--- library_name: transformers license: other base_model: nvidia/mit-b0 tags: - vision - image-segmentation - generated_from_trainer model-index: - name: segformer-b0-finetuned-wrinkle results: [] --- # segformer-b0-finetuned-wrinkle This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on the face-wrinkles dataset. It achieves the following results on the evaluation set: - Loss: 0.0188 - Mean Iou: 0.2008 - Mean Accuracy: 0.4015 - Overall Accuracy: 0.4015 - Accuracy Unlabeled: nan - Accuracy Wrinkle: 0.4015 - Iou Unlabeled: 0.0 - Iou Wrinkle: 0.4015 ## 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: 6e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Use adamw_torch 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: 100 - num_epochs: 10 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Accuracy Unlabeled | Accuracy Wrinkle | Iou Unlabeled | Iou Wrinkle | |:-------------:|:------:|:----:|:---------------:|:--------:|:-------------:|:----------------:|:------------------:|:----------------:|:-------------:|:-----------:| | 0.0127 | 0.2174 | 20 | 0.0185 | 0.1901 | 0.3802 | 0.3802 | nan | 0.3802 | 0.0 | 0.3802 | | 0.011 | 0.4348 | 40 | 0.0185 | 0.1886 | 0.3772 | 0.3772 | nan | 0.3772 | 0.0 | 0.3772 | | 0.0109 | 0.6522 | 60 | 0.0190 | 0.1380 | 0.2761 | 0.2761 | nan | 0.2761 | 0.0 | 0.2761 | | 0.0157 | 0.8696 | 80 | 0.0190 | 0.1587 | 0.3174 | 0.3174 | nan | 0.3174 | 0.0 | 0.3174 | | 0.0155 | 1.0870 | 100 | 0.0187 | 0.2034 | 0.4068 | 0.4068 | nan | 0.4068 | 0.0 | 0.4068 | | 0.0185 | 1.3043 | 120 | 0.0184 | 0.1819 | 0.3639 | 0.3639 | nan | 0.3639 | 0.0 | 0.3639 | | 0.0164 | 1.5217 | 140 | 0.0192 | 0.2445 | 0.4890 | 0.4890 | nan | 0.4890 | 0.0 | 0.4890 | | 0.0202 | 1.7391 | 160 | 0.0187 | 0.1624 | 0.3249 | 0.3249 | nan | 0.3249 | 0.0 | 0.3249 | | 0.008 | 1.9565 | 180 | 0.0185 | 0.1828 | 0.3656 | 0.3656 | nan | 0.3656 | 0.0 | 0.3656 | | 0.015 | 2.1739 | 200 | 0.0190 | 0.2415 | 0.4831 | 0.4831 | nan | 0.4831 | 0.0 | 0.4831 | | 0.0119 | 2.3913 | 220 | 0.0186 | 0.2115 | 0.4230 | 0.4230 | nan | 0.4230 | 0.0 | 0.4230 | | 0.0113 | 2.6087 | 240 | 0.0185 | 0.1545 | 0.3090 | 0.3090 | nan | 0.3090 | 0.0 | 0.3090 | | 0.0125 | 2.8261 | 260 | 0.0187 | 0.1798 | 0.3597 | 0.3597 | nan | 0.3597 | 0.0 | 0.3597 | | 0.0197 | 3.0435 | 280 | 0.0195 | 0.1493 | 0.2987 | 0.2987 | nan | 0.2987 | 0.0 | 0.2987 | | 0.0116 | 3.2609 | 300 | 0.0190 | 0.1612 | 0.3224 | 0.3224 | nan | 0.3224 | 0.0 | 0.3224 | | 0.013 | 3.4783 | 320 | 0.0184 | 0.2097 | 0.4193 | 0.4193 | nan | 0.4193 | 0.0 | 0.4193 | | 0.0176 | 3.6957 | 340 | 0.0185 | 0.2268 | 0.4537 | 0.4537 | nan | 0.4537 | 0.0 | 0.4537 | | 0.0122 | 3.9130 | 360 | 0.0185 | 0.2039 | 0.4079 | 0.4079 | nan | 0.4079 | 0.0 | 0.4079 | | 0.014 | 4.1304 | 380 | 0.0185 | 0.1942 | 0.3885 | 0.3885 | nan | 0.3885 | 0.0 | 0.3885 | | 0.0105 | 4.3478 | 400 | 0.0185 | 0.2202 | 0.4403 | 0.4403 | nan | 0.4403 | 0.0 | 0.4403 | | 0.0113 | 4.5652 | 420 | 0.0187 | 0.1648 | 0.3296 | 0.3296 | nan | 0.3296 | 0.0 | 0.3296 | | 0.0092 | 4.7826 | 440 | 0.0185 | 0.2044 | 0.4087 | 0.4087 | nan | 0.4087 | 0.0 | 0.4087 | | 0.0175 | 5.0 | 460 | 0.0185 | 0.2160 | 0.4320 | 0.4320 | nan | 0.4320 | 0.0 | 0.4320 | | 0.0124 | 5.2174 | 480 | 0.0190 | 0.2009 | 0.4018 | 0.4018 | nan | 0.4018 | 0.0 | 0.4018 | | 0.0162 | 5.4348 | 500 | 0.0186 | 0.2431 | 0.4863 | 0.4863 | nan | 0.4863 | 0.0 | 0.4863 | | 0.0203 | 5.6522 | 520 | 0.0185 | 0.2091 | 0.4181 | 0.4181 | nan | 0.4181 | 0.0 | 0.4181 | | 0.0172 | 5.8696 | 540 | 0.0190 | 0.1700 | 0.3401 | 0.3401 | nan | 0.3401 | 0.0 | 0.3401 | | 0.014 | 6.0870 | 560 | 0.0189 | 0.1771 | 0.3541 | 0.3541 | nan | 0.3541 | 0.0 | 0.3541 | | 0.0191 | 6.3043 | 580 | 0.0189 | 0.1788 | 0.3575 | 0.3575 | nan | 0.3575 | 0.0 | 0.3575 | | 0.0157 | 6.5217 | 600 | 0.0188 | 0.1986 | 0.3971 | 0.3971 | nan | 0.3971 | 0.0 | 0.3971 | | 0.0128 | 6.7391 | 620 | 0.0187 | 0.2218 | 0.4436 | 0.4436 | nan | 0.4436 | 0.0 | 0.4436 | | 0.0155 | 6.9565 | 640 | 0.0185 | 0.2099 | 0.4198 | 0.4198 | nan | 0.4198 | 0.0 | 0.4198 | | 0.0132 | 7.1739 | 660 | 0.0189 | 0.1909 | 0.3819 | 0.3819 | nan | 0.3819 | 0.0 | 0.3819 | | 0.0142 | 7.3913 | 680 | 0.0185 | 0.1892 | 0.3784 | 0.3784 | nan | 0.3784 | 0.0 | 0.3784 | | 0.0113 | 7.6087 | 700 | 0.0187 | 0.1914 | 0.3828 | 0.3828 | nan | 0.3828 | 0.0 | 0.3828 | | 0.0111 | 7.8261 | 720 | 0.0187 | 0.2136 | 0.4271 | 0.4271 | nan | 0.4271 | 0.0 | 0.4271 | | 0.0094 | 8.0435 | 740 | 0.0188 | 0.1922 | 0.3843 | 0.3843 | nan | 0.3843 | 0.0 | 0.3843 | | 0.0198 | 8.2609 | 760 | 0.0188 | 0.1911 | 0.3822 | 0.3822 | nan | 0.3822 | 0.0 | 0.3822 | | 0.0164 | 8.4783 | 780 | 0.0189 | 0.1896 | 0.3792 | 0.3792 | nan | 0.3792 | 0.0 | 0.3792 | | 0.0222 | 8.6957 | 800 | 0.0186 | 0.2178 | 0.4355 | 0.4355 | nan | 0.4355 | 0.0 | 0.4355 | | 0.0108 | 8.9130 | 820 | 0.0190 | 0.1855 | 0.3710 | 0.3710 | nan | 0.3710 | 0.0 | 0.3710 | | 0.0128 | 9.1304 | 840 | 0.0187 | 0.2006 | 0.4011 | 0.4011 | nan | 0.4011 | 0.0 | 0.4011 | | 0.0143 | 9.3478 | 860 | 0.0187 | 0.2013 | 0.4026 | 0.4026 | nan | 0.4026 | 0.0 | 0.4026 | | 0.0088 | 9.5652 | 880 | 0.0187 | 0.2020 | 0.4040 | 0.4040 | nan | 0.4040 | 0.0 | 0.4040 | | 0.0127 | 9.7826 | 900 | 0.0188 | 0.2023 | 0.4046 | 0.4046 | nan | 0.4046 | 0.0 | 0.4046 | | 0.0109 | 10.0 | 920 | 0.0188 | 0.2008 | 0.4015 | 0.4015 | nan | 0.4015 | 0.0 | 0.4015 | ### Framework versions - Transformers 4.46.3 - Pytorch 2.5.1+cu121 - Datasets 3.2.0 - Tokenizers 0.20.3