--- license: mit base_model: microsoft/git-base tags: - generated_from_trainer model-index: - name: GenerativeImage2Text-naruto results: [] --- # GenerativeImage2Text-naruto This model is a fine-tuned version of [microsoft/git-base](https://huggingface.co/microsoft/git-base) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.0424 - Wer Score: 20.4909 ## 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: 5e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 2 - total_train_batch_size: 4 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 30 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer Score | |:-------------:|:-------:|:----:|:---------------:|:---------:| | 7.2788 | 0.3344 | 50 | 4.3352 | 10.2091 | | 2.0251 | 0.6689 | 100 | 0.2388 | 0.5182 | | 0.0807 | 1.0033 | 150 | 0.0401 | 0.5091 | | 0.0303 | 1.3378 | 200 | 0.0333 | 0.4455 | | 0.0284 | 1.6722 | 250 | 0.0324 | 1.1273 | | 0.0287 | 2.0067 | 300 | 0.0316 | 23.6909 | | 0.0239 | 2.3411 | 350 | 0.0331 | 23.7455 | | 0.022 | 2.6756 | 400 | 0.0328 | 23.7364 | | 0.0236 | 3.0100 | 450 | 0.0317 | 23.7 | | 0.0208 | 3.3445 | 500 | 0.0325 | 23.7636 | | 0.0206 | 3.6789 | 550 | 0.0324 | 19.4 | | 0.0214 | 4.0134 | 600 | 0.0320 | 23.6818 | | 0.0183 | 4.3478 | 650 | 0.0318 | 23.5636 | | 0.0194 | 4.6823 | 700 | 0.0314 | 23.7 | | 0.0203 | 5.0167 | 750 | 0.0339 | 23.7182 | | 0.0179 | 5.3512 | 800 | 0.0307 | 23.6909 | | 0.0178 | 5.6856 | 850 | 0.0323 | 23.7182 | | 0.0192 | 6.0201 | 900 | 0.0323 | 23.7273 | | 0.0166 | 6.3545 | 950 | 0.0330 | 23.7091 | | 0.0181 | 6.6890 | 1000 | 0.0346 | 23.7182 | | 0.0176 | 7.0234 | 1050 | 0.0335 | 23.7 | | 0.016 | 7.3579 | 1100 | 0.0357 | 23.7182 | | 0.0167 | 7.6923 | 1150 | 0.0341 | 23.7 | | 0.0165 | 8.0268 | 1200 | 0.0341 | 23.6727 | | 0.0152 | 8.3612 | 1250 | 0.0356 | 23.6727 | | 0.0158 | 8.6957 | 1300 | 0.0327 | 23.6273 | | 0.0158 | 9.0301 | 1350 | 0.0352 | 23.6909 | | 0.0148 | 9.3645 | 1400 | 0.0350 | 23.6909 | | 0.0153 | 9.6990 | 1450 | 0.0348 | 23.5182 | | 0.0153 | 10.0334 | 1500 | 0.0349 | 23.6818 | | 0.0142 | 10.3679 | 1550 | 0.0345 | 23.6818 | | 0.0143 | 10.7023 | 1600 | 0.0340 | 23.5727 | | 0.0152 | 11.0368 | 1650 | 0.0341 | 23.6818 | | 0.0137 | 11.3712 | 1700 | 0.0374 | 23.6818 | | 0.0142 | 11.7057 | 1750 | 0.0324 | 23.6545 | | 0.0141 | 12.0401 | 1800 | 0.0353 | 23.7 | | 0.0134 | 12.3746 | 1850 | 0.0362 | 23.1545 | | 0.0138 | 12.7090 | 1900 | 0.0357 | 23.6818 | | 0.0138 | 13.0435 | 1950 | 0.0359 | 22.8909 | | 0.0124 | 13.3779 | 2000 | 0.0372 | 23.4182 | | 0.0132 | 13.7124 | 2050 | 0.0370 | 23.6909 | | 0.0137 | 14.0468 | 2100 | 0.0380 | 23.7182 | | 0.0118 | 14.3813 | 2150 | 0.0358 | 23.6545 | | 0.0129 | 14.7157 | 2200 | 0.0376 | 23.6909 | | 0.013 | 15.0502 | 2250 | 0.0379 | 23.4636 | | 0.0115 | 15.3846 | 2300 | 0.0384 | 23.3909 | | 0.0125 | 15.7191 | 2350 | 0.0369 | 23.2364 | | 0.0126 | 16.0535 | 2400 | 0.0392 | 22.8545 | | 0.0115 | 16.3880 | 2450 | 0.0379 | 23.0455 | | 0.0118 | 16.7224 | 2500 | 0.0384 | 22.9455 | | 0.0119 | 17.0569 | 2550 | 0.0371 | 23.5636 | | 0.0108 | 17.3913 | 2600 | 0.0383 | 23.6455 | | 0.0116 | 17.7258 | 2650 | 0.0378 | 23.6818 | | 0.0114 | 18.0602 | 2700 | 0.0391 | 23.2091 | | 0.0101 | 18.3946 | 2750 | 0.0399 | 22.1545 | | 0.0113 | 18.7291 | 2800 | 0.0398 | 23.7182 | | 0.0108 | 19.0635 | 2850 | 0.0389 | 23.0 | | 0.01 | 19.3980 | 2900 | 0.0382 | 22.0818 | | 0.0107 | 19.7324 | 2950 | 0.0392 | 22.8636 | | 0.0109 | 20.0669 | 3000 | 0.0399 | 23.1091 | | 0.0096 | 20.4013 | 3050 | 0.0404 | 22.7273 | | 0.01 | 20.7358 | 3100 | 0.0398 | 22.3364 | | 0.0102 | 21.0702 | 3150 | 0.0402 | 22.3636 | | 0.0092 | 21.4047 | 3200 | 0.0395 | 21.7636 | | 0.0096 | 21.7391 | 3250 | 0.0396 | 21.8727 | | 0.0095 | 22.0736 | 3300 | 0.0394 | 21.4909 | | 0.0089 | 22.4080 | 3350 | 0.0407 | 20.3455 | | 0.009 | 22.7425 | 3400 | 0.0390 | 21.3727 | | 0.009 | 23.0769 | 3450 | 0.0398 | 20.9273 | | 0.0085 | 23.4114 | 3500 | 0.0406 | 20.8364 | | 0.0087 | 23.7458 | 3550 | 0.0399 | 21.8727 | | 0.0086 | 24.0803 | 3600 | 0.0409 | 20.9818 | | 0.0082 | 24.4147 | 3650 | 0.0412 | 20.8455 | | 0.0082 | 24.7492 | 3700 | 0.0406 | 21.5273 | | 0.0083 | 25.0836 | 3750 | 0.0409 | 21.4545 | | 0.008 | 25.4181 | 3800 | 0.0408 | 21.2818 | | 0.0077 | 25.7525 | 3850 | 0.0414 | 21.1273 | | 0.0075 | 26.0870 | 3900 | 0.0422 | 21.6727 | | 0.007 | 26.4214 | 3950 | 0.0421 | 20.6636 | | 0.0074 | 26.7559 | 4000 | 0.0428 | 21.2818 | | 0.0072 | 27.0903 | 4050 | 0.0429 | 20.8 | | 0.0066 | 27.4247 | 4100 | 0.0424 | 19.6182 | | 0.0066 | 27.7592 | 4150 | 0.0423 | 19.9364 | | 0.0066 | 28.0936 | 4200 | 0.0423 | 20.8636 | | 0.0059 | 28.4281 | 4250 | 0.0425 | 20.6545 | | 0.0066 | 28.7625 | 4300 | 0.0422 | 20.4909 | | 0.0065 | 29.0970 | 4350 | 0.0423 | 20.5091 | | 0.0061 | 29.4314 | 4400 | 0.0423 | 20.5 | | 0.0058 | 29.7659 | 4450 | 0.0424 | 20.4909 | ### Framework versions - Transformers 4.40.2 - Pytorch 2.3.0+cu121 - Datasets 2.19.1 - Tokenizers 0.19.1