--- license: other base_model: nvidia/segformer-b5-finetuned-cityscapes-1024-1024 tags: - vision - image-segmentation - generated_from_trainer model-index: - name: segformer-b5-cityscapes-finetuned-coastTrain results: [] --- # segformer-b5-cityscapes-finetuned-coastTrain This model is a fine-tuned version of [nvidia/segformer-b5-finetuned-cityscapes-1024-1024](https://huggingface.co./nvidia/segformer-b5-finetuned-cityscapes-1024-1024) on the peldrak/coastTrain dataset. It achieves the following results on the evaluation set: - Loss: 0.4253 - Mean Iou: 0.5585 - Mean Accuracy: 0.6197 - Overall Accuracy: 0.8740 - Accuracy Water: 0.9765 - Accuracy Whitewater: 0.0159 - Accuracy Sediment: 0.6122 - Accuracy Other Natural Terrain: 0.0 - Accuracy Vegetation: 0.9255 - Accuracy Development: 0.8619 - Accuracy Unknown: 0.9457 - Iou Water: 0.8021 - Iou Whitewater: 0.0158 - Iou Sediment: 0.5787 - Iou Other Natural Terrain: 0.0 - Iou Vegetation: 0.8069 - Iou Development: 0.7835 - Iou Unknown: 0.9224 - F1 Score: 0.8596 ## 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: 4 - eval_batch_size: 4 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 20 ### Training results | Training Loss | Epoch | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Accuracy Water | Accuracy Whitewater | Accuracy Sediment | Accuracy Other Natural Terrain | Accuracy Vegetation | Accuracy Development | Accuracy Unknown | Iou Water | Iou Whitewater | Iou Sediment | Iou Other Natural Terrain | Iou Vegetation | Iou Development | Iou Unknown | F1 Score | |:-------------:|:-----:|:----:|:---------------:|:--------:|:-------------:|:----------------:|:--------------:|:-------------------:|:-----------------:|:------------------------------:|:-------------------:|:--------------------:|:----------------:|:---------:|:--------------:|:------------:|:-------------------------:|:--------------:|:---------------:|:-----------:|:--------:| | 1.7218 | 0.16 | 20 | 1.5229 | 0.3639 | 0.4826 | 0.6959 | 0.6530 | 0.0033 | 0.6952 | 0.0067 | 0.8127 | 0.2824 | 0.9249 | 0.5955 | 0.0030 | 0.3817 | 0.0063 | 0.5740 | 0.2453 | 0.7415 | 0.6829 | | 1.3084 | 0.31 | 40 | 1.1275 | 0.4131 | 0.5060 | 0.7566 | 0.7956 | 0.0011 | 0.5967 | 0.0 | 0.9496 | 0.3131 | 0.8858 | 0.6821 | 0.0011 | 0.4697 | 0.0 | 0.6092 | 0.2891 | 0.8406 | 0.7379 | | 1.1538 | 0.47 | 60 | 0.8108 | 0.4722 | 0.5593 | 0.8123 | 0.8802 | 0.0001 | 0.6946 | 0.0 | 0.9169 | 0.5078 | 0.9153 | 0.7833 | 0.0001 | 0.4768 | 0.0 | 0.7188 | 0.4376 | 0.8889 | 0.8002 | | 0.9791 | 0.62 | 80 | 0.6995 | 0.5264 | 0.6143 | 0.8518 | 0.9168 | 0.0000 | 0.7471 | 0.0 | 0.8479 | 0.8453 | 0.9433 | 0.8301 | 0.0000 | 0.5727 | 0.0 | 0.7406 | 0.6249 | 0.9164 | 0.8421 | | 1.0426 | 0.78 | 100 | 0.5931 | 0.5280 | 0.6063 | 0.8523 | 0.8932 | 0.0003 | 0.6361 | 0.0 | 0.9550 | 0.8282 | 0.9309 | 0.8097 | 0.0003 | 0.5481 | 0.0 | 0.7440 | 0.6697 | 0.9243 | 0.8402 | | 0.8008 | 0.93 | 120 | 0.4687 | 0.5485 | 0.6225 | 0.8706 | 0.9263 | 0.0 | 0.7444 | 0.0 | 0.9248 | 0.8212 | 0.9410 | 0.8404 | 0.0 | 0.5924 | 0.0 | 0.7871 | 0.6857 | 0.9337 | 0.8595 | | 1.0298 | 1.09 | 140 | 0.4732 | 0.5527 | 0.6244 | 0.8726 | 0.9421 | 0.0000 | 0.8164 | 0.0 | 0.9047 | 0.7891 | 0.9185 | 0.8289 | 0.0000 | 0.6400 | 0.0 | 0.7976 | 0.6991 | 0.9036 | 0.8617 | | 0.4902 | 1.24 | 160 | 0.3911 | 0.5713 | 0.6310 | 0.8868 | 0.9694 | 0.0 | 0.7543 | 0.0 | 0.9348 | 0.8241 | 0.9344 | 0.8366 | 0.0 | 0.6816 | 0.0 | 0.8102 | 0.7408 | 0.9295 | 0.8744 | | 0.8204 | 1.4 | 180 | 0.4865 | 0.5210 | 0.5894 | 0.8522 | 0.9765 | 0.0 | 0.4534 | 0.0 | 0.9521 | 0.8103 | 0.9336 | 0.7833 | 0.0 | 0.4303 | 0.0 | 0.7921 | 0.7097 | 0.9313 | 0.8322 | | 1.1865 | 1.55 | 200 | 0.3980 | 0.5668 | 0.6352 | 0.8838 | 0.9644 | 0.0000 | 0.7632 | 0.0 | 0.8985 | 0.8816 | 0.9385 | 0.8442 | 0.0000 | 0.6333 | 0.0 | 0.8133 | 0.7431 | 0.9338 | 0.8722 | | 0.5676 | 1.71 | 220 | 0.3955 | 0.5598 | 0.6352 | 0.8750 | 0.9299 | 0.0 | 0.8440 | 0.0 | 0.9085 | 0.8890 | 0.8747 | 0.8160 | 0.0 | 0.6601 | 0.0 | 0.8209 | 0.7499 | 0.8721 | 0.8647 | | 0.9343 | 1.86 | 240 | 0.3969 | 0.5809 | 0.6445 | 0.8944 | 0.9593 | 0.0001 | 0.8201 | 0.0 | 0.9120 | 0.8658 | 0.9539 | 0.8589 | 0.0001 | 0.6678 | 0.0 | 0.8327 | 0.7744 | 0.9326 | 0.8829 | | 0.5811 | 2.02 | 260 | 0.4253 | 0.5585 | 0.6197 | 0.8740 | 0.9765 | 0.0159 | 0.6122 | 0.0 | 0.9255 | 0.8619 | 0.9457 | 0.8021 | 0.0158 | 0.5787 | 0.0 | 0.8069 | 0.7835 | 0.9224 | 0.8596 | ### Framework versions - Transformers 4.37.0 - Pytorch 2.1.2 - Datasets 2.18.0 - Tokenizers 0.15.1