Aldo
Aldo
From "_Advances in Financial Machine Learning_" book by Marcos López de Prado the implemented version of Combinatorial Cross Validation with Purging and Embargoing  explaining video: https://www.youtube.com/watch?v=hDQssGntmFA
With **contiguous** test sets: ``` cv_orig = GapKFold(n_splits=5, gap_before=1, gap_after=1) for train_index, test_index in cv_orig.split(np.arange(10)): print("TRAIN:", train_index, "TEST:", test_index) ... TRAIN: [3 4 5 6 7 8 9] TEST: [0...
### Description According to some studies reported on papers (like this: https://www.researchgate.net/publication/220804144_Overfitting_in_Wrapper-Based_Feature_Subset_Selection_The_Harder_You_Try_the_Worse_it_Gets), the feature selection methodologies known as Wrapper suffer from overfitting as the number of explored states increases. A...
[LabBuildApp] JupyterLab 4.0.0 [LabBuildApp] Building jupyterlab assets (production, minimized) [LabBuildApp] WARNING | The extension "jupyterlab_onedarkpro" is outdated.
Upgrading keras to version 2.1.x it gives this error: ``` Starting training Epoch 1/3000 Exception: Error when checking target: expected rpn_out_class to have shape (None, None, None, 9) but got...