fixing tests and package versions. test work only black import is buggy.

Author: mo2561057

environment.yml

@@ -12,7 +12,7 @@ dependencies:
   - conda-build
   - conda-verify
   - doc8
-  - estimagic>=0.1.2
+  - estimagic=0.4.3
   - fastparquet
   - hypothesis
   - joblib
@@ -49,4 +49,4 @@ dependencies:
     - bump2version
     - pytest-randomly
     - sphinx-tabs
-    - chaospy>=4.2.3
+    - chaospy==3.3.8

respy/conditional_draws.py

@@ -1,6 +1,6 @@
 """Everything related to conditional draws for the maximum likelihood estimation."""
 import numpy as np
-from estimagic.optimization.utilities import robust_cholesky
+from estimagic.utilities import robust_cholesky
 from numba import guvectorize
 
 from respy.config import MAX_FLOAT
@@ -177,7 +177,7 @@ def update_mean_and_evaluate_likelihood(
             invariant
             - log_wage_observed
             - np.log(sigma)
-            - shock ** 2 / (2 * sigma_squared)
+            - shock**2 / (2 * sigma_squared)
         )
     else:
         for i in range(n_choices):

respy/pre_processing/model_processing.py

@@ -9,10 +9,10 @@
 import numpy as np
 import pandas as pd
 import yaml
-from estimagic.optimization.utilities import chol_params_to_lower_triangular_matrix
-from estimagic.optimization.utilities import cov_params_to_matrix
-from estimagic.optimization.utilities import robust_cholesky
-from estimagic.optimization.utilities import sdcorr_params_to_matrix
+from estimagic.utilities import chol_params_to_lower_triangular_matrix
+from estimagic.utilities import cov_params_to_matrix
+from estimagic.utilities import robust_cholesky
+from estimagic.utilities import sdcorr_params_to_matrix
 
 from respy.config import DEFAULT_OPTIONS
 from respy.config import MAX_FLOAT
@@ -189,7 +189,7 @@ def _parse_exogenous_processes(optim_paras, params):
     names = _parse_observable_or_exog_process_names(params, "exogenous_process")
 
     for exog_proc in names:
-        regex_pattern = fr"\bexogenous_process_{exog_proc}_([0-9a-z_]+)\b"
+        regex_pattern = rf"\bexogenous_process_{exog_proc}_([0-9a-z_]+)\b"
         parsed_parameters = _parse_probabilities_or_logit_coefficients(
             params, regex_pattern
         )
@@ -205,7 +205,7 @@ def _parse_observables(optim_paras, params):
     names = _parse_observable_or_exog_process_names(params, "observable")
 
     for observable in names:
-        regex_pattern = fr"\bobservable_{observable}_([0-9a-z_]+)\b"
+        regex_pattern = rf"\bobservable_{observable}_([0-9a-z_]+)\b"
         parsed_parameters = _parse_probabilities_or_logit_coefficients(
             params, regex_pattern
         )
@@ -286,7 +286,7 @@ def _parse_choice_parameters(optim_paras, params):
 def _parse_initial_and_max_experience(optim_paras, params, options):
     """Process initial experience distributions and maximum experience."""
     for choice in optim_paras["choices_w_exp"]:
-        regex_for_levels = fr"\binitial_exp_{choice}_([0-9]+)\b"
+        regex_for_levels = rf"\binitial_exp_{choice}_([0-9]+)\b"
         parsed_parameters = _parse_probabilities_or_logit_coefficients(
             params, regex_for_levels
         )
@@ -470,7 +470,7 @@ def _infer_choices_with_prefix(params, prefix):
     """
     return sorted(
         params.index.get_level_values(0)
-        .str.extract(fr"\b{prefix}_([A-Za-z_]+)\b", expand=False)
+        .str.extract(rf"\b{prefix}_([A-Za-z_]+)\b", expand=False)
         .dropna()
         .unique()
     )
@@ -544,7 +544,7 @@ def _parse_lagged_choices(optim_paras, options, params):
     # Add existing lagged choice parameters to ``optim_paras``.
     for lag in range(1, n_lc_covariates + 1):
         parsed_parameters = _parse_probabilities_or_logit_coefficients(
-            params, fr"lagged_choice_{lag}_([A-Za-z_]+)"
+            params, rf"lagged_choice_{lag}_([A-Za-z_]+)"
         )
 
         # If there are no parameters for the specific lag, assume equiprobable choices.

respy/shared.py

@@ -456,7 +456,7 @@ def calculate_expected_value_functions(
     The underlying process in this function is called `Monte Carlo integration
     <https://en.wikipedia.org/wiki/Monte_Carlo_integration>`_. The goal is to
     approximate an integral by evaluating the integrand at randomly chosen points. In
-    this setting, one wants to approximate the expected maximum utility of the current
+    this setting, one wants to approximate the m maximum utility of the current
     state.
 
     Note that ``wages`` have the same length as ``nonpecs`` despite that wages are only
@@ -589,8 +589,8 @@ def pandas_dot(x, beta, out=None):
     if not received_out:
         out = np.zeros(x.shape[0])
 
-    for covariate, beta in beta.items():
-        out += beta * x[covariate].values
+    for covariate, beta_ in beta.items():
+        out += beta_ * x[covariate].values
 
     if not received_out:
         return out

respy/state_space.py

@@ -705,7 +705,7 @@ def create_is_inadmissible(df, optim_paras, options):
         for formula in options["negative_choice_set"][choice]:
             try:
                 df[f"_{choice}"] |= df.eval(formula)
-            except pd.core.computation.ops.UndefinedVariableError:
+            except pd.errors.UndefinedVariableError:
                 pass
 
     return df

respy/tests/random_model.py

@@ -4,9 +4,9 @@
 
 import numpy as np
 import pandas as pd
-from estimagic.optimization.utilities import cov_matrix_to_params
-from estimagic.optimization.utilities import cov_matrix_to_sdcorr_params
-from estimagic.optimization.utilities import number_of_triangular_elements_to_dimension
+from estimagic.utilities import cov_matrix_to_params
+from estimagic.utilities import cov_matrix_to_sdcorr_params
+from estimagic.utilities import number_of_triangular_elements_to_dimension
 
 from respy.config import DEFAULT_OPTIONS
 from respy.config import ROOT_DIR
@@ -264,7 +264,7 @@ def _update_nested_dictionary(dict_, other):
 
     """
     for key, value in other.items():
-        if isinstance(value, collections.Mapping):
+        if isinstance(value, collections.abc.Mapping):
             dict_[key] = _update_nested_dictionary(dict_.get(key, {}), value)
         else:
             dict_[key] = value