Rename suppress_output -> verbose, and verbose -> renorm_warnings

ddm/functions.py

@@ -106,7 +106,7 @@ def fit_model(sample,
               method=None,
               overlay=OverlayNone(),
               lossfunction=LossLikelihood,
-              suppress_output=False,
+              verbose=True,
               name="fit_model",
               verify=False):
     """Fit a model to reaction time data.
@@ -146,8 +146,8 @@ def fit_model(sample,
     may prevent crashes.  If verification is already disabled, this
     does not re-enable it.
     
-    `suppress_output` disables out-of-boundaries warnings and suppresses 
-    printing the model information at each evaluation of the fitness function.
+    `verbose` enables out-of-boundaries warnings and prints the model
+    information at each evaluation of the fitness function.
 
     Returns a "Model()" object with the specified `drift`, `noise`,
     `bound`, `IC`, and `overlay`.
@@ -172,11 +172,11 @@ def fit_model(sample,
     # of the model.
     m = copy.deepcopy(Model(name=name, drift=drift, noise=noise, bound=bound, IC=IC, overlay=overlay, T_dur=T_dur, dt=dt, dx=dx))
     return fit_adjust_model(sample, m, fitparams=fitparams, fitting_method=fitting_method, 
-                            method=method, lossfunction=lossfunction, suppress_output=suppress_output)
+                            method=method, lossfunction=lossfunction, verbose=verbose)
 
 
 def fit_adjust_model(sample, model, fitparams=None, fitting_method="differential_evolution",
-                     lossfunction=LossLikelihood, verify=False, method=None, suppress_output=False):
+                     lossfunction=LossLikelihood, verify=False, method=None, verbose=True):
     """Modify parameters of a model which has already been fit.
     
     The data `sample` should be a Sample object of the reaction times
@@ -215,8 +215,8 @@ def fit_adjust_model(sample, model, fitparams=None, fitting_method="differential
     `method` gives the method used to solve the model, and can be
     "analytical", "numerical", "cn", "implicit", or "explicit".
 
-    `suppress_output` disables out-of-boundaries warnings and suppresses 
-    printing the model information at each evaluation of the fitness function.
+    `verbose` enables out-of-boundaries warnings and prints the model
+    information at each evaluation of the fitness function.
 
     Returns the same model object that was passed to it as an
     argument.  However, the parameters will be modified.  The model is
@@ -233,10 +233,10 @@ def fit_adjust_model(sample, model, fitparams=None, fitting_method="differential
     """
     # Disable paranoid if `verify` is False.
     paranoid_state = paranoid_settings.get('enabled')
-    verbose_state = param.verbose
+    renorm_warnings_state = param.renorm_warnings
     if paranoid_state and not verify:
         paranoid_settings.set(enabled=False)
-        param.verbose = False
+        param.renorm_warnings = False
     # Loop through the different components of the model and get the
     # parameters that are fittable.  Save the "Fittable" objects in
     # "params".  Create a list of functions to set the value of these
@@ -318,16 +318,16 @@ def _fit_model(xs):
             # they will give 1.000000000001.  This fixes that problem
             # to make sure the model is within its domain.
             if x > p.maxval:
-                if not suppress_output:
+                if verbose:
                     print("Warning: optimizer went out of bounds.  Setting %f to %f" % (x, p.maxval))
                 x = p.maxval
             if x < p.minval:
-                if not suppress_output:
+                if verbose:
                     print("Warning: optimizer went out of bounds.  Setting %f to %f" % (x, p.minval))
                 x = p.minval
             s(m, x)
         lossf = lf.loss(m)
-        if not suppress_output:
+        if verbose:
             print(repr(m), "loss="+ str(lossf))
         return lossf
     # Cast to a dictionary if necessary
@@ -359,7 +359,7 @@ def _fit_model(xs):
         s(m, x)
     if not verify:
         paranoid_settings.set(enabled=paranoid_state)
-        param.verbose = verbose_state
+        param.renorm_warnings = renorm_warnings_state
     return m
 
 def evolution_strategy(fitness, x_0, mu=1, lmbda=3, copyparents=True, mutate_var=.002, mutate_prob=.5, evals=100):

ddm/model.py

@@ -482,7 +482,7 @@ def solve_analytical(self, conditions={}):
         # Fix numerical errors
         pdfsum = (np.sum(anal_pdf_corr) + np.sum(anal_pdf_err))*self.dt
         if pdfsum > 1:
-            if pdfsum > 1.01 and param.verbose:
+            if pdfsum > 1.01 and param.renorm_warnings:
                 print("Warning: renormalizing probability density from", pdfsum, "to 1.  " \
                       "Try decreasing dt.  If that doesn't eliminate this warning, it may be due to " \
                       "extreme parameter values and/or bugs in your model speficiation.")
@@ -646,11 +646,11 @@ def solve_numerical(self, method="cn", conditions={}, return_evolution=False):
         if minval < 0:
             sum_negative_strength = np.sum(pdf_corr[pdf_corr<0]) + np.sum(pdf_err[pdf_err<0])
             sum_negative_strength_undec = np.sum(pdf_undec[pdf_undec<0])
-            if sum_negative_strength < -.01 and param.verbose:
+            if sum_negative_strength < -.01 and param.renorm_warnings:
                 print("Warning: probability density included values less than zero "
                       "(minimum=%f, total=%f).  "  \
                       "Please decrease dt and/or avoid extreme parameter values." % (minval, sum_negative_strength))
-            if sum_negative_strength_undec < -.01 and param.verbose:
+            if sum_negative_strength_undec < -.01 and param.renorm_warnings:
                 print("Warning: remaining FP distribution included values less than zero " \
                       "(minimum=%f, total=%f).  " \
                       "Please decrease dt and/or avoid extreme parameter values." % (minval, sum_negative_strength_undec))
@@ -660,7 +660,7 @@ def solve_numerical(self, method="cn", conditions={}, return_evolution=False):
         # Fix numerical errors
         pdfsum = np.sum(pdf_corr) + np.sum(pdf_err) + np.sum(pdf_undec)
         if pdfsum > 1:
-            if pdfsum > 1.01 and param.verbose:
+            if pdfsum > 1.01 and param.renorm_warnings:
                 print("Warning: renormalizing probability density from", pdfsum, "to 1.  " \
                       "Try decreasing dt or using the implicit (backward Euler) method instead.  " \
                       "If that doesn't eliminate this warning, it may be due to " \
@@ -870,7 +870,7 @@ def solve_numerical_cn(self, conditions={}):
         if minval < 0:
             sum_negative_strength = np.sum(pdf_corr[pdf_corr<0]) + np.sum(pdf_err[pdf_err<0])
             # For small errors, don't bother alerting the user
-            if sum_negative_strength < -.01 and param.verbose:
+            if sum_negative_strength < -.01 and param.renorm_warnings:
                 print("Warning: probability density included values less than zero "
                       "(minimum=%f, total=%f).  " \
                       "Please decrease dt and/or avoid extreme parameter values." % (minval, sum_negative_strength))
@@ -880,7 +880,7 @@ def solve_numerical_cn(self, conditions={}):
         pdfsum = np.sum(pdf_corr) + np.sum(pdf_err)
         if pdfsum > 1:
             # If it is only a small renormalization, don't bother alerting the user.
-            if pdfsum > 1.01 and param.verbose:
+            if pdfsum > 1.01 and param.renorm_warnings:
                 print("Warning: renormalizing probability density from", pdfsum, "to 1.  " \
                       "Try decreasing dt.  If that doesn't eliminate this warning, it may be due to " \
                       "extreme parameter values and/or bugs in your model speficiation.")

ddm/parameters.py

@@ -13,4 +13,4 @@
 dt = .005 #0.005 # [s] Time-step.
 
 # Display warnings
-verbose = True
+renorm_warnings = True