Alexander Popkov

Hi, in WIndows OS case check that: - MSVC++ is in Path environment variable - 80 lines of further output doesn't contain errors of finding C/C++ headers

Hi, @jeongyoonlee , Thanks for the review, I pushed necessary changes. I faced the thing that `min_impurity_split` removal in latest scikit-learn breaks tree building. It turned out that new version...

Hi, @jeongyoonlee, It seems that now everything is fixed :ok_hand:.

Hi, `CausalTreeRegressor` and `CausalRandomForestRegressor` in #522 could be interpreted directly via `shap` (`shap.TreeExplainer`) with minor package update. See [example](https://github.com/alexander-pv/causalml/blob/feature/causal_trees_interpretation/examples/causal_trees_interpretation.ipynb). I could open new PR for this feature and PR for...

Hi, the related PR https://github.com/uber/causalml/pull/536 is open now.

Hi, @VeeranjaneyuluToka , I made tests with 256, 512 image sizes. So there is a lot of research to make. You can change input size in src/common/config.py as well as...

Hi, @VeeranjaneyuluToka, Could you please, provide more information about the issue to speedup the process: 1) What's happening with losses according to tensorboard? 2) What images were loaded in the...

Hi, @VeeranjaneyuluToka, Thanks for the information! I conducted a small review of the repository after a long absence, updated the general structure of the model class, and added epsilons, where...

Hi, @b-nils, There are `FPNClassifier` and `FPNMaskLayer` in the subclassed version of the model which are classes inheriting from `tf.keras.layers.Layer`. AFAIR, the reason is that fpn classifier and fpn mask...

Hi, @b-nils, thanks, I can also agree with @VeeranjaneyuluToka that with tf.data.Dataset, data processing is faster. However, in Sequence, you can just increase the queue size for the purpose. I...