Leon Sixt

Results 9 issues of Leon Sixt

Hello developers of sacred! First, I want to thank you for this awesome project. It is so helpful for keeping track of your experiments. I currently run my experiments from...

feature

Hi Avanti, I modified how DeepLIFT handles model parameters. The current behaviour is to store them as numpy arrays. I converted them to tensorflow variables. This change resulted in a...

Hi Avanti, for my paper [[1]](https://arxiv.org/abs/1912.09818), I computed the cosine similarity using the values of `self._target_contrib_vars` as it includes both positive and negative contributions. However for the MaxPool2D, `self._target_contrib_vars` was...

There is a problem with installing pysurvival under python 3.9. (Maybe #39 is related) The solution is to recompile the cython module with a recent cython version. I did this...

- [x] fix bibtex - [x] fix pytorch example - [x] fix installation instruction - [x] add tensorflow example - [x] link documentation - [x] include GIF - [x] add...

A few remarks to the PyTorch IBA API: - [x] `IBA.heatmap` rename to `IBA.analyze`. Add flag to switch between `saliency_map` (in bits, channels summed, scaled to input image) or `capacity`...

In the Readout example, a closure is handed into the `analyze` function: https://github.com/BioroboticsLab/IBA/blob/master/notebooks/pytorch_IBA_train_readout.ipynb ``` model_loss_closure = lambda x: -torch.log_softmax(model(x), 1)[:, target].mean() heatmap = iba.analyze(img[None].to(dev), model_loss_closure) ``` This is probably not...

Hi, I want to re-open a bug report: Vim cursor does not move with "Go to Next Problem" #5746 The problem is that after "Go to Next Problem" the vim...