convnext-Mistral-RSICD-without-captioning

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

• Loss: 1.7262
• Accuracy: 82.42
• Bleu-1: 0.6502
• Bleu-2: 0.5182
• Bleu-3: 0.4283
• Bleu-4: 0.3617
• Meteor: 0.6100
• Rouge-l: 0.5789
• Cider: 1.9325

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	768	1.3567	79.43	0.5159	0.3505	0.2526	0.1874	0.4593	0.4496	0.6915
2.4285	2.0	1536	1.2939	79.72	0.5731	0.4267	0.3287	0.2551	0.5497	0.5140	1.1645
0.9411	3.0	2304	1.2522	81.95	0.6864	0.5638	0.4696	0.3948	0.6723	0.6212	2.1220
0.7487	4.0	3072	1.3267	82.15	0.6896	0.5674	0.4718	0.3962	0.6822	0.6352	2.1771
0.7487	5.0	3840	1.4171	81.5	0.6543	0.5272	0.4349	0.3663	0.6425	0.5917	1.9704
0.5341	6.0	4608	1.5311	81.76	0.6421	0.5107	0.4177	0.3488	0.6200	0.5742	1.9267
0.4147	7.0	5376	1.6184	81.19	0.6232	0.4906	0.3963	0.3264	0.6132	0.5581	1.7155
0.3572	8.0	6144	1.6795	81.16	0.6154	0.4823	0.3871	0.3168	0.6009	0.5490	1.7236
0.3572	9.0	6912	1.7262	82.42	0.6502	0.5182	0.4283	0.3617	0.6100	0.5789	1.9325

Framework versions

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