Ensemble: Distribution: Add parameters checkers.

src/Model/Ensembles/Distribution.py

@@ -46,6 +46,8 @@ class Distribution(SQLSubModel):
         self._script = script
         self._labels = []
         self._default_values = []
+        self._params_checkers = []
+        self._params_tips = []
 
     @property
     def name(self):
@@ -145,6 +147,21 @@ class Distribution(SQLSubModel):
         if predicate(self, data):
             modifier(self, data)
 
+    def is_valid_parameters(self, parameters):
+        valid = True
+        tips = []
+
+        for i, checker in enumerate(self._params_checkers):
+            ok = checker(parameters)
+            if not ok:
+                tips.append(self._params_tips[i])
+                valid = False
+
+        return (valid, tips)
+
+    def check_params(self, params):
+        return True
+
     def random(self, r):
         raise NotImplementedMethodeError(cls, cls._load)
 
@@ -166,6 +183,11 @@ class Uniform(Distribution):
 
         self._labels = ["low", "high"]
         self._default_values = [20.0, 40.0]
+        self._params_checkers = [self.check_params, self.check_params]
+        self._params_tips = ["uniform_low_tips", "uniform_high_tips"]
+
+    def check_params(self, params):
+        return (params[0] < params[1])
 
     def random(self, params, nb):
         lst = params.copy()
@@ -187,6 +209,11 @@ class Normal(Distribution):
 
         self._labels = ["loc", "scale"]
         self._default_values = [30.0, 5.0]
+        self._params_checkers = [self.check_params, self.check_scale]
+        self._params_tips = ["normal_loc_tips", "normal_scale_tips"]
+
+    def check_scale(self, params):
+        return 0 <= params[1]
 
     def random(self, params, nb):
         lst = params.copy()
@@ -208,6 +235,16 @@ class Binomial(Distribution):
 
         self._labels = ["binomial_n", "binomial_p"]
         self._default_values = [100.0, 0.3]
+        self._params_checkers = [self.check_n, self.check_p]
+        self._params_tips = ["binomial_n_tips", "binomial_p_tips"]
+
+    def check_n(self, params):
+        n = params[0]
+        return (0 <= n) and (n == round(n))
+
+    def check_p(self, params):
+        p = params[1]
+        return 0 <= p <= 1
 
     def random(self, params, nb):
         lst = params.copy()
@@ -229,6 +266,12 @@ class Poisson(Distribution):
 
         self._labels = ["poisson_lam"]
         self._default_values = [5.0]
+        self._params_checkers = [self.check_lam]
+        self._params_tips = ["poisson_lam_tips"]
+
+    def check_lam(self, params):
+        lam = params[0]
+        return 0 <= lam
 
     def random(self, params, nb):
         lst = params.copy()
@@ -251,6 +294,8 @@ class Custom(Distribution):
         self._labels = ["custom"]
         self._type = "custom"
         self._default_values = [""]
+        self._params_checkers = [self.check_params]
+        self._params_tips = ["custom_params_tips"]
 
     def random(self, params, nb):
         try: