bugfix

Author: mathias-von-ottenbreit

API_REFERENCE_FOR_REGRESSION.md

@@ -17,7 +17,7 @@ Used to randomly split training observations into cv_folds if ***cv_observations
 Determines the loss function used. Allowed values are "mse", "binomial", "poisson", "gamma", "tweedie", "group_mse", "group_mse_cycle","mae", "quantile", "negative_binomial", "cauchy", "weibull" and "custom_function". This is used together with ***link_function***. When ***loss_function*** is "group_mse" then the "group" argument in the ***fit*** method must be provided. In the latter case APLR will try to minimize group MSE when training the model. When using "group_mse_cycle", ***group_mse_cycle_min_obs_in_bin*** controls the minimum amount of observations in each group. For a description of "group_mse_cycle" see ***group_mse_cycle_min_obs_in_bin***. The ***loss_function*** "quantile" is used together with the ***quantile*** constructor parameter. When ***loss_function*** is "custom_function" then the constructor parameters ***calculate_custom_loss_function*** and ***calculate_custom_negative_gradient_function***, both described below, must be provided.
 
 #### link_function (default = "identity")
-Determines how the linear predictor is transformed to predictions. Allowed values are "identity", "logit", "log" and "custom_function". For an ordinary regression model use ***loss_function*** "mse" and ***link_function*** "identity". For logistic regression use ***loss_function*** "binomial" and ***link_function*** "logit". For a multiplicative model use the "log" ***link_function***. The "log" ***link_function*** often works best with a "poisson", "gamma", "tweedie", "negative_binomial" or "weibull" ***loss_function***, depending on the data. The ***loss_function*** "poisson", "gamma", "tweedie", "negative_binomial" or "weibull" should only be used with the "log" ***link_function***. Inappropriate combinations of ***loss_function*** and ***link_function*** may result in a warning message when fitting the model and/or a poor model fit. Please note that values other than "identity" typically require a significantly higher ***m*** (or ***v***) in order to converge. When ***link_function*** is "custom_function" then the constructor parameters ***calculate_custom_transform_linear_predictor_to_predictions_function*** and ***calculate_custom_differentiate_predictions_wrt_linear_predictor_function***, both described below, must be provided.
+Determines how the linear predictor is transformed to predictions. Allowed values are "identity", "logit", "log" and "custom_function". For an ordinary regression model use ***loss_function*** "mse" and ***link_function*** "identity". For logistic regression use ***loss_function*** "binomial" and ***link_function*** "logit". For a multiplicative model use the "log" ***link_function***. The "log" ***link_function*** often works best with a "poisson", "gamma", "tweedie", "negative_binomial" or "weibull" ***loss_function***, depending on the data. The ***loss_function*** "poisson", "gamma", "tweedie", "negative_binomial" or "weibull" should only be used with the "log" ***link_function***. Inappropriate combinations of ***loss_function*** and ***link_function*** may result in a warning message when fitting the model and/or a poor model fit. Please note that values other than "identity" may require a higher ***m*** (or ***v***) in order to converge. When ***link_function*** is "custom_function" then the constructor parameters ***calculate_custom_transform_linear_predictor_to_predictions_function*** and ***calculate_custom_differentiate_predictions_wrt_linear_predictor_function***, both described below, must be provided.
 
 #### n_jobs (default = 0)
 Multi-threading parameter. If ***0*** then uses all available cores for multi-threading. Any other positive integer specifies the number of cores to use (***1*** means single-threading).

cpp/APLRRegressor.h

@@ -1149,7 +1149,7 @@ VectorXd APLRRegressor::calculate_neg_gradient_current_for_group_mse(GroupData &
 VectorXd APLRRegressor::differentiate_predictions_wrt_linear_predictor()
 {
     if (link_function == "logit")
-        return 1.0 / 4.0 * (linear_predictor_current.array() / 2.0).cosh().array().pow(-2);
+        return 10.0 / 4.0 * (linear_predictor_current.array() / 2.0).cosh().array().pow(-2);
     else if (link_function == "log")
     {
         return linear_predictor_current.array().exp();

cpp/tests.cpp

@@ -1055,7 +1055,7 @@ class Tests
         // Model
         APLRRegressor model{APLRRegressor()};
         model.m = 100;
-        model.v = 0.5;
+        model.v = 0.05;
         model.bins = 300;
         model.n_jobs = 0;
         model.loss_function = "binomial";
@@ -1614,7 +1614,7 @@ class Tests
         // Model
         APLRClassifier model{APLRClassifier()};
         model.m = 100;
-        model.v = 0.5;
+        model.v = 0.05;
         model.bins = 300;
         model.n_jobs = 0;
         model.verbosity = 3;
@@ -1684,7 +1684,7 @@ class Tests
         // Model
         APLRClassifier model{APLRClassifier()};
         model.m = 100;
-        model.v = 0.5;
+        model.v = 0.05;
         model.bins = 300;
         model.n_jobs = 0;
         model.verbosity = 3;
@@ -1754,7 +1754,7 @@ class Tests
         // Model
         APLRClassifier model{APLRClassifier()};
         model.m = 100;
-        model.v = 0.5;
+        model.v = 0.05;
         model.bins = 300;
         model.n_jobs = 0;
         model.verbosity = 3;
@@ -1824,7 +1824,7 @@ class Tests
         // Model
         APLRClassifier model{APLRClassifier()};
         model.m = 100;
-        model.v = 0.5;
+        model.v = 0.05;
         model.bins = 300;
         model.n_jobs = 0;
         model.verbosity = 3;
@@ -1894,7 +1894,7 @@ class Tests
         // Model
         APLRClassifier model{APLRClassifier()};
         model.m = 100;
-        model.v = 0.5;
+        model.v = 0.05;
         model.bins = 300;
         model.n_jobs = 0;
         model.verbosity = 3;
@@ -1962,7 +1962,7 @@ class Tests
         // Model
         APLRClassifier model{APLRClassifier()};
         model.m = 100;
-        model.v = 0.5;
+        model.v = 0.05;
         model.bins = 300;
         model.n_jobs = 0;
         model.verbosity = 3;
@@ -2020,15 +2020,15 @@ class Tests
 
         std::cout << "cv_error\n"
                   << model.get_cv_error() << "\n\n";
-        tests.push_back(is_approximately_equal(model.get_cv_error(), 0.15984656957508173, 0.000001));
+        tests.push_back(is_approximately_equal(model.get_cv_error(), 0.15942686880196807, 0.000001));
 
         std::cout << "predicted_class_prob_mean\n"
                   << predicted_class_probabilities.mean() << "\n\n";
         tests.push_back(is_approximately_equal(predicted_class_probabilities.mean(), 0.5, 0.00001));
 
         std::cout << "local_feature_importance_mean\n"
                   << local_feature_contribution.mean() << "\n\n";
-        tests.push_back(is_approximately_equal(local_feature_contribution.mean(), 0.052181259967961045, 0.00001));
+        tests.push_back(is_approximately_equal(local_feature_contribution.mean(), 0.05891072116542774, 0.00001));
         tests.push_back(base_predictors_in_the_second_affiliation == correct_base_predictors_in_the_second_affiliation);
         tests.push_back(the_second_unique_term_affiliation == the_correct_second_unique_term_affiliation);
     }
@@ -2038,7 +2038,7 @@ class Tests
         // Model
         APLRClassifier model{APLRClassifier()};
         model.m = 100;
-        model.v = 0.5;
+        model.v = 0.05;
         model.bins = 300;
         model.n_jobs = 0;
         model.verbosity = 3;
@@ -2092,23 +2092,23 @@ class Tests
 
         std::cout << "cv_error\n"
                   << model.get_cv_error() << "\n\n";
-        tests.push_back(is_approximately_equal(model.get_cv_error(), 0.17250319103503037, 0.000001));
+        tests.push_back(is_approximately_equal(model.get_cv_error(), 0.14420733842494515, 0.000001));
 
         std::cout << "predicted_class_prob_mean\n"
                   << predicted_class_probabilities.mean() << "\n\n";
         tests.push_back(is_approximately_equal(predicted_class_probabilities.mean(), 0.5, 0.00001));
 
         std::cout << "local_feature_importance_mean\n"
                   << local_feature_contribution.mean() << "\n\n";
-        tests.push_back(is_approximately_equal(local_feature_contribution.mean(), 0.07920242388299352, 0.00001));
+        tests.push_back(is_approximately_equal(local_feature_contribution.mean(), 0.10357828243742498, 0.00001));
     }
 
     void test_aplrclassifier_two_class_max_terms()
     {
         // Model
         APLRClassifier model{APLRClassifier()};
         model.m = 100;
-        model.v = 0.5;
+        model.v = 0.05;
         model.bins = 300;
         model.n_jobs = 0;
         model.verbosity = 3;

documentation/APLR 10.5.0.pdf renamed to documentation/APLR 10.5.1.pdf

@@ -27,7 +27,7 @@
 
 setuptools.setup(
     name="aplr",
-    version="10.5.0",
+    version="10.5.1",
     description="Automatic Piecewise Linear Regression",
     ext_modules=[sfc_module],
     author="Mathias von Ottenbreit",