Vit Resnet Mistral UCM With 7 Captioning

--- library_name: transformers tags: - generated_from_trainer metrics: - accuracy model-index: - name: vit-resnet-mistral-UCM-with-7-captioning results: [] --- # vit-resnet-mistral-UCM-with-7-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: 0.4441 - Accuracy: 75.27 - Bleu-1: 0.8363 - Bleu-2: 0.7755 - Bleu-3: 0.7257 - Bleu-4: 0.6775 - Meteor: 0.8123 - Rouge-l: 0.8096 - Cider: 3.4695 ## 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 | 0.6536 | 72.42 | 0.4105 | 0.2785 | 0.2020 | 0.1494 | 0.3113 | 0.3459 | 0.3319 | | No log | 2.0 | 296 | 0.4697 | 72.06 | 0.2955 | 0.2054 | 0.1355 | 0.0899 | 0.2216 | 0.2446 | 0.3533 | | No log | 3.0 | 444 | 0.4138 | 70.64 | 0.6360 | 0.5542 | 0.4987 | 0.4588 | 0.5969 | 0.6060 | 2.0977 | | No log | 4.0 | 592 | 0.3570 | 73.61 | 0.8194 | 0.7496 | 0.6934 | 0.6434 | 0.7781 | 0.7675 | 3.1170 | | No log | 5.0 | 740 | 0.3480 | 73.93 | 0.8319 | 0.7783 | 0.7334 | 0.6936 | 0.8076 | 0.8054 | 3.3582 | | No log | 6.0 | 888 | 0.3498 | 73.56 | 0.8317 | 0.7786 | 0.7311 | 0.6870 | 0.8191 | 0.8100 | 3.2555 | | 0.4509 | 7.0 | 1036 | 0.3420 | 74.67 | 0.8585 | 0.8024 | 0.7525 | 0.7051 | 0.8420 | 0.8206 | 3.4377 | | 0.4509 | 8.0 | 1184 | 0.3477 | 74.31 | 0.8443 | 0.7879 | 0.7395 | 0.6951 | 0.8125 | 0.8037 | 3.3959 | | 0.4509 | 9.0 | 1332 | 0.3564 | 74.63 | 0.8561 | 0.8057 | 0.7614 | 0.7211 | 0.8249 | 0.8191 | 3.5346 | | 0.4509 | 10.0 | 1480 | 0.3581 | 74.49 | 0.8547 | 0.8023 | 0.7557 | 0.7141 | 0.8276 | 0.8197 | 3.4898 | | 0.4509 | 11.0 | 1628 | 0.3627 | 73.54 | 0.8290 | 0.7593 | 0.6998 | 0.6428 | 0.8323 | 0.8019 | 3.2609 | | 0.4509 | 12.0 | 1776 | 0.3632 | 73.98 | 0.8399 | 0.7772 | 0.7225 | 0.6721 | 0.8128 | 0.8043 | 3.2854 | | 0.4509 | 13.0 | 1924 | 0.3915 | 73.71 | 0.8273 | 0.7655 | 0.7138 | 0.6683 | 0.7949 | 0.7856 | 3.3208 | | 0.1648 | 14.0 | 2072 | 0.3855 | 75.05 | 0.8563 | 0.8039 | 0.7556 | 0.7094 | 0.8355 | 0.8179 | 3.3844 | | 0.1648 | 15.0 | 2220 | 0.3761 | 74.95 | 0.8625 | 0.8091 | 0.7633 | 0.7195 | 0.8317 | 0.8215 | 3.6767 | | 0.1648 | 16.0 | 2368 | 0.3893 | 75.3 | 0.8545 | 0.7973 | 0.7498 | 0.7053 | 0.8227 | 0.8134 | 3.4774 | | 0.1648 | 17.0 | 2516 | 0.3932 | 74.59 | 0.8589 | 0.8052 | 0.7596 | 0.7167 | 0.8454 | 0.8300 | 3.5445 | | 0.1648 | 18.0 | 2664 | 0.3982 | 74.15 | 0.8455 | 0.7890 | 0.7438 | 0.7021 | 0.8243 | 0.8119 | 3.4521 | | 0.1648 | 19.0 | 2812 | 0.3968 | 74.61 | 0.8570 | 0.7998 | 0.7537 | 0.7108 | 0.8373 | 0.8204 | 3.5498 | | 0.1648 | 20.0 | 2960 | 0.4207 | 74.59 | 0.8324 | 0.7716 | 0.7228 | 0.6807 | 0.8106 | 0.7982 | 3.3747 | | 0.1395 | 21.0 | 3108 | 0.4240 | 73.63 | 0.8482 | 0.7943 | 0.7485 | 0.7055 | 0.8319 | 0.8198 | 3.4832 | | 0.1395 | 22.0 | 3256 | 0.4294 | 74.56 | 0.8344 | 0.7754 | 0.7270 | 0.6825 | 0.8162 | 0.7966 | 3.4112 | | 0.1395 | 23.0 | 3404 | 0.4330 | 74.23 | 0.8553 | 0.7997 | 0.7526 | 0.7077 | 0.8297 | 0.8136 | 3.4253 | | 0.1395 | 24.0 | 3552 | 0.4256 | 74.43 | 0.8323 | 0.7781 | 0.7330 | 0.6903 | 0.8142 | 0.8045 | 3.3820 | | 0.1395 | 25.0 | 3700 | 0.4323 | 74.19 | 0.8438 | 0.7890 | 0.7397 | 0.6931 | 0.8339 | 0.8093 | 3.3894 | | 0.1395 | 26.0 | 3848 | 0.4420 | 74.66 | 0.8626 | 0.8058 | 0.7564 | 0.7125 | 0.8367 | 0.8243 | 3.5029 | | 0.1395 | 27.0 | 3996 | 0.4441 | 75.27 | 0.8363 | 0.7755 | 0.7257 | 0.6775 | 0.8123 | 0.8096 | 3.4695 | ### Framework versions - Transformers 5.12.1 - Pytorch 2.12.1+cu130 - Datasets 5.0.0 - Tokenizers 0.22.2