Vit Resnet Mistral SYDNEY With All Captioning

--- library_name: transformers tags: - generated_from_trainer metrics: - accuracy model-index: - name: vit-resnet-mistral-SYDNEY-with-all-captioning results: [] --- # vit-resnet-mistral-SYDNEY-with-all-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: 0.5411 - Accuracy: 69.9 - Bleu-1: 0.8300 - Bleu-2: 0.7620 - Bleu-3: 0.6937 - Bleu-4: 0.6263 - Meteor: 0.7810 - Rouge-l: 0.7509 - Cider: 2.6447 ## 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 | 44 | 0.9216 | 65.16 | 0.5177 | 0.3973 | 0.3172 | 0.2491 | 0.4243 | 0.4448 | 0.8130 | | No log | 2.0 | 88 | 0.5326 | 60.67 | 0.5480 | 0.4380 | 0.3590 | 0.2933 | 0.4270 | 0.4606 | 0.7265 | | No log | 3.0 | 132 | 0.4546 | 66.45 | 0.6671 | 0.5678 | 0.4812 | 0.4100 | 0.5853 | 0.5987 | 1.5583 | | No log | 4.0 | 176 | 0.4336 | 67.28 | 0.7962 | 0.6981 | 0.5962 | 0.4938 | 0.6924 | 0.6647 | 2.1904 | | No log | 5.0 | 220 | 0.4188 | 66.94 | 0.7878 | 0.7010 | 0.6149 | 0.5279 | 0.6961 | 0.6574 | 2.1957 | | No log | 6.0 | 264 | 0.4144 | 67.86 | 0.7966 | 0.7073 | 0.6256 | 0.5551 | 0.7115 | 0.7046 | 2.2405 | | No log | 7.0 | 308 | 0.4158 | 68.52 | 0.8088 | 0.7386 | 0.6737 | 0.6119 | 0.7577 | 0.7377 | 2.6949 | | No log | 8.0 | 352 | 0.4071 | 66.37 | 0.8344 | 0.7731 | 0.7173 | 0.6636 | 0.7652 | 0.7330 | 2.8783 | | No log | 9.0 | 396 | 0.4220 | 68.17 | 0.8119 | 0.7336 | 0.6583 | 0.5891 | 0.7480 | 0.7217 | 2.5704 | | No log | 10.0 | 440 | 0.4312 | 67.91 | 0.8265 | 0.7582 | 0.6949 | 0.6342 | 0.7704 | 0.7377 | 2.6809 | | No log | 11.0 | 484 | 0.4222 | 67.84 | 0.8088 | 0.7316 | 0.6630 | 0.6021 | 0.7504 | 0.7023 | 2.5702 | | No log | 12.0 | 528 | 0.4490 | 68.89 | 0.7666 | 0.6743 | 0.5951 | 0.5249 | 0.7054 | 0.6801 | 2.1787 | | No log | 13.0 | 572 | 0.4569 | 66.22 | 0.8226 | 0.7500 | 0.6837 | 0.6151 | 0.7536 | 0.7050 | 2.4849 | | No log | 14.0 | 616 | 0.4779 | 67.71 | 0.8036 | 0.7170 | 0.6350 | 0.5555 | 0.7338 | 0.7170 | 2.4574 | | No log | 15.0 | 660 | 0.4687 | 69.29 | 0.7940 | 0.7129 | 0.6450 | 0.5839 | 0.7581 | 0.7203 | 2.3877 | | No log | 16.0 | 704 | 0.4756 | 69.24 | 0.8029 | 0.7365 | 0.6745 | 0.6124 | 0.7568 | 0.7258 | 2.5659 | | No log | 17.0 | 748 | 0.4831 | 68.54 | 0.7953 | 0.7138 | 0.6447 | 0.5834 | 0.7512 | 0.7187 | 2.5023 | | No log | 18.0 | 792 | 0.4797 | 69.4 | 0.8362 | 0.7713 | 0.7099 | 0.6546 | 0.7829 | 0.7694 | 2.8833 | | No log | 19.0 | 836 | 0.5081 | 68.1 | 0.7938 | 0.7088 | 0.6442 | 0.5880 | 0.7390 | 0.7002 | 2.5298 | | No log | 20.0 | 880 | 0.5061 | 69.23 | 0.8101 | 0.7385 | 0.6736 | 0.6146 | 0.7733 | 0.7545 | 2.6758 | | No log | 21.0 | 924 | 0.4991 | 67.62 | 0.7925 | 0.7205 | 0.6606 | 0.6027 | 0.7377 | 0.6856 | 2.3866 | | No log | 22.0 | 968 | 0.5088 | 68.65 | 0.7868 | 0.7032 | 0.6369 | 0.5757 | 0.7550 | 0.7296 | 2.5237 | | No log | 23.0 | 1012 | 0.5557 | 67.61 | 0.7805 | 0.6971 | 0.6255 | 0.5587 | 0.7281 | 0.6993 | 2.3368 | | 0.2764 | 24.0 | 1056 | 0.5111 | 68.14 | 0.8103 | 0.7408 | 0.6742 | 0.6135 | 0.7895 | 0.7631 | 2.6267 | | 0.2764 | 25.0 | 1100 | 0.5345 | 68.61 | 0.7773 | 0.7014 | 0.6391 | 0.5815 | 0.7571 | 0.7217 | 2.4595 | | 0.2764 | 26.0 | 1144 | 0.5483 | 68.49 | 0.8128 | 0.7309 | 0.6575 | 0.5921 | 0.7493 | 0.7341 | 2.5915 | | 0.2764 | 27.0 | 1188 | 0.5501 | 68.18 | 0.7516 | 0.6482 | 0.5656 | 0.4888 | 0.7216 | 0.6727 | 2.2985 | | 0.2764 | 28.0 | 1232 | 0.5411 | 69.9 | 0.8300 | 0.7620 | 0.6937 | 0.6263 | 0.7810 | 0.7509 | 2.6447 | ### Framework versions - Transformers 5.12.1 - Pytorch 2.12.1+cu130 - Datasets 5.0.0 - Tokenizers 0.22.2