-
For some datasets (typically modeling physical properties) one knows that some montone constrain can be applied between a feature and the prediction which can help bring down noise and ensure meaningf…
-
Normally，it is essential for the bins besides missing to be monotonically increasing or decreasing on the badrate for better explainability. For exmple， the bigger a certain feature X1，the better chan…
-
That's basically the summary of https://github.com/ulid/spec/issues/11
Reasons mentioned are e.g.:
- Makes it stateful
- Worst case scenario leads to just one ULID per millisecond
- Can only wor…
-
For whenNext without threshold, results must always increase, if input increases. This has to be checked at runtime. Use a table to store previous inputs/outputs. Use #86 on a new result every and eve…
-
The simple portfolio model from Saltelli et al. (2004) is in the pipeline (see Issue #319). The model is a linear three-dimensional model. The modulus version of the portfolio model focuses on the dev…
-
add support for multinomial xgboost and gbm with monotonicity constraints
-
User would like to have automatic binning and monotonicity constraints available for them in GLM
Example:
x is linear
transform it to x_1, x_2, x_3:
* if x x_1 = 0
* if 0.2 < x < 0.4 -> x_2 = 0.4
* …
-
Hi,
Sorry to bother. I had a problem when understanding the monotonicity of the coral network's predict layer. Here's the only statement I found modifying the bias layer adding to the fc result:
sel…
-
- [ ] *features/burst.py/compute_monotonicity* : might be able to change `monotonicity = np.ones(cycles) * np.nan` so that the np.ones is not being initialized just to be replaced by nans
- [ ] *feat…
-
Thank you for sharing a good project.
When I study your code, I cannot understand why heur_score_table[row] is calculated in a such way in init_tables().
heur_score_table[row] = SCORE_LOST_PENAL…