	2.2432
0.1111	19.0	14592	1.0749	83.69	0.7129	0.5909	0.4955	0.4213	0.6677	0.6364	2.2411

Framework versions

—Transformers 5.12.1
—Pytorch 2.12.1+cu130
—Datasets 5.0.0
—Tokenizers 0.22.2