handle fit errors in cross validation

2 files changed, 26 insertions, 4 deletions

diff --git a/lib/functions.py b/lib/functions.py
index 698d68c..b1477da 100644
--- a/lib/functions.py
+++ b/lib/functions.py
@@ -75,7 +75,7 @@ class ParamFunction:
     error measure.
     """
 
-    def __init__(self, param_function, validation_function, num_vars):
+    def __init__(self, param_function, validation_function, num_vars, repr_str=None):
         """
         Create function object suitable for regression analysis.
 
@@ -97,6 +97,12 @@ class ParamFunction:
         self._param_function = param_function
         self._validation_function = validation_function
         self._num_variables = num_vars
+        self.repr_str = repr_str
+
+    def __repr__(self) -> str:
+        if self.repr_str:
+            return f"ParamFunction<{self.repr_str}>"
+        return f"ParamFunction<{self._param_function}, {self.validation_function}, {self._num_variables}>"
 
     def is_valid(self, arg: float) -> bool:
         """
@@ -691,24 +697,28 @@ class analytic:
                 + reg_param[1] * model_param,
                 lambda model_param: True,
                 2,
+                repr_str="β₀ + β₁ * x",
             ),
             "logarithmic": ParamFunction(
                 lambda reg_param, model_param: reg_param[0]
                 + reg_param[1] * np.log(model_param),
                 lambda model_param: model_param > 0,
                 2,
+                repr_str="β₀ + β₁ * np.log(x)",
             ),
             "logarithmic1": ParamFunction(
                 lambda reg_param, model_param: reg_param[0]
                 + reg_param[1] * np.log(model_param + 1),
                 lambda model_param: model_param > -1,
                 2,
+                repr_str="β₀ + β₁ * np.log(x+1)",
             ),
             "exponential": ParamFunction(
                 lambda reg_param, model_param: reg_param[0]
                 + reg_param[1] * np.exp(model_param),
                 lambda model_param: model_param <= 64,
                 2,
+                repr_str="β₀ + β₁ * np.exp(x)",
             ),
             #'polynomial' : lambda reg_param, model_param: reg_param[0] + reg_param[1] * model_param + reg_param[2] * model_param ** 2,
             "square": ParamFunction(
@@ -716,18 +726,21 @@ class analytic:
                 + reg_param[1] * model_param ** 2,
                 lambda model_param: True,
                 2,
+                repr_str="β₀ + β₁ * x²",
             ),
             "inverse": ParamFunction(
                 lambda reg_param, model_param: reg_param[0]
                 + reg_param[1] / model_param,
                 lambda model_param: model_param != 0,
                 2,
+                repr_str="β₀ + β₁ * 1/x",
             ),
             "sqrt": ParamFunction(
                 lambda reg_param, model_param: reg_param[0]
                 + reg_param[1] * np.sqrt(model_param),
                 lambda model_param: model_param >= 0,
                 2,
+                repr_str="β₀ + β₁ * np.sqrt(x)",
             ),
             # "num0_8": ParamFunction(
             #    lambda reg_param, model_param: reg_param[0]
@@ -759,6 +772,7 @@ class analytic:
                 + reg_param[1] * analytic._safe_log(model_param),
                 lambda model_param: True,
                 2,
+                repr_str="β₀ + β₁ * safe_log(x)",
             )
             functions.pop("inverse")
             functions["safe_inv"] = ParamFunction(
@@ -766,6 +780,7 @@ class analytic:
                 + reg_param[1] * analytic._safe_inv(model_param),
                 lambda model_param: True,
                 2,
+                repr_str="β₀ + β₁ * safe(1/x)",
             )
             functions.pop("sqrt")
             functions["safe_sqrt"] = ParamFunction(
@@ -773,6 +788,7 @@ class analytic:
                 + reg_param[1] * analytic._safe_sqrt(model_param),
                 lambda model_param: True,
                 2,
+                repr_str="β₀ + β₁ * safe_sqrt(x)",
             )
 
         if bool(int(os.getenv("DFATOOL_FIT_LINEAR_ONLY", "0"))):
diff --git a/lib/paramfit.py b/lib/paramfit.py
index 8cfd486..9923e1a 100644
--- a/lib/paramfit.py
+++ b/lib/paramfit.py
@@ -202,9 +202,15 @@ def _try_fits(
                 if function_name not in raw_results:
                     raw_results[function_name] = dict()
                 error_function = param_function.error_function
-                res = optimize.least_squares(
-                    error_function, [0, 1], args=(X, Y), xtol=2e-15
-                )
+                try:
+                    res = optimize.least_squares(
+                        error_function, [0, 1], args=(X, Y), xtol=2e-15
+                    )
+                except FloatingPointError as e:
+                    logger.warning(
+                        f"optimize.least_squares threw '{e}' when fitting {param_function} on {X}, {Y}"
+                    )
+                    continue
                 measures = regression_measures(param_function.eval(res.x, X), Y)
                 raw_results_by_param[other_parameters][function_name] = measures
                 for measure, error_rate in measures.items():