diff options
| -rw-r--r-- | README.md | 6 | ||||
| -rw-r--r-- | lib/parameters.py | 16 |
2 files changed, 14 insertions, 8 deletions
@@ -32,8 +32,10 @@ The following variables may be set to alter the behaviour of dfatool components. | `DFATOOL_DTREE_ENABLED` | 0, **1** | Use decision trees in get\_fitted | | `DFATOOL_DTREE_FUNCTION_LEAVES` | 0, **1** | Use functions (fitted via linear regression) in decision tree leaves when modeling numeric parameters with at least three distinct values. If 0, integer parameters are treated as enums instead. | | `DFATOOL_DTREE_SKLEARN_CART` | **0**, 1 | Use sklearn CART ("Decision Tree Regression") algorithm for decision tree generation. Uses binary nodes and supports splits on scalar variables. Overrides `FUNCTION_LEAVES` (=0) and `NONBINARY_NODES` (=0). | -| `DFATOOL_CART_MAX_DEPTH` | **0** ... *n* | maximum depth for sklearn CART. Default: unlimited. | -| `DFATOOL_USE_XGBOOST` | **0**, 1 | Use Extreme Gradient Boosting algorithm for decision tree generation. | +| `DFATOOL_CART_MAX_DEPTH` | **0** .. *n* | maximum depth for sklearn CART. Default: unlimited. | +| `DFATOOL_USE_XGBOOST` | **0**, 1 | Use Extreme Gradient Boosting algorithm for decision forest generation. | +| `DFATOOL_XGB_N_ESTIMATORS` | 1 .. **100** .. *n* | Number of estimators (i.e., trees) for XGBoost. | +| `DFATOOL_XGB_MAX_DEPTH` | 2 .. **10** ** *n* | Maximum XGBoost tree depth. | | `DFATOOL_KCONF_WITH_CHOICE_NODES` | 0, **1** | Generate enum parameters from kconfig choice nodes; ignore corresponding boolean config options. | | `DFATOOL_KCONF_IGNORE_NUMERIC` | **0**, 1 | Ignore numeric (int/hex) configuration options. Useful for comparison with CART/DECART. | | `DFATOOL_KCONF_IGNORE_STRING` | **0**, 1 | Ignore string configuration options. Useful for comparison with CART/DECART. | diff --git a/lib/parameters.py b/lib/parameters.py index 401c7c6..1239046 100644 --- a/lib/parameters.py +++ b/lib/parameters.py @@ -936,16 +936,20 @@ class ModelAttribute: return if with_xgboost: - from xgboost import XGBRegressor + import xgboost # TODO retrieve parameters from env - xgb = XGBRegressor( - n_estimators=100, - max_depth=10, - eta=0.2, + # <https://xgboost.readthedocs.io/en/stable/python/python_api.html#module-xgboost.sklearn> + # n_estimators := number of trees in forest + # max_depth := maximum tree depth + # eta <=> learning_rate + xgb = xgboost.XGBRegressor( + n_estimators=int(os.getenv("DFATOOL_XGB_N_ESTIMATORS", "100")), + max_depth=int(os.getenv("DFATOOL_XGB_MAX_DEPTH", "10")), + learning_rate=0.2, subsample=0.7, gamma=0.01, - alpha=0.0006, + reg_alpha=0.0006, ) fit_parameters, category_to_index, ignore_index = param_to_ndarray( parameters, with_nan=False, categorial_to_scalar=categorial_to_scalar |