vit-resnet-Mistral-UCM-without-captioning

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

• Loss: 0.7723
• Accuracy: 74.72
• Bleu-1: 0.8542
• Bleu-2: 0.8011
• Bleu-3: 0.7571
• Bleu-4: 0.7146
• Meteor: 0.8359
• Rouge-l: 0.8142
• Cider: 3.4768

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	0.8708	72.22	0.4211	0.2916	0.2159	0.1619	0.3291	0.3586	0.3669
No log	2.0	296	0.7341	72.27	0.4540	0.3400	0.2660	0.2165	0.3409	0.3912	0.9591
No log	3.0	444	0.5835	73.39	0.8091	0.7497	0.7010	0.6580	0.7778	0.7731	3.1833
No log	4.0	592	0.5842	73.43	0.8425	0.7781	0.7243	0.6759	0.8107	0.7959	3.3423
No log	5.0	740	0.6087	72.81	0.8297	0.7712	0.7224	0.6793	0.8108	0.7996	3.3063
No log	6.0	888	0.6218	74.11	0.8368	0.7832	0.7366	0.6943	0.8268	0.8166	3.4069
0.6500	7.0	1036	0.6235	74.98	0.8882	0.8454	0.8049	0.7684	0.8426	0.8398	3.6889
0.6500	8.0	1184	0.6569	73.84	0.8362	0.7828	0.7366	0.6927	0.8170	0.8073	3.4078
0.6500	9.0	1332	0.6680	74.68	0.8602	0.8159	0.7755	0.7382	0.8410	0.8288	3.5358
0.6500	10.0	1480	0.6681	74.36	0.8586	0.8006	0.7512	0.7072	0.8380	0.8178	3.4580
0.6500	11.0	1628	0.6793	74.5	0.8436	0.7810	0.7303	0.6836	0.8207	0.8042	3.4041
0.6500	12.0	1776	0.7174	74.67	0.8538	0.7997	0.7542	0.7126	0.8395	0.8242	3.4757
0.6500	13.0	1924	0.7296	73.88	0.8275	0.7718	0.7216	0.6750	0.8194	0.8010	3.4220
0.2971	14.0	2072	0.7463	74.55	0.8506	0.7966	0.7507	0.7072	0.8293	0.8148	3.4686
0.2971	15.0	2220	0.7314	74.73	0.8516	0.7928	0.7409	0.6905	0.8305	0.8164	3.5680
0.2971	16.0	2368	0.7657	74.19	0.8514	0.8010	0.7539	0.7081	0.8363	0.8183	3.4412
0.2971	17.0	2516	0.7723	74.72	0.8542	0.8011	0.7571	0.7146	0.8359	0.8142	3.4768

Framework versions

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