We evaluate submissions using the [surface dice](https://github.com/google-deepmind/surface-distance/tree/master) metric with a tolerance of `0.0`.

You can find the code for the metric in this notebook: [Surface Dice Metric](https://www.kaggle.com/metric/surface-dice-metric).

## Submission File

For each `id` in the test set, you must predict `rle`, a [run-length encoded](https://www.kaggle.com/code/paulorzp/run-length-encode-and-decode/script) instance segmentation mask, where `id` represents `{dataset}_{slice}` for an image with path `test/{dataset}/images/{slice}.tif`. Represent the RLE for an empty mask as `1 0`.

Your submission should contain a header and have the following format:
```
id,rle
kidney_5_0,1 1 100 10
kidney_5_1,1 1 100 10
kidney_6_0,1 0
kidney_6_1,1 0
...
```