Mike Boss

RasterDataset intersection loads multiple files if overlapped

6

As NAIP tiles have an overlap, the `intersection` call in in `__getitem__` returns multiple filepaths if a `patch` is loaded on the edge of a `tile`. However, at least one...

Why Sampler Index intersection with roi

2

Why do the `Sampler`s call `self.index.intersection(tuple(self.roi), objects=True):` (line 116, single.py)? This seems to be called in `GeoSampler` as well which they all derive but with the `dataset.index`.

Multiple roi in Sampler

2

I have an issue with multiple `roi`'s passed to a `Sampler`. I am trying to use the same `train/val/test` set as `Tile2Vec`. In the paper they split a singular `roi`...

Unclear error message in get_rotation_matrix2d

1

### Describe the bug Both device and `dtype` are checked in `get_rotation_matrix2d` for equality of all three tensors but the error message only informs about device (`dtype` is returned but...

### 🐛 Describe the bug When cycle with later caching is used it works the very first time but afterwards it crashes because of the demux because it checks infinitely...

Support pyarrow LargeListType

2

Fixes #6834

### Describe the bug The following code fails because LargeListType is not supported. This is especially a problem for .from_polars since polars uses LargeListType. ### Steps to reproduce the bug...

Make Image cast storage faster

8

PR for issue #6782. Makes `cast_storage` of the `Image` class faster by removing the slow call to `.pylist`. Instead directly convert each `ListArray` item to either `Array2DExtensionType` or `Array3DExtensionType`. This...

Image cast_storage very slow for arrays (e.g. numpy, tensors)

3

Update: see comments below ### Describe the bug Operations that save an image from a path are very slow. I believe the reason for this is that the image data...

Cache only changed columns?

2

### Feature request Cache only the actual changes to the dataset i.e. changed columns. ### Motivation I realized that caching actually saves the complete dataset again. This is especially problematic...