--- license: apache-2.0 base_model: google/vit-base-patch16-224-in21k tags: - generated_from_trainer model-index: - name: ryan03312024_lr_2e-5_wd_001 results: [] --- # ryan03312024_lr_2e-5_wd_001 This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.1952 - Ordinal Mae: 0.4087 - Ordinal Accuracy: 0.6887 - Na Accuracy: 0.8574 ## 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: 2e-05 - train_batch_size: 16 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 1.5 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Ordinal Mae | Ordinal Accuracy | Na Accuracy | |:-------------:|:-----:|:----:|:---------------:|:-----------:|:----------------:|:-----------:| | 0.4436 | 0.04 | 100 | 0.3698 | 0.8706 | 0.3332 | 0.7990 | | 0.3143 | 0.07 | 200 | 0.3215 | 0.8555 | 0.4017 | 0.8093 | | 0.3385 | 0.11 | 300 | 0.2997 | 0.8303 | 0.4485 | 0.8591 | | 0.3127 | 0.14 | 400 | 0.2889 | 0.8013 | 0.4881 | 0.8746 | | 0.3054 | 0.18 | 500 | 0.2804 | 0.7619 | 0.5325 | 0.8780 | | 0.3051 | 0.22 | 600 | 0.2752 | 0.7215 | 0.5235 | 0.9158 | | 0.2833 | 0.25 | 700 | 0.2653 | 0.6807 | 0.5487 | 0.8969 | | 0.2907 | 0.29 | 800 | 0.2550 | 0.6432 | 0.5618 | 0.8351 | | 0.2468 | 0.32 | 900 | 0.2522 | 0.6119 | 0.5972 | 0.8058 | | 0.2199 | 0.36 | 1000 | 0.2437 | 0.6023 | 0.6062 | 0.8127 | | 0.2219 | 0.4 | 1100 | 0.2361 | 0.5574 | 0.5959 | 0.9038 | | 0.2071 | 0.43 | 1200 | 0.2387 | 0.5439 | 0.6175 | 0.7715 | | 0.2214 | 0.47 | 1300 | 0.2341 | 0.5257 | 0.6232 | 0.7955 | | 0.2627 | 0.5 | 1400 | 0.2315 | 0.5152 | 0.6124 | 0.7990 | | 0.2067 | 0.54 | 1500 | 0.2247 | 0.5026 | 0.6396 | 0.8110 | | 0.2086 | 0.58 | 1600 | 0.2192 | 0.4955 | 0.6589 | 0.8041 | | 0.1993 | 0.61 | 1700 | 0.2182 | 0.4738 | 0.6522 | 0.8127 | | 0.1962 | 0.65 | 1800 | 0.2211 | 0.4858 | 0.6232 | 0.9141 | | 0.1882 | 0.69 | 1900 | 0.2045 | 0.4669 | 0.6632 | 0.8625 | | 0.1895 | 0.72 | 2000 | 0.2082 | 0.4696 | 0.6316 | 0.8608 | | 0.1979 | 0.76 | 2100 | 0.2270 | 0.4791 | 0.6373 | 0.9003 | | 0.2643 | 0.79 | 2200 | 0.2069 | 0.4663 | 0.6414 | 0.8557 | | 0.2279 | 0.83 | 2300 | 0.2030 | 0.4581 | 0.6543 | 0.8694 | | 0.1965 | 0.87 | 2400 | 0.2109 | 0.4446 | 0.6820 | 0.8007 | | 0.1637 | 0.9 | 2500 | 0.2005 | 0.4439 | 0.6763 | 0.8557 | | 0.1705 | 0.94 | 2600 | 0.1964 | 0.4321 | 0.6748 | 0.8540 | | 0.2412 | 0.97 | 2700 | 0.1958 | 0.4345 | 0.6730 | 0.8780 | | 0.1438 | 1.01 | 2800 | 0.1972 | 0.4301 | 0.6784 | 0.8471 | | 0.123 | 1.05 | 2900 | 0.1995 | 0.4231 | 0.6753 | 0.8419 | | 0.1411 | 1.08 | 3000 | 0.1946 | 0.4220 | 0.6817 | 0.8454 | | 0.1443 | 1.12 | 3100 | 0.1916 | 0.4221 | 0.6828 | 0.8591 | | 0.208 | 1.15 | 3200 | 0.1942 | 0.4163 | 0.6740 | 0.8677 | | 0.1343 | 1.19 | 3300 | 0.1962 | 0.4182 | 0.6889 | 0.8471 | | 0.1347 | 1.23 | 3400 | 0.1938 | 0.4161 | 0.6900 | 0.8660 | | 0.1076 | 1.26 | 3500 | 0.1970 | 0.4181 | 0.6943 | 0.8471 | | 0.1248 | 1.3 | 3600 | 0.1951 | 0.4151 | 0.6959 | 0.8471 | | 0.1455 | 1.33 | 3700 | 0.1952 | 0.4147 | 0.6851 | 0.8814 | | 0.131 | 1.37 | 3800 | 0.1953 | 0.4172 | 0.6948 | 0.8454 | | 0.1307 | 1.41 | 3900 | 0.1932 | 0.4127 | 0.6928 | 0.8643 | | 0.1198 | 1.44 | 4000 | 0.1947 | 0.4110 | 0.6941 | 0.8574 | | 0.1363 | 1.48 | 4100 | 0.1952 | 0.4087 | 0.6887 | 0.8574 | ### Framework versions - Transformers 4.39.1 - Pytorch 2.2.1+cu121 - Datasets 2.18.0 - Tokenizers 0.15.2