|
--- |
|
license: other |
|
base_model: nvidia/mit-b0 |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: image_segmentation_text_v2 |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# image_segmentation_text_v2 |
|
|
|
This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.2273 |
|
- Mean Iou: 0.7888 |
|
- Mean Accuracy: 0.8815 |
|
- Overall Accuracy: 0.9466 |
|
- Per Category Iou: [0.9411436688097194, 0.6365339140286779] |
|
- Per Category Accuracy: [0.9666298272627839, 0.7963078307432842] |
|
|
|
## 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: 2 |
|
- eval_batch_size: 2 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 5 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Per Category Iou | Per Category Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:-------------:|:----------------:|:-----------------------------------------:|:----------------------------------------:| |
|
| 0.6854 | 0.4 | 20 | 0.6497 | 0.5097 | 0.8417 | 0.7443 | [0.7115032459977992, 0.3077993102868545] | [0.7144421527142827, 0.9689304620217858] | |
|
| 0.5904 | 0.8 | 40 | 0.5143 | 0.5981 | 0.8004 | 0.8478 | [0.8333935175312267, 0.3627987944393932] | [0.862384168080342, 0.7383421777739057] | |
|
| 0.4687 | 1.2 | 60 | 0.4546 | 0.6050 | 0.8161 | 0.8493 | [0.8342992363445556, 0.37569323794278087] | [0.8595097956337959, 0.7727067629087438] | |
|
| 0.424 | 1.6 | 80 | 0.3844 | 0.6584 | 0.8272 | 0.8881 | [0.8773645633114495, 0.43953238690068724] | [0.9067963946134603, 0.7476733768251818] | |
|
| 0.3485 | 2.0 | 100 | 0.3535 | 0.6942 | 0.8808 | 0.8995 | [0.8882922581800861, 0.5000111990627714] | [0.9052579907882474, 0.8563377644392374] | |
|
| 0.3402 | 2.4 | 120 | 0.3424 | 0.7021 | 0.9165 | 0.8977 | [0.8850157203428243, 0.5191490567411473] | [0.891946363487535, 0.9410955056545037] | |
|
| 0.3037 | 2.8 | 140 | 0.3071 | 0.7357 | 0.9150 | 0.9173 | [0.9073510172037412, 0.5639900768760655] | [0.9179667207588744, 0.9119966243321158] | |
|
| 0.2669 | 3.2 | 160 | 0.2528 | 0.7659 | 0.8552 | 0.9410 | [0.9353899879693552, 0.5964963563086936] | [0.9673518346317019, 0.7429811204460667] | |
|
| 0.225 | 3.6 | 180 | 0.2523 | 0.7695 | 0.8894 | 0.9375 | [0.9308386361574114, 0.6082593419593063] | [0.9523221012048093, 0.826397566814428] | |
|
| 0.2703 | 4.0 | 200 | 0.2430 | 0.7812 | 0.8943 | 0.9419 | [0.9355827639904635, 0.6268385945303665] | [0.9564666637969618, 0.8320982324555644] | |
|
| 0.2607 | 4.4 | 220 | 0.2352 | 0.7875 | 0.8918 | 0.9448 | [0.9389026387973957, 0.6361696692483538] | [0.9610553391463873, 0.822527121722995] | |
|
| 0.2314 | 4.8 | 240 | 0.2273 | 0.7888 | 0.8815 | 0.9466 | [0.9411436688097194, 0.6365339140286779] | [0.9666298272627839, 0.7963078307432842] | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.35.2 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.15.0 |
|
- Tokenizers 0.15.0 |
|
|
