Siem de Jong

My data consists of bitmap (.bmp) images. Is there any good reason only tif and png can be loaded? Otherwise, including bmp in the file checking logic allows for bmp...

**Is your feature request related to a problem? Please describe.** Currently, in the context of pathology, `PatchWSIDataset` accepts ```python [ {"image": "path/to/image1.tiff", "location": [200, 500], "label": 0}, {"image": "path/to/image2.tiff", "location":...

Allow for unnormalized attribution heatmap

2

Currently, when using `captum.attr.visualization.visualize_image_attr`, attributions are always normalized: https://github.com/pytorch/captum/blob/2c9dcc1b31400eaa32ed6a5b0e5ca4c7ec0c3741/captum/attr/_utils/visualization.py#L263. I have a use case where I would like to show the positive and negative signs (so set `sign='all'`), but show...

Add Precision-Recall-Gain curve, Area Under Precision Recall Gain curve, and FGain1 score

5

## 🚀 Feature Add Precision-Recall-Gain (PRG) curve as a new feature with the same interface as the Precision-Recall (PR) curve. Along with PRG, the Area Under the Precision Recall Gain...

Add .bmp support

1

Add bitmap (.bmp) support.

Horizontal scroll

1

Currently, streamlit-pdf-viewer only allows for vertical scrolling, but for some layouts it might be useful to show pages side by side and do horizontal scrolling.

Expose RandomResizedCrop ratio argument

5

Currently, https://github.com/lightly-ai/lightly/blob/d893cb592936d80e3d7b921da13e6f513efebf53/lightly/transforms/dino_transform.py#L219-L224 doesn't allow for changing the aspect ratio argument. If users would like to use the DINO transform but change the aspect ratio of the `RandomResizedCrop`, they would have...

Use torchvision transforms instead of PIL operations for solarization and gaussian blurring

1

**Problem** When using a transform that uses `lightly.transforms.gaussian_blur.GaussianBlur` or `lightly.transforms.solarization.RandomSolarization`, e.g. `DINOTransform`, the input must be transformed to a `PIL.Image` first, because the Gaussian blur and the solarization use `PIL.ImageFilter.GaussianBlur`...

Optionally color paired plots

4

**Is your feature request related to a problem? Please describe.** ```python groups_baseline = dabest.load(df, idx=("1", "2", "3", "4", "5"), paired="baseline", idx="id") ``` doesn't show a colored plot **Describe the solution...

Jitter brightness, contrast, and saturation for multichannel (2/>3) input with v2.ColorJitter

4

### 🚀 The feature Let `v2.ColorJitter` apply brightness, contrast, and saturation jitter on images with 2, or more than 3 channels. This may be specifically enabled via `v2.ColorJitter(allow_multichannel=True)`. ### Motivation,...