Amund Vedal

Results 13 issues of Amund Vedal

## 🐛 Bug Using the wrong type for input `np.array` leads to hard-to-understand error message ## To Reproduce ``` import numpy as np import albumentations as A image = np.zeros((32,32)).astype('int32')...

enhancement

Hi, thanks for a great and useful library! Is there a way to ensure that no keypoints are dropped, similar to `RandomSizedBBoxSafeCrop` for bboxes? I tried making a bbox by...

bug

## 📚 Documentation The behavior of [clipping](https://github.com/albumentations-team/albumentations/blob/89a675cbfb2b76f6be90e7049cd5211cb08169a5/albumentations/augmentations/functional.py#L1246) values to [0,1] after `GaussNoise` and _some_ other similar functions should be documented under each transform. In the current state of the library,...

Thanks a lot for tika-python. its fast and awesome! 🥇 I suggest the following change to make the command line tool `$ tika-python parse all file.pdf` behave more similarly to...

In the current version `quantus==0.1.4`, the following: ```python import torch from torch import nn from torchvision import models import quantus class Net(nn.Module): def __init__(self): super().__init__() resnet18 = models.resnet18() children =...

Hi! Thanks for this very useful extension! Is it possible to click a shortcut to see next image in same folder? Would be very useful. Best

Currently, `Quantus` [converts](https://github.com/understandable-machine-intelligence-lab/Quantus/blob/daa9ba313f97a5ca52a2125563399461ca5d3bf1/quantus/helpers/explanation_func.py#L391) explanations to `numpy` arrays. This is understandable, but maybe it could be optional with a default-value kwarg `convert_to_numpy=True`, such that its possible to work with tensors throughout?

Sorry for all the questions. Please feel free to ignore it. I wonder if its possible to define, and then override/extend, yaml defaults inside other yamls. This is supported in...

enhancement
question

## ❓ Questions and Help In most docstrings of captum attribution methods, there is an example where the model is said to return class probabilities: ``` >>> # ImageClassifier takes...