Simon Graham

Hi! Thanks for the great effort with this PR. Is there any update on getting it verified and merged? Also, in the `GNNExplainer` class there is an option to select...

Hi @jjhbw , Thanks a lot for this - we really appreciate it. We are pleased that you and your lab are enjoying working with the code. Also thanks for...

Hi, to give you a proper diagnosis you would need to exactly reproduce the setup that we use. I.e we use a batch size of 8 for the first 50...

How long did it take to extract patches? Maybe it didn't actually extract any patches and it just created the directories. Make sure the path to your original data is...

Hi, We have provided two scripts ([run_tile.sh](https://github.com/vqdang/hover_net/blob/master/run_tile.sh) and [run_wsi.sh](https://github.com/vqdang/hover_net/blob/master/run_wsi.sh)) that can be used to guide you how to perform inference. We have also provided detailed instructions on the [README](https://github.com/vqdang/hover_net#readme) page....

As far as I am aware, `h5` weights are usually used in a Keras framework. Can you tell me where you got these `h5` weights from?

Miscellaneous is a very noisy class. Sometimes you will get some unexpected behaviour. Do you have the dice over time during training? You should have this in your tensorboard output.

There are different ways of doing it. You can grab a tile from the WSI and plot the contours like we show in `examples/usage.ipynb`. Alternatively there are other approaches to...

Looks like there is a problem with the formatting of the string to the data path

Hi - this error normally comes from the path to the data not being correct. Can you please double check the path to your data. This is specified in the...