added parameter copy

src/scgenerator/parameter.py

@@ -3,7 +3,7 @@ from __future__ import annotations
 import datetime as datetime_module
 import json
 import os
-from copy import copy
+from copy import copy, deepcopy
 from dataclasses import dataclass, field, fields
 from functools import lru_cache, wraps
 from math import isnan
@@ -468,9 +468,12 @@ class Parameters:
             )
         object.__setattr__(self, k, v)
 
+    def copy(self) -> Parameters:
+        return Parameters(**deepcopy(self.strip_params_dict()))
+
     def to_json(self) -> str:
         d = self.dump_dict()
-        return json.dumps(d, cls=DatetimeEncoder)
+        return json.dumps(d, cls=DatetimeEncoder, default=list)
 
     def get_evaluator(self):
         evaluator = Evaluator.default(self.full_field)
@@ -478,7 +481,7 @@ class Parameters:
         return evaluator
 
     def dump_dict(self, add_metadata=True) -> dict[str, Any]:
-        param = Parameters.strip_params_dict()
+        param = self.strip_params_dict()
         if add_metadata:
             param["datetime"] = datetime_module.datetime.now()
             param["version"] = __version__
@@ -579,15 +582,15 @@ class Parameters:
         max_right = max(len(el[1]) for el in p_pairs)
         return "\n".join("{:>{l}} = {:{r}}".format(*p, l=max_left, r=max_right) for p in p_pairs)
 
-    def strip_params_dict(self) -> dict[str, Any]:
+    def strip_params_dict(self, copy=False) -> dict[str, Any]:
         """
         prepares a dictionary for serialization. Some keys may not be preserved
         (dropped because they take a lot of space and can be exactly reconstructed)
 
         Parameters
         ----------
-        dico : dict
-            dictionary
+        copy : bool, optional
+            whether to deepcopy each value, by default False
         """
         forbiden_keys = {
             "_param_dico",
@@ -608,25 +611,19 @@ class Parameters:
             "linear_op",
             "c_to_a_factor",
         }
-        types = (np.ndarray, float, int, str, list, tuple, dict, Path)
+        types = (np.ndarray, float, int, str, list, tuple, Path)
+        c = deepcopy if copy else lambda x: x
         out = {}
         for key, value in self._param_dico.items():
             if key in forbiden_keys or key not in self._p_names:
                 continue
             if not isinstance(value, types):
                 continue
-            if isinstance(value, dict):
-                out[key] = Parameters.strip_params_dict(value)
             elif isinstance(value, Path):
                 out[key] = str(value)
             elif isinstance(value, np.ndarray) and value.dtype == complex:
                 continue
-            elif isinstance(value, np.ndarray):
-                out[key] = value.tolist()
             else:
-                out[key] = value
-
-        if "variable" in out and len(out["variable"]) == 0:
-            del out["variable"]
+                out[key] = c(value)
 
         return out