Added TdCondTdGridDistribution

pyrecest/distributions/conditional/td_cond_td_grid_distribution.py

@@ -0,0 +1,37 @@
+import numpy as np
+import warnings
+from .abstract_conditional_distribution import AbstractConditionalDistribution
+
+class TdCondTdGridDistribution(AbstractConditionalDistribution):
+
+    def multiply(self, other):
+        assert np.all(self.grid == other.grid), "Multiply:IncompatibleGrid: Can only multiply for equal grids."
+        warnings.warn("Multiply:UnnormalizedResult: Multiplication does not yield normalized result.")
+        sdg = self
+        sdg.fvals = sdg.fvals * other.fvals
+        return sdg
+
+    def marginalize_out(self, first_or_second):
+        assert first_or_second in [1, 2], "firstOrSecond must be 1 or 2."
+        raise NotImplementedError("Method is not implemented yet.")
+
+    def fix_dim(self, first_or_second, point):
+        assert first_or_second in [1, 2], "firstOrSecond must be 1 or 2."
+        raise NotImplementedError("Method is not implemented yet.")
+
+    def plot(self):
+        if self.dim > 6:
+            raise ValueError("Can currently only plot for T1, T2, and T3 torus.")
+        raise NotImplementedError("Method is not implemented yet.")
+
+    def plot_interpolated(self):
+        if self.dim > 6:
+            raise ValueError("Can currently only plot for T1, T2, and T3 torus.")
+        raise NotImplementedError("Method is not implemented yet.")
+
+    def get_manifold_size(self):
+        raise ValueError("Not defined for conditional distributions because interpretation may not be 100% obvious.")
+
+    @classmethod
+    def from_function(cls, fun, no_of_grid_points, fun_does_cartesian_product, grid_type, dim):
+        raise NotImplementedError("Method is not implemented yet.")