You can deploy this model and start earning money today!

swadhindas324/
vit-convnext-mistral-UCM-with-4-tag-supressed-captioning

Name: vit-convnext-mistral-UCM-with-4-tag-supressed-captioning
Availability: InStock

transformers

safetensors

vision-encoder-decoder

generated_from_trainer

endpoints_compatible

Views

No views yet

vit-convnext-mistral-UCM-with-4-tag-supressed-captioning

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

Loss: 0.4353
Accuracy: 74.53
Bleu-1: 0.8349
Bleu-2: 0.7817
Bleu-3: 0.7349
Bleu-4: 0.6937
Meteor: 0.8192
Rouge-l: 0.8069
Cider: 3.4020

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
train_batch_size: 64
eval_batch_size: 64
seed: 50
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 0.1
num_epochs: 128
mixed_precision_training: 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	148	0.6413	72.29	0.4049	0.2728	0.1983	0.1465	0.3077	0.3382	0.3223
No log	2.0	296	0.4634	72.03	0.2490	0.1745	0.1133	0.0706	0.1848	0.2111	0.1811
No log	3.0	444	0.4135	69.96	0.6404	0.5529	0.4920	0.4477	0.5748	0.5899	2.0471
No log	4.0	592	0.3537	73.32	0.8223	0.7554	0.6993	0.6491	0.7916	0.7786	3.1124
No log	5.0	740	0.3440	73.75	0.8319	0.7669	0.7137	0.6657	0.8133	0.8007	3.3197
No log	6.0	888	0.3533	72.93	0.8111	0.7580	0.7093	0.6641	0.8164	0.7915	3.2107
0.4486	7.0	1036	0.3393	74.59	0.8646	0.8115	0.7642	0.7197	0.8366	0.8188	3.5237
0.4486	8.0	1184	0.3451	74.63	0.8464	0.7919	0.7454	0.7039	0.8089	0.8049	3.4166
0.4486	9.0	1332	0.3492	74.53	0.8657	0.8199	0.7794	0.7420	0.8342	0.8232	3.6122
0.4486	10.0	1480	0.3477	74.3	0.8398	0.7822	0.7316	0.6853	0.8198	0.8101	3.3844
0.4486	11.0	1628	0.3486	73.51	0.8541	0.7993	0.7528	0.7101	0.8368	0.8318	3.5081
0.4486	12.0	1776	0.3693	74.16	0.8520	0.7992	0.7528	0.7102	0.8199	0.8178	3.4034
0.4486	13.0	1924	0.3938	74.09	0.8247	0.7707	0.7259	0.6852	0.8093	0.7963	3.4500
0.1648	14.0	2072	0.3674	75.47	0.8727	0.8247	0.7831	0.7452	0.8503	0.8351	3.6181
0.1648	15.0	2220	0.3872	74.44	0.8297	0.7734	0.7272	0.6830	0.8161	0.8011	3.4549
0.1648	16.0	2368	0.3958	74.61	0.8422	0.7844	0.7370	0.6910	0.8052	0.7935	3.3137
0.1648	17.0	2516	0.3912	74.82	0.8552	0.8011	0.7570	0.7158	0.8164	0.8104	3.4658
0.1648	18.0	2664	0.4050	74.21	0.8359	0.7783	0.7299	0.6843	0.8187	0.8085	3.3851
0.1648	19.0	2812	0.4009	74.9	0.8647	0.8116	0.7673	0.7244	0.8310	0.8202	3.4782
0.1648	20.0	2960	0.3937	75.06	0.8421	0.7855	0.7384	0.6962	0.8318	0.8176	3.5099
0.1395	21.0	3108	0.4152	73.89	0.8411	0.7824	0.7362	0.6956	0.8065	0.7973	3.4434
0.1395	22.0	3256	0.4222	74.8	0.8297	0.7696	0.7210	0.6777	0.8153	0.7932	3.3972
0.1395	23.0	3404	0.4127	73.95	0.8517	0.7929	0.7438	0.6984	0.8171	0.8090	3.3821
0.1395	24.0	3552	0.4353	74.53	0.8349	0.7817	0.7349	0.6937	0.8192	0.8069	3.4020

Framework versions

Transformers 5.12.1
Pytorch 2.12.1+cu130
Datasets 5.0.0
Tokenizers 0.22.2

swadhindas324/vit-convnext-mistral-UCM-with-4-tag-supressed-captioning

vit-convnext-mistral-UCM-with-4-tag-supressed-captioning

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

swadhindas324/
vit-convnext-mistral-UCM-with-4-tag-supressed-captioning