feat: completes overhaul to v0.5.0

BREAKING CHANGE: mostly all APIs affected, resolves #23, #20, #19, #17, #16, #12, #11, #10, and JANUS-Institute/HallThrusterPEM#6

src/amisc/__init__.py

@@ -98,16 +98,17 @@ class Compression {
 We recommend using the built-in `YamlLoader` for this purpose, as it includes custom YAML tags for reading/writing
 `amisc` objects from file.
 """
-from abc import ABC as _ABC, abstractmethod as _abstractmethod
+from abc import ABC as _ABC
+from abc import abstractmethod as _abstractmethod
 from pathlib import Path as _Path
 from typing import Any as _Any
 
 import yaml as _yaml
 
-from amisc.variable import Variable, VariableList
 from amisc.component import Component
 from amisc.system import System
 from amisc.utils import to_model_dataset, to_surrogate_dataset
+from amisc.variable import Variable, VariableList
 
 __version__ = "0.4.0"
 __all__ = ['System', 'Component', 'Variable', 'VariableList', 'FileLoader', 'YamlLoader',

src/amisc/component.py

@@ -39,13 +39,21 @@
 from pydantic import BaseModel, ConfigDict, ValidationInfo, field_validator
 from typing_extensions import TypedDict
 
-from amisc.interpolator import InterpolatorState, Interpolator, Lagrange
-from amisc.serialize import YamlSerializable, PickleSerializable, Serializable, StringSerializable
-from amisc.training import TrainingData, SparseGrid
-from amisc.typing import MultiIndex, Dataset, LATENT_STR_ID
-from amisc.utils import get_logger, format_inputs, format_outputs, search_for_file, to_model_dataset, \
-    to_surrogate_dataset
-from amisc.utils import _get_yaml_path, _inspect_assignment, _inspect_function
+from amisc.interpolator import Interpolator, InterpolatorState, Lagrange
+from amisc.serialize import PickleSerializable, Serializable, StringSerializable, YamlSerializable
+from amisc.training import SparseGrid, TrainingData
+from amisc.typing import LATENT_STR_ID, Dataset, MultiIndex
+from amisc.utils import (
+    _get_yaml_path,
+    _inspect_assignment,
+    _inspect_function,
+    format_inputs,
+    format_outputs,
+    get_logger,
+    search_for_file,
+    to_model_dataset,
+    to_surrogate_dataset,
+)
 from amisc.variable import Variable, VariableList
 
 __all__ = ["ModelKwargs", "StringKwargs", "IndexSet", "MiscTree", "Component"]
@@ -706,15 +714,16 @@ def _match_index_set(self, index_set, misc_coeff):
                 case 'test':
                     misc_coeff = self.misc_coeff_test
                 case other:
-                    raise ValueError(f"Index set must be 'train' or 'test' if you do not provide `misc_coeff`.")
+                    raise ValueError(f"Index set must be 'train' or 'test' if you do not provide `misc_coeff`. "
+                                     f"{other} not recognized.")
         if isinstance(index_set, str):
             match index_set:
                 case 'train':
                     index_set = self.active_set
                 case 'test':
                     index_set = self.active_set.union(self.candidate_set)
                 case other:
-                    raise ValueError(f"Index set must be 'train' or 'test'.")
+                    raise ValueError(f"Index set must be 'train' or 'test'. {other} not recognized.")
 
         return index_set, misc_coeff
 
@@ -868,7 +877,7 @@ def call_model(self, inputs: dict | Dataset,
                             ret = (ret,) if not isinstance(ret, tuple) else ret
                             ret = {out_var.name: ret[i] for i, out_var in enumerate(self.outputs)}
                         results.append(ret)
-                    except Exception as e:
+                    except Exception:
                         results.append({'inputs': {k: v[i] for k, v in inputs.items()}, 'index': i,
                                         'model_kwargs': kwargs.copy(), 'error': traceback.format_exc()})
             else:  # Parallel
@@ -893,7 +902,7 @@ def call_model(self, inputs: dict | Dataset,
                             ret = (ret,) if not isinstance(ret, tuple) else ret
                             ret = {out_var.name: ret[i] for i, out_var in enumerate(self.outputs)}
                         results.append(ret)
-                    except Exception as e:
+                    except Exception:
                         results.append({'inputs': {k: v[i] for k, v in inputs.items()}, 'index': i,
                                         'model_kwargs': kwargs.copy(), 'error': traceback.format_exc()})
 

src/amisc/distribution.py

@@ -14,7 +14,7 @@
 """
 from __future__ import annotations
 
-from abc import abstractmethod, ABC
+from abc import ABC, abstractmethod
 
 import numpy as np
 

src/amisc/examples/tutorial.py

@@ -28,7 +28,7 @@ def fun2(y):
 
 def single_component():
     # --8<-- [start:single]
-    from amisc import Variable, Component, System
+    from amisc import Component, System, Variable
 
     def fun(inputs):
         return {'y': inputs['x'] ** 2}
@@ -69,7 +69,7 @@ def field_quantity():
     # --8<-- [start:field_qty]
     import numpy as np
 
-    from amisc import Variable, Component, System, to_model_dataset
+    from amisc import Component, System, Variable, to_model_dataset
     from amisc.compression import SVD
 
     def my_model(inputs):

src/amisc/interpolator.py

@@ -18,8 +18,8 @@
 
 import numpy as np
 
-from amisc.serialize import Serializable, Base64Serializable, StringSerializable
-from amisc.typing import MultiIndex, Dataset
+from amisc.serialize import Base64Serializable, Serializable, StringSerializable
+from amisc.typing import Dataset, MultiIndex
 
 __all__ = ["InterpolatorState", "LagrangeState", "Interpolator", "Lagrange"]
 
@@ -120,7 +120,7 @@ def from_dict(cls, config: dict) -> Interpolator:
             case 'lagrange':
                 return Lagrange(**config)
             case other:
-                raise NotImplementedError(f"Unknown interpolator method: {method}")
+                raise NotImplementedError(f"Unknown interpolator method: {other}")
 
 
 @dataclass
@@ -401,7 +401,8 @@ def hessian(self, x: Dataset, state: LagrangeState, training_data: tuple[Dataset
                                 curr_j_idx = div_zero_idx[..., k, j[k]]
                                 other_j_idx = np.any(other_pts[..., k, :], axis=-1)
                                 dLJ_dx[curr_j_idx] = -np.nansum((w_j[k, p_idx] / w_j[k, j[k]]) /
-                                                                (x_arr[curr_j_idx, k, np.newaxis] - x_j[k, p_idx]), axis=-1)
+                                                                (x_arr[curr_j_idx, k, np.newaxis] - x_j[k, p_idx]),
+                                                                axis=-1)
                                 dLJ_dx[other_j_idx] = ((w_j[k, j[k]] / w_j_large[other_pts[..., k, :]]) /
                                                        (x_arr[other_j_idx, k] - x_j[k, j[k]]))
 
@@ -430,11 +431,11 @@ def hessian(self, x: Dataset, state: LagrangeState, training_data: tuple[Dataset
 
                                 # if these points are at the current j interpolation point
                                 d2LJ_dx2[curr_j_idx] = (2 * np.nansum((w_j[m, p_idx] / w_j[m, j[m]]) /
-                                                                      (x_arr[curr_j_idx, m, np.newaxis] - x_j[m, p_idx]),
-                                                                      axis=-1) ** 2 +  # noqa: E501
+                                                                      (x_arr[curr_j_idx, m, np.newaxis] - x_j[m, p_idx]), # noqa: E501
+                                                                      axis=-1) ** 2 +
                                                         2 * np.nansum((w_j[m, p_idx] / w_j[m, j[m]]) /
-                                                                      (x_arr[curr_j_idx, m, np.newaxis] - x_j[m, p_idx]) ** 2,
-                                                                      axis=-1))  # noqa: E501
+                                                                      (x_arr[curr_j_idx, m, np.newaxis] - x_j[m, p_idx]) ** 2, # noqa: E501
+                                                                      axis=-1))
 
                                 # if these points are at any other interpolation point
                                 other_pts_inv = other_pts.copy()

src/amisc/system.py

@@ -31,11 +31,11 @@
 import random
 import string
 import time
-from collections import ChainMap, deque, UserList
-from concurrent.futures import Executor, wait, ALL_COMPLETED
+from collections import ChainMap, UserList, deque
+from concurrent.futures import ALL_COMPLETED, Executor, wait
 from datetime import timezone
 from pathlib import Path
-from typing import ClassVar, Annotated, Optional, Callable, Literal
+from typing import Annotated, Callable, ClassVar, Literal, Optional
 
 import matplotlib.pyplot as plt
 import networkx as nx
@@ -46,10 +46,18 @@
 
 from amisc.component import Component, IndexSet, MiscTree
 from amisc.serialize import Serializable, _builtin
-from amisc.utils import get_logger, format_inputs, format_outputs, constrained_lls, relative_error, to_model_dataset, \
-    to_surrogate_dataset, _combine_latent_arrays
+from amisc.typing import LATENT_STR_ID, Dataset, MultiIndex, TrainIteration
+from amisc.utils import (
+    _combine_latent_arrays,
+    constrained_lls,
+    format_inputs,
+    format_outputs,
+    get_logger,
+    relative_error,
+    to_model_dataset,
+    to_surrogate_dataset,
+)
 from amisc.variable import VariableList
-from amisc.typing import Dataset, TrainIteration, MultiIndex, LATENT_STR_ID
 
 __all__ = ['TrainHistory', 'System']
 
@@ -650,8 +658,8 @@ def fit(self, targets: list = None,
                         var.update_domain(var.denormalize(new_domain), override=True)
                 del y_samples
             else:
-                self.logger.warning(f'Could not estimate bounds for coupling variables: no test set provided. '
-                                    f'Make sure you manually provide (good) coupling variable domains.')
+                self.logger.warning('Could not estimate bounds for coupling variables: no test set provided. '
+                                    'Make sure you manually provide (good) coupling variable domains.')
 
         # Track convergence progress on the error indicator and test set (plot to file)
         if self.root_dir is not None:
@@ -1153,8 +1161,8 @@ def _end_conditions_met():
                         start_idx = 0
                         for j, var in enumerate(coupling_prev):
                             end_idx = start_idx + xdims[j]
-                            coupling_snap[:, start_idx:end_idx, i] = coupling_iter[var][samples.curr_idx, ...].reshape((N_curr, -1))
-                            res_snap[:, start_idx:end_idx, i] = residual_iter[var][samples.curr_idx, ...].reshape((N_curr, -1))
+                            coupling_snap[:, start_idx:end_idx, i] = coupling_iter[var][samples.curr_idx, ...].reshape((N_curr, -1))  # noqa: E501
+                            res_snap[:, start_idx:end_idx, i] = residual_iter[var][samples.curr_idx, ...].reshape((N_curr, -1))  # noqa: E501
                             start_idx = end_idx
                     C = np.ones((N_curr, 1, mk))
                     b = np.zeros((N_curr, N_couple, 1))
@@ -1164,7 +1172,7 @@ def _end_conditions_met():
                     start_idx = 0
                     for j, var in enumerate(coupling_prev):
                         end_idx = start_idx + xdims[j]
-                        coupling_prev[var][samples.curr_idx, ...] = coupling_new[:, start_idx:end_idx].reshape((N_curr, *var_shapes[j]))
+                        coupling_prev[var][samples.curr_idx, ...] = coupling_new[:, start_idx:end_idx].reshape((N_curr, *var_shapes[j]))  # noqa: E501
                         start_idx = end_idx
                     k += 1
 

src/amisc/training.py

@@ -11,18 +11,18 @@
 import copy
 import itertools
 from abc import ABC, abstractmethod
-from dataclasses import field, dataclass
+from dataclasses import dataclass, field
 from typing import Any, ClassVar
 
 import numpy as np
+from numpy.typing import ArrayLike
 from scipy.optimize import direct
 from sklearn.linear_model import Ridge
 from sklearn.pipeline import Pipeline
 from sklearn.preprocessing import MaxAbsScaler
-from numpy.typing import ArrayLike
 
 from amisc.serialize import PickleSerializable, Serializable
-from amisc.typing import MultiIndex, Dataset, LATENT_STR_ID
+from amisc.typing import LATENT_STR_ID, Dataset, MultiIndex
 
 __all__ = ['TrainingData', 'SparseGrid']
 
@@ -124,7 +124,7 @@ def from_dict(cls, config: dict) -> TrainingData:
             case 'sparse-grid':
                 return SparseGrid(**config)
             case other:
-                raise NotImplementedError(f"Unknown training data method: {method}")
+                raise NotImplementedError(f"Unknown training data method: {other}")
 
 
 @dataclass
@@ -245,7 +245,7 @@ def set(self, alpha: MultiIndex, beta: MultiIndex, coords: list, yi_dict: dict[s
             new_yi = {}
             for var, yi in yi_dict.items():
                 yi = np.atleast_1d(yi[i])
-                new_yi[var] = (float(yi[0]) if self._is_numeric(yi) else yi[0]) if self._is_singleton(yi) else yi.tolist()
+                new_yi[var] = (float(yi[0]) if self._is_numeric(yi) else yi[0]) if self._is_singleton(yi) else yi.tolist()  # noqa: E501
             self.yi_map[alpha][coord] = copy.deepcopy(new_yi)
 
     def impute_missing_data(self, alpha: MultiIndex, beta: MultiIndex):
@@ -419,7 +419,7 @@ def _numeric_outputs(cls, yi_dict: dict[str, ArrayLike]) -> list[str]:
             try:
                 if cls._is_numeric(np.atleast_1d(yi_dict[var])):
                     output_vars.append(var)
-            except:
+            except Exception:
                 continue
         return output_vars
 
@@ -507,6 +507,6 @@ def collocation_1d(N: int, z_bds: tuple, z_pts: np.ndarray = None,
                     z_star = res.x
                     z_pts = np.concatenate((z_pts, z_star))
             case other:
-                raise NotImplementedError(f"Unknown collocation method: {method}")
+                raise NotImplementedError(f"Unknown collocation method: {other}")
 
         return z_pts

src/amisc/transform.py

@@ -49,7 +49,8 @@ def from_string(cls, transform_spec: str | list[str]) -> list[Transform] | None:
                        default, `(a, b)` is the Variable's domain and `(l, u)` is `(0, 1)`. Use simply as `minmax`
                        to use all defaults.
         - **zscore** — $x_{norm} = \\frac{x - m}{s}$ specified as `zscore(m, s)` or `zscore(mu=m, std=s)`. If the
-                       Variable is specified as `distribution=normal`, then `zscore` defaults to the Variable's own `mu, std`.
+                       Variable is specified as `distribution=normal`, then `zscore` defaults to the Variable's
+                       own `mu, std`.
 
         !!! Example
             ```python

src/amisc/typing.py

@@ -12,11 +12,9 @@
 from pathlib import Path as _Path
 from typing import Optional as _Optional
 
-from typing_extensions import TypedDict as _TypedDict
-
 import numpy as _np
 from numpy.typing import ArrayLike as _ArrayLike
-
+from typing_extensions import TypedDict as _TypedDict
 
 __all__ = ["MultiIndex", "Dataset", "TrainIteration", "CompressionData", "LATENT_STR_ID"]
 

src/amisc/utils.py

@@ -21,15 +21,18 @@
 import re
 import sys
 from pathlib import Path
+from typing import TYPE_CHECKING
 
 import numpy as np
 import yaml
 
 __all__ = ['parse_function_string', 'relative_error', 'get_logger', 'format_inputs', 'format_outputs',
            'search_for_file', 'constrained_lls', 'to_surrogate_dataset', 'to_model_dataset']
 
-import amisc.variable
-from amisc.typing import Dataset, LATENT_STR_ID
+from amisc.typing import LATENT_STR_ID, Dataset
+
+if TYPE_CHECKING:
+    import amisc.variable
 
 LOG_FORMATTER = logging.Formatter(u"%(asctime)s — [%(levelname)s] — %(name)-15s — %(message)s")
 
@@ -44,7 +47,7 @@ def _combine_latent_arrays(arr):
             del arr[var]
 
 
-def to_surrogate_dataset(dataset: Dataset, variables: amisc.variable.VariableList, del_fields: bool = True,
+def to_surrogate_dataset(dataset: Dataset, variables: 'amisc.variable.VariableList', del_fields: bool = True,
                          **field_coords) -> tuple[Dataset, list[str]]:
     """Convert true model input/output dataset to a form usable by the surrogate. Primarily, compress field
     quantities and normalize.
@@ -80,7 +83,7 @@ def to_surrogate_dataset(dataset: Dataset, variables: amisc.variable.VariableLis
     return dataset, surr_vars
 
 
-def to_model_dataset(dataset: Dataset, variables: amisc.variable.VariableList, del_latent: bool = True,
+def to_model_dataset(dataset: Dataset, variables: 'amisc.variable.VariableList', del_latent: bool = True,
                      **field_coords) -> tuple[Dataset, Dataset]:
     """Convert surrogate input/output dataset to a form usable by the true model. Primarily, reconstruct
     field quantities and denormalize.
@@ -181,7 +184,7 @@ def visit_Return(self, node):
         return_visitor.visit(tree)
 
         return pos_args, return_visitor.return_values
-    except:
+    except Exception:
         return [], []
 
 
@@ -221,7 +224,7 @@ def __init__(self):
                 if isinstance(assignment.value, ast.Call) and isinstance(assignment.value.func, ast.Name):
                     if assignment.value.func.id == class_name:
                         variable_name = target_name
-    except:
+    except Exception:
         variable_name = None
     finally:
         # del current_frame, caller_frame

src/amisc/variable.py

@@ -26,11 +26,11 @@
 from pydantic import BaseModel, ConfigDict, ValidationInfo, field_validator
 
 from amisc.compression import Compression
-from amisc.distribution import Distribution, Normal, Uniform, LogUniform
+from amisc.distribution import Distribution, LogUniform, Normal, Uniform
 from amisc.serialize import Serializable
-from amisc.transform import Transform, Minmax, Zscore
-from amisc.utils import search_for_file, _get_yaml_path, _inspect_assignment
-from amisc.typing import CompressionData, LATENT_STR_ID
+from amisc.transform import Minmax, Transform, Zscore
+from amisc.typing import LATENT_STR_ID, CompressionData
+from amisc.utils import _get_yaml_path, _inspect_assignment, search_for_file
 
 __all__ = ['Variable', 'VariableList']
 _TransformLike = Union[str, Transform, list[str | Transform]]  # something that can be converted to a Transform

tests/profile_fit.py

@@ -5,7 +5,7 @@
 import numpy as np
 from matplotlib import pyplot as plt
 
-from amisc import Variable, Component, System
+from amisc import Component, System, Variable
 
 
 def io_bound_model(inputs, model_cost=1):