metacountregressor 0.1.176__tar.gz → 0.1.202__tar.gz

{metacountregressor-0.1.176 → metacountregressor-0.1.202}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: metacountregressor
-Version: 0.1.176
+Version: 0.1.202
 Summary: Extensive Testing for Estimation of Data Count Models
 Home-page: https://github.com/zahern/CountDataEstimation
 Author: Zeke Ahern
@@ -16,7 +16,7 @@ Requires-Dist: latextable
 Requires-Dist: pandas
 Requires-Dist: scikit_learn>=1.4.1.post1
 Requires-Dist: statsmodels
-Requires-Dist: psustil
+Requires-Dist: psutil
 Dynamic: author
 Dynamic: author-email
 Dynamic: description

{metacountregressor-0.1.176 → metacountregressor-0.1.202}/metacountregressor/helperprocess.py

@@ -271,7 +271,17 @@ def guess_low_medium_high(column_name, series):
     # Compute the tertiles (33rd and 66th percentiles)
     #print('did it make it...')
     #mode_value = st.mode(series)  # Get the most frequent value
-    #print('good')
+    #i dont think this works cayse its not a seriers any other way
+    is_binary = series.isin([0, 1]).all()
+    if is_binary:
+        return {
+            'type': 'binary',
+            'bins': [0,1],
+            'labels': ['Off', 'On'],
+            'prefix': f'{column_name}'
+
+        }
+
    # series = pd.to_numeric(series, errors='coerce').fillna(mode_value)
     low_threshold = np.quantile(series, 0.33)
     high_threshold = np.quantile(series,0.66)

{metacountregressor-0.1.176 → metacountregressor-0.1.202}/metacountregressor/solution.py

@@ -159,7 +159,7 @@ class ObjectiveFunction(object):
         self.full_model = None
         self.GP_parameter = 0
         self.is_multi = kwargs.get('is_multi', False)
-        self.complexity_level = 6
+        self.complexity_level = kwargs('complexity_level', 6)
         self._max_iterations_improvement = 10000
         self.generated_sln = set()
         self.ave_mae = 0
@@ -256,10 +256,11 @@ class ObjectiveFunction(object):
                 self.is_multi = False
 
             if 'panels' in kwargs and not (kwargs.get('panels') == None):
-                self.group_names = np.asarray(x_data[kwargs['group']].astype('category').cat._parent.dtype.categories)
+                if kwargs.get('group') is not None:
+                    self.group_names = np.asarray(x_data[kwargs['group']].astype('category').cat._parent.dtype.categories)
 
-                x_data[kwargs['group']] = x_data[kwargs['group']].astype(
-                    'category').cat.codes
+                    x_data[kwargs['group']] = x_data[kwargs['group']].astype(
+                        'category').cat.codes
                 self.complexity_level = 6
                 # create test dataset
 
@@ -309,10 +310,13 @@ class ObjectiveFunction(object):
                 df_train[kwargs['panels']]) if kwargs['panels'] is not None else None
             self.ids_test = np.asarray(
                 df_test[kwargs['panels']]) if kwargs['panels'] is not None else None
-            groupll = np.asarray(df_train[kwargs['group']].astype(
-                'category').cat.codes)
-            group_test = np.asarray(df_test[kwargs['group']].astype(
-                'category').cat.codes)
+            if kwargs.get('group') is not None:
+                groupll = np.asarray(df_train[kwargs['group']].astype(
+                    'category').cat.codes)
+                group_test = np.asarray(df_test[kwargs['group']].astype(
+                    'category').cat.codes)
+            else:
+                groupll = None
             X, Y, panel, group = self._arrange_long_format(
                 df_train, y_train, self.ids, self.ids, groupll)
             self.group_halton = group.copy()
@@ -501,7 +505,7 @@ class ObjectiveFunction(object):
         self._max_hurdle = 4
 
         #Manually fit from analyst specification
-        manual_fit = kwargs.get('Manual_Fit')
+        manual_fit = kwargs.get('Manual_Fit', None)
         if manual_fit is not None:
             print('fitting manual')
             self.process_manual_fit(manual_fit)
@@ -538,7 +542,7 @@ class ObjectiveFunction(object):
                 if self.is_multi:
                     self._offsets_test = self._x_data_test[:, :, val_od]
                     self._x_data_test = self.remove_offset(self._x_data_test, val_od)
-                print(self._offsets)
+                #print(self._offsets)
             else:
                 self.initialize_empty_offsets()
 
@@ -1712,6 +1716,11 @@ class ObjectiveFunction(object):
                         vector[get_rdm_i] -= 1
                         only_ints_vals[get_rdm_i] -= 1
 
+                    elif vector[get_rdm_i] == 1:
+                        vector[get_rdm_i] -= 1
+                        only_ints_vals[get_rdm_i] -= 1
+
+
                     if vector.count(5) == 1:
                         idx = vector.index(5)
                         vector[idx] = 0
@@ -2361,7 +2370,7 @@ class ObjectiveFunction(object):
             sorted(my_dict, key=lambda x: x[0]['pval_percentage'])
 
     def get_fitness(self, vector, multi=False, verbose=False, max_routine=3):
-        obj_1 = 10.0 ** 5
+        obj_1 = 10.0 ** 4
         obj_best = None
         sub_slns = list()
 
@@ -2369,12 +2378,14 @@ class ObjectiveFunction(object):
             vector)  # just added to grab the fixed fit TODO: Clean up
         dispersion = model_nature.get('dispersion')
         self.define_selfs_fixed_rdm_cor(model_nature)
+
         try:
             self.repair(vector)
         except Exception as e:
-            print('prolem repairing here')
+            print('problem repairing here')
             print(vector)
             print(e)
+
         layout = vector.copy()
         trial_run = 0
         max_trial = 0
@@ -2453,10 +2464,10 @@ class ObjectiveFunction(object):
 
 
             if not self.is_quanitifiable_num(obj_1[self._obj_1]):
-                obj_1[self._obj_1] = 10 ** 9
+                obj_1[self._obj_1] = 10 ** 5
             else:
                 if obj_1[self._obj_1] <= 0:
-                    obj_1[self._obj_1] = 10 ** 9
+                    obj_1[self._obj_1] = 10 ** 5
 
         if multi:
 
@@ -2487,10 +2498,10 @@ class ObjectiveFunction(object):
 
         self.reset_sln()
         if not self.is_quanitifiable_num(obj_1[self._obj_1]):
-            obj_1[self._obj_1] = 10 ** 9
+            obj_1[self._obj_1] = 10 ** 5
         else:
             if obj_1[self._obj_1] == 0:
-                obj_1[self._obj_1] = 10 ** 9
+                obj_1[self._obj_1] = 10 **5
         if verbose:
             print('The best solution iteratively is of objective value:', obj_1)
 
@@ -3029,6 +3040,39 @@ class ObjectiveFunction(object):
         # print('log_lik poisson', log_lik)
         return -log_lik
 
+    def extract_parameters(self, betas, Kf, Kr, Kchol_a, Krb_a):
+        """
+        Extracts parameters from the `betas` array based on the given sizes.
+
+        Parameters:
+            betas (numpy.ndarray): The array of betas.
+            Kf (int): Size of Bf (first Kf elements of betas).
+            Kr (int): Size of Br.
+            Kchol_a (int): Part of the size for brstd.
+            Krb_a (int): Part of the size for brstd.
+
+        Returns:
+            tuple: A tuple containing:
+                - Bf (numpy.ndarray): The first Kf elements of betas.
+                - Br (numpy.ndarray): The next Kr elements of betas after Bf.
+                - brstd (numpy.ndarray): The next Kchol_a + Krb_a elements of betas after Br.
+                - remaining_betas (numpy.ndarray): Any remaining elements in betas after brstd.
+        """
+        # Step 1: Extract Bf
+        Bf = betas[:Kf]  # First Kf elements
+
+        # Step 2: Extract Br
+        Br = betas[Kf:Kf + Kr]  # Next Kr elements after Bf
+
+        # Step 3: Extract brstd
+        brstd_size = Kchol_a + Krb_a  # Total size of brstd
+        brstd = betas[Kf + Kr:Kf + Kr + brstd_size]  # Next brstd_size elements after Br
+
+        # Step 4: Extract remaining betas
+        remaining_betas = betas[Kf + Kr + brstd_size:]  # Remaining elements in betas
+
+        return Bf, Br, brstd, remaining_betas
+
     def convert_nbinom_params(self, mu, theta):
         """
             Convert mean/dispersion parameterization of a negative binomial to the ones scipy supports
@@ -3561,8 +3605,11 @@ class ObjectiveFunction(object):
         # Compute: betas = mean + sd*draws
         if len(br_sd) != draws.shape[1]:
             #get the same size as the mean
-            betas_random = self.Br.copy()
-
+            #if hasattr(self.Br):
+            #    betas_random = self.Br.copy()
+            #else:
+            idx = self.get_X_draw_tril()
+            betas_random = br_mean[None, :, None] + draws[:,idx, :] * br_sd[None, :, None]
             '''
             c = self.get_num_params()[3:5]
             
@@ -4716,10 +4763,10 @@ class ObjectiveFunction(object):
                 n_coeff = self.get_param_num(dispersion)
                 Kf_a, Kr_a, Kr_c, Kr_b_a, Kchol_a, Kh = self.get_num_params()
                 if Kchol_a != Kchol:
-                    print('hold')
+                    print('hold qhy')
 
                 if Kr_b != Kr_b_a:
-                    print('hold')
+                    print('hold qhy')
 
 
 
@@ -4735,13 +4782,32 @@ class ObjectiveFunction(object):
                 Bf = betas[0:Kf]  # Fixed betas
 
 
-
+           # Bf_new, Br_new, Br_std_new, Br_rema = self.extract_parameters(betas, Kf, Kr, Kchol_a, Kr_b_a)
 
 
             Vdf = dev.np.einsum('njk,k -> nj', Xdf, Bf, dtype=np.float64)  # (N, P)
             br = betas[Kf:Kf + Kr]
 
+
+
+            #i have an array of betas, Kf represents the first kf of the betas array
+            # now return Bf where size of bf = kf
+
+            # size of br needs to be Kr
+            #Kr
+            #now extract from betas, after all the Bf
+            # cakk
+
+            #the next array is brstd
+
+            # size of brstd needs to be
+            # Kchol_a + Krb_a
+            #its grabbing from the
+
+
+
             brstd = betas[Kf + Kr:Kf + Kr + Kr_b + Kchol]
+
             # initialises size matrix
             proba = []  # Temp batching storage
 
@@ -4755,6 +4821,8 @@ class ObjectiveFunction(object):
                 if len(self.none_handler(self.rdm_cor_fit)) == 0:
                     # Br = self._transform_rand_betas(br, np.abs(
                     #     brstd), draws_)  # Get random coefficients, old method
+                    #TODO
+
                     Br = self._transform_rand_betas(br,
                                                     brstd, draws_)  # Get random coefficients
                     self.naming_for_printing(betas, dispersion=dispersion, model_nature=model_nature)
@@ -5844,6 +5912,11 @@ class ObjectiveFunction(object):
 
                     if self.no_extra_param:
                         dispersion_poisson = 0
+                        print('b :', len(b))
+                        print(self.get_param_num())
+                        baby = self.get_param_num()
+                        if len(b) != baby:
+                            print('modify')
                         betas_est = self._minimize(self._loglik_gradient, b, args=(
                             X, y, draws, X, Xr, self.batch_size, self.grad_yes, self.hess_yes, dispersion_poisson, 0, False, 0,
                             self.rdm_cor_fit, None, None, draws_grouped, XG, mod),
@@ -6205,8 +6278,9 @@ class ObjectiveFunction(object):
                             transform, distribution, None, dispersion=dispersion)
 
     def get_named_indices(self, names):
-        indices = [i for i, name in enumerate(self._characteristics_names) if name in names]
-
+        # Change substrings issue
+        indices = [i for i, name in enumerate(self._characteristics_names) if name == names]
+        indices = [i for i, name in enumerate(self._characteristics_names) if name in names and isinstance(name, str)]
         return indices
 
     """
@@ -6482,10 +6556,7 @@ class ObjectiveFunction(object):
             else:
                 rv_indices.append(rv_count_all - 1)
 
-        # for s.d.: gr_w = (Obs prob. minus predicted probability) * obs. var * random draw
-        draws_tril_idx = np.array([corr_indices[j]
-                                   for i in range(len(self.none_handler(self.rdm_cor_fit)))
-                                   for j in range(i + 1)])  # varnames pos.
+
         X_tril_idx = np.array([corr_indices[i]
                                for i in range(len(self.none_handler(self.rdm_cor_fit)))
                                for j in range(i + 1)])
@@ -6494,12 +6565,55 @@ class ObjectiveFunction(object):
         range_var = [x for x in
                      range(len(self.none_handler(var_uncor)))]
         range_var = sorted(range_var)
-        draws_tril_idx = np.array(np.concatenate((range_var, draws_tril_idx)))
+
         X_tril_idx = np.array(np.concatenate((range_var, X_tril_idx)))
-        draws_tril_idx = draws_tril_idx.astype(int)
+
         X_tril_idx = X_tril_idx.astype(int)
         return  X_tril_idx
 
+    def get_X_draw_tril(self):
+        '''For correlations find the repeating terms'''
+        varnames = self.none_join([self.rdm_grouped_fit, self.rdm_fit, self.rdm_cor_fit])
+        rv_count_all = 0
+        chol_count = 0
+        rv_count = 0
+        corr_indices = []
+        rv_indices = []
+        for ii, var in enumerate(varnames):  # TODO: BUGFIXf
+            if var in self.none_handler(self.rdm_cor_fit):
+                is_correlated = True
+            else:
+                is_correlated = False
+
+            rv_count_all += 1
+            if is_correlated:
+                chol_count += 1
+            else:
+                rv_count += 1
+
+            if var in self.none_handler(self.rdm_cor_fit):
+
+                corr_indices.append(rv_count_all - 1)  # TODO: what does tis do
+
+            else:
+                rv_indices.append(rv_count_all - 1)
+
+        # for s.d.: gr_w = (Obs prob. minus predicted probability) * obs. var * random draw
+        draws_tril_idx = np.array([corr_indices[j]
+                                   for i in range(len(self.none_handler(self.rdm_cor_fit)))
+                                   for j in range(i + 1)])  # varnames pos.
+
+        # Find the s.d. for random variables that are not correlated
+        var_uncor = self.none_join([self.rdm_grouped_fit, self.rdm_fit])
+        range_var = [x for x in
+                     range(len(self.none_handler(var_uncor)))]
+        range_var = sorted(range_var)
+        draws_tril_idx = np.array(np.concatenate((range_var, draws_tril_idx)))
+
+        draws_tril_idx = draws_tril_idx.astype(int)
+
+        return draws_tril_idx
+
 
 
     def modifyn(self, data):

{metacountregressor-0.1.176 → metacountregressor-0.1.202}/metacountregressor.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: metacountregressor
-Version: 0.1.176
+Version: 0.1.202
 Summary: Extensive Testing for Estimation of Data Count Models
 Home-page: https://github.com/zahern/CountDataEstimation
 Author: Zeke Ahern
@@ -16,7 +16,7 @@ Requires-Dist: latextable
 Requires-Dist: pandas
 Requires-Dist: scikit_learn>=1.4.1.post1
 Requires-Dist: statsmodels
-Requires-Dist: psustil
+Requires-Dist: psutil
 Dynamic: author
 Dynamic: author-email
 Dynamic: description

{metacountregressor-0.1.176 → metacountregressor-0.1.202}/metacountregressor.egg-info/requires.txt

@@ -5,4 +5,4 @@ latextable
 pandas
 scikit_learn>=1.4.1.post1
 statsmodels
-psustil
+psutil

{metacountregressor-0.1.176 → metacountregressor-0.1.202}/setup.py

@@ -72,6 +72,6 @@ setuptools.setup(
         'pandas',
         'scikit_learn>=1.4.1.post1',
         'statsmodels',
-        'psustil'
+        'psutil'
     ]
 )