update the name of the estimator class in all files

README.md

@@ -69,13 +69,13 @@ Install it using `pip install catsim`.
 ```python
 from catsim.initialization import RandomInitializer
 from catsim.selection import MaxInfoSelector
-from catsim.estimation import HillClimbingEstimator
+from catsim.estimation import NumericalSearchEstimator
 from catsim.stopping import MaxItemStopper
 from catsim.simulation import Simulator
 from catsim.cat import generate_item_bank
 initializer = RandomInitializer()
 selector = MaxInfoSelector()
-estimator = HillClimbingEstimator()
+estimator = NumericalSearchEstimator()
 stopper = MaxItemStopper(20)
 Simulator(generate_item_bank(100), 10).simulate(initializer, selector, estimator, stopper)
 ```

catsim/estimation.py

@@ -7,7 +7,7 @@
 from catsim.simulation import Estimator
 
 
-class UnimodalIntervalSearchEstimator(Estimator):
+class NumericalSearchEstimator(Estimator):
     methods = [
         "hillclimbing",
         "ternary",
@@ -22,7 +22,7 @@ class UnimodalIntervalSearchEstimator(Estimator):
     golden_ratio = (1 + 5**0.5) / 2
 
     def __str__(self):
-        return "Unimodal Interval Search Estimator ({})".format(self.__search_method)
+        return "Numerical Search Estimator ({})".format(self.__search_method)
 
     def __init__(
         self,
@@ -56,11 +56,9 @@ def __init__(
                 format(precision)
             )
 
-        if method not in UnimodalIntervalSearchEstimator.methods:
+        if method not in NumericalSearchEstimator.methods:
             raise ValueError(
-                "Parameter 'method' must be one of {}".format(
-                    UnimodalIntervalSearchEstimator.methods
-                )
+                "Parameter 'method' must be one of {}".format(NumericalSearchEstimator.methods)
             )
 
         self._epsilon = float("1e-" + str(precision))
@@ -326,8 +324,8 @@ def _solve_fibonacci(self, b, a, response_vector, items, administered_items):
         return (b + a) / 2
 
     def _solve_golden_section(self, b, a, response_vector, items, administered_items):
-        c = b + (a - b) / UnimodalIntervalSearchEstimator.golden_ratio
-        d = a + (b - a) / UnimodalIntervalSearchEstimator.golden_ratio
+        c = b + (a - b) / NumericalSearchEstimator.golden_ratio
+        d = a + (b - a) / NumericalSearchEstimator.golden_ratio
 
         left_side_ll = irt.log_likelihood(c, response_vector, items[administered_items])
         right_side_ll = irt.log_likelihood(d, response_vector, items[administered_items])
@@ -338,14 +336,14 @@ def _solve_golden_section(self, b, a, response_vector, items, administered_items
             if left_side_ll >= right_side_ll:
                 b = d
                 d = c
-                c = b + (a - b) / UnimodalIntervalSearchEstimator.golden_ratio
+                c = b + (a - b) / NumericalSearchEstimator.golden_ratio
 
                 right_side_ll = left_side_ll
                 left_side_ll = irt.log_likelihood(c, response_vector, items[administered_items])
             else:
                 a = c
                 c = d
-                d = a + (b - a) / UnimodalIntervalSearchEstimator.golden_ratio
+                d = a + (b - a) / NumericalSearchEstimator.golden_ratio
 
                 left_side_ll = right_side_ll
                 right_side_ll = irt.log_likelihood(d, response_vector, items[administered_items])

catsim/plot.py

@@ -195,12 +195,12 @@ def item_exposure(
         from catsim import plot
         from catsim.initialization import RandomInitializer
         from catsim.selection import MaxInfoSelector
-        from catsim.estimation import HillClimbingEstimator
+        from catsim.estimation import NumericalSearchEstimator
         from catsim.stopping import MaxItemStopper
         from catsim.simulation import Simulator
 
         s = Simulator(generate_item_bank(100), 10)
-        s.simulate(RandomInitializer(), MaxInfoSelector(), HillClimbingEstimator(), MaxItemStopper(20))
+        s.simulate(RandomInitializer(), MaxInfoSelector(), NumericalSearchEstimator(), MaxItemStopper(20))
         plot.item_exposure(title='Exposures', simulator=s, hist=True)
         plot.item_exposure(title='Exposures', simulator=s, par='b')
 
@@ -302,12 +302,12 @@ def test_progress(
         from catsim import plot
         from catsim.initialization import RandomInitializer
         from catsim.selection import MaxInfoSelector
-        from catsim.estimation import HillClimbingEstimator
+        from catsim.estimation import NumericalSearchEstimator
         from catsim.stopping import MaxItemStopper
         from catsim.simulation import Simulator
 
         s = Simulator(generate_item_bank(100), 10)
-        s.simulate(RandomInitializer(), MaxInfoSelector(), HillClimbingEstimator(), MaxItemStopper(20))
+        s.simulate(RandomInitializer(), MaxInfoSelector(), NumericalSearchEstimator(), MaxItemStopper(20))
         plot.test_progress(simulator=s, index=0)
         plot.test_progress(simulator=s, index=0, info=True, var=True, see=True)
 

catsim/simulation.py

@@ -396,13 +396,13 @@ def simulate(
 
         >>> from catsim.initialization import RandomInitializer
         >>> from catsim.selection import MaxInfoSelector
-        >>> from catsim.estimation import HillClimbingEstimator
+        >>> from catsim.estimation import NumericalSearchEstimator
         >>> from catsim.stopping import MaxItemStopper
         >>> from catsim.simulation import Simulator
         >>> from catsim.cat import generate_item_bank
         >>> initializer = RandomInitializer()
         >>> selector = MaxInfoSelector()
-        >>> estimator = HillClimbingEstimator()
+        >>> estimator = NumericalSearchEstimator()
         >>> stopper = MaxItemStopper(20)
         >>> Simulator(generate_item_bank(100), 10).simulate(initializer, selector, estimator, stopper)
         """

catsim/testing.py

@@ -27,7 +27,7 @@ def test_cism():
         for test_size in test_sizes:
 
             initializers = [RandomInitializer("uniform", (-5, 5))]
-            estimators = [UnimodalIntervalSearchEstimator()]
+            estimators = [NumericalSearchEstimator()]
             stoppers = [MaxItemStopper(test_size), MinErrorStopper(0.4)]
 
             for initializer in initializers:
@@ -68,8 +68,7 @@ def test_simulations():
             for logistic_model in logistic_models:
                 for initializer in [FixedPointInitializer(0)]:
                     for estimator in [
-                        UnimodalIntervalSearchEstimator(method=m)
-                        for m in UnimodalIntervalSearchEstimator.methods
+                        NumericalSearchEstimator(method=m) for m in NumericalSearchEstimator.methods
                     ]:
                         for stopper in [MaxItemStopper(test_size)]:
                             for selector in finite_selectors:
@@ -129,7 +128,7 @@ def test_plots():
 
     initializer = RandomInitializer()
     selector = MaxInfoSelector()
-    estimator = UnimodalIntervalSearchEstimator()
+    estimator = NumericalSearchEstimator()
     stopper = MaxItemStopper(20)
     s = Simulator(generate_item_bank(100), 10)
     s.simulate(initializer, selector, estimator, stopper, verbose=True)

sphinx/estimation.rst

@@ -57,7 +57,7 @@ necessarily) more efficient estimators.
     thetas = np.arange(-6.,6.,.1)
 
     for estimator in [
-            UnimodalIntervalSearchEstimator(method=m) for m in UnimodalIntervalSearchEstimator.methods
+            NumericalSearchEstimator(method=m) for m in NumericalSearchEstimator.methods
         ]:
         plt.figure()
 

sphinx/examples.rst

@@ -149,7 +149,7 @@ create the other objects…
 
     initializer = RandomInitializer()
     selector = MaxInfoSelector()
-    estimator = HillClimbingEstimator()
+    estimator = NumericalSearchEstimator()
     stopper = MaxItemStopper(20)
 
 catsim provides different options for each of the aforementioned types
@@ -177,7 +177,7 @@ Here we will use an integer.
 
 .. code:: 
 
-    s = Simulator(items, 10, RandomInitializer(), MaxInfoSelector(), HillClimbingEstimator(), MaxItemStopper(50))
+    s = Simulator(items, 10, RandomInitializer(), MaxInfoSelector(), NumericalSearchEstimator(), MaxItemStopper(50))
 
 
 Starting the simulation
@@ -263,7 +263,7 @@ test information to it.
 .. code:: 
 
     examinees = numpy.random.normal(size=10)
-    s = Simulator(items, examinees, RandomInitializer(), MaxInfoSelector(), HillClimbingEstimator(), MinErrorStopper(.3))
+    s = Simulator(items, examinees, RandomInitializer(), MaxInfoSelector(), NumericalSearchEstimator(), MinErrorStopper(.3))
     s.simulate(verbose=True)
     catplot.test_progress(simulator=s,index=0, info=True)
 
@@ -297,7 +297,7 @@ them in order to all examinees.
 
     indexes = numpy.random.choice(items.shape[0], 50, replace=False)
     print('The following items will be applied to the examinees in this order:', indexes)
-    s = Simulator(items, 10, RandomInitializer(), LinearSelector(indexes), HillClimbingEstimator(), MaxItemStopper(50))
+    s = Simulator(items, 10, RandomInitializer(), LinearSelector(indexes), NumericalSearchEstimator(), MaxItemStopper(50))
     s.simulate(verbose=True)
 
 
@@ -352,7 +352,7 @@ applied to the examinee.
     
     initializer = RandomInitializer()
     selector = MaxInfoSelector()
-    estimator = HillClimbingEstimator()
+    estimator = NumericalSearchEstimator()
     stopper = MaxItemStopper(20)
 
 This dummy data means that the examinee has answered items 1435, 3221,

sphinx/readme_body.md

@@ -64,13 +64,13 @@ Install it using `pip install catsim`.
 ```python
 from catsim.initialization import RandomInitializer
 from catsim.selection import MaxInfoSelector
-from catsim.estimation import HillClimbingEstimator
+from catsim.estimation import NumericalSearchEstimator
 from catsim.stopping import MaxItemStopper
 from catsim.simulation import Simulator
 from catsim.cat import generate_item_bank
 initializer = RandomInitializer()
 selector = MaxInfoSelector()
-estimator = HillClimbingEstimator()
+estimator = NumericalSearchEstimator()
 stopper = MaxItemStopper(20)
 Simulator(generate_item_bank(100), 10).simulate(initializer, selector, estimator, stopper)
 ```