swadhindas324/vit-Mistral-SYDNEY-without-captioning

vit-Mistral-SYDNEY-captioning

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

• Loss: 0.8482
• Accuracy: 67.54
• Bleu-1: 0.7351
• Bleu-2: 0.6309
• Bleu-3: 0.5564
• Bleu-4: 0.4927
• Meteor: 0.6677
• Rouge-l: 0.6505
• Cider: 2.0083

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: 8
• 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: 1024
• 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	44	1.5693	62.43	0.5118	0.3929	0.3181	0.2620	0.4362	0.4382	0.7064
No log	2.0	88	0.7227	64.47	0.5855	0.4692	0.3913	0.3316	0.5060	0.5060	1.2204
No log	3.0	132	0.6898	64.72	0.6252	0.5163	0.4403	0.3815	0.5523	0.5521	1.5630
No log	4.0	176	0.6758	65.03	0.7419	0.6427	0.5589	0.4862	0.6804	0.6683	2.1625
No log	5.0	220	0.6506	65.85	0.7711	0.6877	0.6176	0.5563	0.7144	0.7058	2.5448
No log	6.0	264	0.6865	65.56	0.7449	0.6559	0.5804	0.5151	0.7033	0.6794	2.3313
No log	7.0	308	0.6974	65.74	0.7336	0.6468	0.5791	0.5212	0.6921	0.6775	2.3155
No log	8.0	352	0.7350	66.05	0.7702	0.6816	0.6095	0.5482	0.7081	0.6973	2.3791
No log	9.0	396	0.7854	65.8	0.7346	0.6479	0.5756	0.5127	0.6798	0.6617	2.1718
No log	10.0	440	0.7893	65.37	0.7412	0.6419	0.5654	0.5017	0.6757	0.6570	2.1446
No log	11.0	484	0.7860	66.74	0.7447	0.6614	0.5973	0.5446	0.6896	0.6689	2.3501
No log	12.0	528	0.8342	67.07	0.7396	0.6334	0.5508	0.4877	0.6492	0.6422	2.0472
No log	13.0	572	0.8145	65.79	0.7446	0.6635	0.5905	0.5232	0.7071	0.6929	2.2397
No log	14.0	616	0.8722	65.85	0.7403	0.6608	0.5946	0.5355	0.6837	0.6835	2.4422
No log	15.0	660	0.8482	67.54	0.7351	0.6309	0.5564	0.4927	0.6677	0.6505	2.0083

Framework versions

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