transfer adam_forecaster

config-i1 · Feb 13, 2024 · c0b48b2 · c0b48b2
1 parent d1641dc
commit c0b48b2
Show file tree

Hide file tree

Showing 3 changed files with 93 additions and 4 deletions.
diff --git a/python/smooth/adam_general/sma.py b/python/smooth/adam_general/sma.py
@@ -1,6 +1,6 @@
 import numpy as np
 
-from smooth.adam_general._adam_general import adam_fitter
+from smooth.adam_general._adam_general import adam_fitter, adam_forecaster
 from smooth.adam_general.adam_profile import adam_profile_creator
 
 
@@ -75,6 +75,52 @@ def creator_sma(order):
             backcast=True,
         )
 
-        return adam_fitted
+        fitted_args = dict(
+            matrixVt=mat_Vt,
+            matrixWt=mat_Wt,
+            matrixF=mat_F,
+            vectorG=vec_G,
+            lags=lags_model_all,
+            indexLookupTable=index_lookup_table,
+            profilesRecent=profiles_recent_table,
+            E=E_type,
+            T=T_type,
+            S=S_type,
+            nNonSeasonal=components_num_ETS,
+            nSeasonal=components_num_ETS_seasonal,
+            nArima=order,
+            nXreg=xreg_number,
+            constant=constant_required,
+            vectorYt=y_in_sample,
+            vectorOt=ot,
+            backcast=True,
+        )
+
+        return adam_fitted, fitted_args
+
+    sma_fitted, fitted_args = creator_sma(order=order)
+
+    # need to convert some inputs to the expected dtypes. This is a temporary fix.
+    fitted_args["lags"] = np.array(fitted_args["lags"], dtype="uint64")
+    fitted_args["indexLookupTable"] = np.array(
+        fitted_args["indexLookupTable"], dtype="uint64"
+    )
+
+    sma_forecast = adam_forecaster(
+        matrixWt=fitted_args["matrixWt"],
+        matrixF=fitted_args["matrixF"],
+        lags=fitted_args["lags"],
+        indexLookupTable=fitted_args["indexLookupTable"],
+        profilesRecent=sma_fitted["profile"],
+        E=fitted_args["E"],
+        T=fitted_args["T"],
+        S=fitted_args["S"],
+        nNonSeasonal=fitted_args["nNonSeasonal"],
+        nSeasonal=fitted_args["nSeasonal"],
+        nArima=fitted_args["nArima"],
+        nXreg=fitted_args["nXreg"],
+        constant=fitted_args["constant"],
+        horizon=h,
+    )
 
-    return creator_sma(order=order)
+    return sma_forecast
diff --git a/python/smooth/adam_general/test_script.py b/python/smooth/adam_general/test_script.py
@@ -5,4 +5,4 @@
 if __name__ == "__main__":
     y = np.arange(0, 100)
     results = sma(y, order=5)
-    print(results["yFitted"])
+    print(results)
diff --git a/src/python_examples/adamGeneral.cpp b/src/python_examples/adamGeneral.cpp
@@ -159,6 +159,34 @@ py::dict adamFitter(arma::mat &matrixVt, arma::mat const &matrixWt, arma::mat &m
     return result;
 }
 
+/* # Function produces the point forecasts for the specified model */
+arma::vec adamForecaster(arma::mat const &matrixWt, arma::mat const &matrixF,
+                         arma::uvec lags, arma::umat const &indexLookupTable, arma::mat profilesRecent,
+                         char const &E, char const &T, char const &S,
+                         unsigned int const &nNonSeasonal, unsigned int const &nSeasonal,
+                         unsigned int const &nArima, unsigned int const &nXreg, bool const &constant,
+                         unsigned int const &horizon)
+{
+    // unsigned int lagslength = lags.n_rows;
+    unsigned int nETS = nNonSeasonal + nSeasonal;
+    unsigned int nComponents = indexLookupTable.n_rows;
+
+    arma::vec vecYfor(horizon, arma::fill::zeros);
+
+    /* # Fill in the new xt matrix using F. Do the forecasts. */
+    for (unsigned int i = 0; i < horizon; i = i + 1)
+    {
+        vecYfor.row(i) = adamWvalue(profilesRecent(indexLookupTable.col(i)), matrixWt.row(i), E, T, S,
+                                    nETS, nNonSeasonal, nSeasonal, nArima, nXreg, nComponents, constant);
+
+        profilesRecent(indexLookupTable.col(i)) = adamFvalue(profilesRecent(indexLookupTable.col(i)),
+                                                             matrixF, E, T, S, nETS, nNonSeasonal, nSeasonal, nArima, nComponents, constant);
+    }
+
+    // return List::create(Named("matVt") = matrixVtnew, Named("yForecast") = vecYfor);
+    return vecYfor;
+}
+
 PYBIND11_MODULE(_adam_general, m)
 {
     m.doc() = "Adam code"; // module docstring
@@ -184,4 +212,19 @@ PYBIND11_MODULE(_adam_general, m)
         py::arg("vectorYt"),
         py::arg("vectorOt"),
         py::arg("backcast"));
+    m.def("adam_forecaster", &adamForecaster, "forecasts the adam model",
+          py::arg("matrixWt"),
+          py::arg("matrixF"),
+          py::arg("lags"),
+          py::arg("indexLookupTable"),
+          py::arg("profilesRecent"),
+          py::arg("E"),
+          py::arg("T"),
+          py::arg("S"),
+          py::arg("nNonSeasonal"),
+          py::arg("nSeasonal"),
+          py::arg("nArima"),
+          py::arg("nXreg"),
+          py::arg("constant"),
+          py::arg("horizon"));
 }