Project Description and KPIs¶

Case: Analyzing Government Credit Card Expenditures¶

(In progress)

Data 📊 - Description and modeling¶

  1. Data Collection: Collect credit card expenditure data through API calls from government sources;
  2. Data Cleaning, Validation and Inspection: Inspection to identify and address potential data quality issues: Missing values, inconsistencies, and duplications.
  3. Data Transformation: Transform the data into a usable format - standardizing data formats, resolving data types, and combining relevant datasets.

Step 1: Data Collection¶

Retrieve data through API calls

In [ ]:
import json
import requests
import pandas as pd
import numpy as np
import plotly.express as px
import locale
from decimal import Decimal
In [ ]:
def _get_br_gov_data(url, filters, page):
    complete_url = url + filters + "&pagina=" + str(page)
    headers = {
      'chave-api-dados': '7048f20892f6e59dc78cf7707e8ad843',
      'Cookie': 'JSESSIONID=358C2ECB5CB444691D06750877E67077'
    }

    response = requests.get(complete_url, headers=headers)
    return response

Jan-July 2023 Data

In [ ]:
url = "https://api.portaldatransparencia.gov.br/api-de-dados/cartoes"
filters = "?dataTransacaoFim=30%2F07%2F2023&dataTransacaoInicio=01%2F01%2F2023"

all_data = []
data_by_page =_get_br_gov_data(url, filters, 1).json()
page = 2

while len(data_by_page) > 0:
    all_data.extend(data_by_page)
    data_by_page =_get_br_gov_data(url, filters, page).json()
    if len(data_by_page) == 0:
        print("total pages" + str(page-1))
        break
    all_data.extend(data_by_page)

    page += 1

df_2023 = pd.json_normalize(all_data)
df_2023

total pages375
Out[ ]:
id mesExtrato dataTransacao valorTransacao tipoCartao.id tipoCartao.codigo tipoCartao.descricao estabelecimento.id estabelecimento.cpfFormatado estabelecimento.cnpjFormatado ... unidadeGestora.orgaoVinculado.codigoSIAFI unidadeGestora.orgaoVinculado.cnpj unidadeGestora.orgaoVinculado.sigla unidadeGestora.orgaoVinculado.nome unidadeGestora.orgaoMaximo.codigo unidadeGestora.orgaoMaximo.sigla unidadeGestora.orgaoMaximo.nome portador.cpfFormatado portador.nis portador.nome
0 346550462 03/2023 06/02/2023 366,00 1 1 Cartão de Pagamento do Governo Federal - CPGF 42308874 03.417.622/0001-08 ... 26280 45358058000140 FUF/São Carlos Fundação Universidade Federal de São Carlos 26000 MEC Ministério da Educação ***.046.888-** ADEMIR PADILHA ARRUDA JUNIOR
1 346550464 03/2023 13/02/2023 60,00 1 1 Cartão de Pagamento do Governo Federal - CPGF -2 ... 25000 00394460000141 MFAZ Ministério da Fazenda - Unidades com vínculo d... 25000 MFAZ Ministério da Fazenda ***.114.656-** GERALDO PAMPLONA FILHO
2 346550465 03/2023 10/02/2023 425,00 1 1 Cartão de Pagamento do Governo Federal - CPGF -1 ... 25205 33787094000140 IBGE Fundação Instituto Brasileiro de Geografia e E... 47000 MPO Ministério do Planejamento e Orçamento ***.753.217-** BRUNO BRANDAO PEREIRA MACHADO
3 346550469 03/2023 03/02/2023 340,00 1 1 Cartão de Pagamento do Governo Federal - CPGF 46626839 85.076.198/0001-40 ... 30108 00394494000136 PF Polícia Federal 30000 JUSTIÇA Ministério da Justiça e Segurança Pública ***.474.899-** LUIZ ADONIS KUHL
4 346550470 03/2023 09/02/2023 1.000,00 1 1 Cartão de Pagamento do Governo Federal - CPGF -2 ... 25205 33787094000140 IBGE Fundação Instituto Brasileiro de Geografia e E... 47000 MPO Ministério do Planejamento e Orçamento ***.222.849-** DELMO DE CARVALHO
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
11210 375255596 03/2023 10/02/2023 106,27 1 1 Cartão de Pagamento do Governo Federal - CPGF 6619790 38.616.553/0001-46 ... 25205 33787094000140 IBGE Fundação Instituto Brasileiro de Geografia e E... 47000 MPO Ministério do Planejamento e Orçamento ***.805.196-** OBEDE EDON DA SILVA
11211 375255601 03/2023 09/02/2023 2.147,47 1 1 Cartão de Pagamento do Governo Federal - CPGF 6019904 07.575.651/0001-59 ... 30108 00394494000136 PF Polícia Federal 30000 JUSTIÇA Ministério da Justiça e Segurança Pública ***.376.011-** LUCIARA NEVES DE SOUZA
11212 375255602 03/2023 09/02/2023 2.231,76 1 1 Cartão de Pagamento do Governo Federal - CPGF 6019904 07.575.651/0001-59 ... 30108 00394494000136 PF Polícia Federal 30000 JUSTIÇA Ministério da Justiça e Segurança Pública ***.376.011-** LUCIARA NEVES DE SOUZA
11213 375255603 03/2023 09/02/2023 1.112,73 1 1 Cartão de Pagamento do Governo Federal - CPGF 6019904 07.575.651/0001-59 ... 30108 00394494000136 PF Polícia Federal 30000 JUSTIÇA Ministério da Justiça e Segurança Pública ***.376.011-** LUCIARA NEVES DE SOUZA
11214 375255604 03/2023 09/02/2023 3.158,83 1 1 Cartão de Pagamento do Governo Federal - CPGF 6019904 07.575.651/0001-59 ... 30108 00394494000136 PF Polícia Federal 30000 JUSTIÇA Ministério da Justiça e Segurança Pública ***.376.011-** LUCIARA NEVES DE SOUZA

11215 rows × 28 columns

Jan-July 2022 Data

In [ ]:
url = "https://api.portaldatransparencia.gov.br/api-de-dados/cartoes"
filters = "?dataTransacaoFim=30%2F07%2F2022&dataTransacaoInicio=01%2F01%2F2022"

all_data = []
data_by_page = []
data_by_page =_get_br_gov_data(url, filters, 1).json()
page = 2

while len(data_by_page) > 0:
    # print (all_data)
    all_data.extend(data_by_page)
    data_by_page =_get_br_gov_data(url, filters, page).json()
    if len(data_by_page) == 0:
        print("total pages" + str(page-1))
        break
    all_data.extend(data_by_page)
    page += 1

df_2022 = pd.json_normalize(all_data)
df_2022

Merging 2022 DF and 2023 DFs

In [ ]:
df_2022 = df_2022.assign(Year=2022)
df_2023 = df_2023.assign(Year=2023)
df = pd.concat([df_2022,df_2023])
df
Out[ ]:
id mesExtrato valorTransacao tipoCartao.descricao unidadeGestora.orgaoMaximo.nome portador.nome Year
0 346550115 09/2022 139,90 Cartão de Pagamento do Governo Federal - CPGF Ministério de Minas e Energia GABRIELA COSTA STOLL 2022
1 346550116 04/2022 1.040,00 Cartão de Pagamento do Governo Federal - CPGF Ministério do Planejamento e Orçamento ELIAS ANTONIO DOS SANTOS 2022
2 346550117 06/2022 75,00 Cartão de Pagamento do Governo Federal - CPGF Ministério da Defesa DJALMA SILVA DOS SANTOS 2022
3 346550120 07/2022 200,00 Cartão de Pagamento do Governo Federal - CPGF Ministério do Planejamento e Orçamento EMIVAL LUDOVINO DE SANTANA 2022
4 346550124 05/2022 202,50 Cartão de Pagamento do Governo Federal - CPGF Ministério da Fazenda ANTONIO BARBOSA ABREU JUNIOR 2022
... ... ... ... ... ... ... ...
11205 375255596 03/2023 106,27 Cartão de Pagamento do Governo Federal - CPGF Ministério do Planejamento e Orçamento OBEDE EDON DA SILVA 2023
11206 375255601 03/2023 2.147,47 Cartão de Pagamento do Governo Federal - CPGF Ministério da Justiça e Segurança Pública LUCIARA NEVES DE SOUZA 2023
11207 375255602 03/2023 2.231,76 Cartão de Pagamento do Governo Federal - CPGF Ministério da Justiça e Segurança Pública LUCIARA NEVES DE SOUZA 2023
11208 375255603 03/2023 1.112,73 Cartão de Pagamento do Governo Federal - CPGF Ministério da Justiça e Segurança Pública LUCIARA NEVES DE SOUZA 2023
11209 375255604 03/2023 3.158,83 Cartão de Pagamento do Governo Federal - CPGF Ministério da Justiça e Segurança Pública LUCIARA NEVES DE SOUZA 2023

52293 rows × 7 columns

Step 2: Data Cleaning, Validation and Inspection¶

Examine and apply techniques to handle with issues like missing values, inconsistencies, errors, duplicates and outliers.

Removing Duplicates¶

Some of the IDs were duplicated (e.g. '366649383' in 2023). We'll remove duplicates (keeping the first one) to avoid redundancy in the analysis.

In [ ]:
df = df.drop_duplicates()

df reduced from 104,622 to 52,326 rows

Identifying/Handling Null and Missing Data¶

In [ ]:
print(df.info())

<class 'pandas.core.frame.DataFrame'>
Index: 52293 entries, 0 to 11209
Data columns (total 7 columns):
 #   Column                           Non-Null Count  Dtype 
---  ------                           --------------  ----- 
 0   id                               52293 non-null  int64 
 1   mesExtrato                       52293 non-null  object
 2   valorTransacao                   52293 non-null  object
 3   tipoCartao.descricao             52293 non-null  object
 4   unidadeGestora.orgaoMaximo.nome  52293 non-null  object
 5   portador.nome                    52293 non-null  object
 6   Year                             52293 non-null  int64 
dtypes: int64(2), object(5)
memory usage: 3.2+ MB
None

df doesn't have any missing data

Identifying/Handling Negative Values¶

Explicar pq estou tirando esse dado

In [ ]:
df = df[df['valorTransacao'].str.contains('-')==False]

33 rows with negative values were deleted

Removing unecessary columns¶

In [ ]:
df = df[['id','mesExtrato','valorTransacao','tipoCartao.descricao', 'unidadeGestora.orgaoMaximo.nome','portador.nome','Year']]

Step 3. Data Transformation¶

Converting data types as needed.

In [ ]:
#df['Fee'] = df['Fee'].map(lambda x: x - (x*10/100))
In [ ]:
def convert_value_to_float(number):
  locale.setlocale(locale.LC_ALL, 'en_US.UTF-8')

  # trocar por 2 replaces
  # replace('.','')
  # replace(',','.')
  
  number = list(number)
  i  = 0
  while i < len(number):
    if number[i] == '.':
      number[i] = ','
    elif number[i] == ',':
      number[i] = '.'
      break
    i +=1

  number = locale.atof(''.join(number))
  return number
In [ ]:
def create_dict_with_conv_values(df):
  dict_transactions = dict(zip(df['id'],df['valorTransacao']))
  for k,v in dict_transactions.items():
    dict_transactions.update({k:convert_value_to_float(v)})
  return dict_transactions
In [ ]:
def convert_dict_to_df(dict, column1, column2):
    dict_df = pd.DataFrame(
    [(k, val) for k, val in dict.items()], 
    columns=[column1, column2])
    return dict_df
In [ ]:
df_converted = convert_dict_to_df(create_dict_with_conv_values(df), 'id', 'valorTransacao')
new_df = (pd.merge(df, df_converted, on='id'))
new_df = new_df.drop(columns='valorTransacao_x')
new_df
Out[ ]:
id mesExtrato tipoCartao.descricao unidadeGestora.orgaoMaximo.nome portador.nome Year valorTransacao_y
0 346550115 09/2022 Cartão de Pagamento do Governo Federal - CPGF Ministério de Minas e Energia GABRIELA COSTA STOLL 2022 139.90
1 346550116 04/2022 Cartão de Pagamento do Governo Federal - CPGF Ministério do Planejamento e Orçamento ELIAS ANTONIO DOS SANTOS 2022 1040.00
2 346550117 06/2022 Cartão de Pagamento do Governo Federal - CPGF Ministério da Defesa DJALMA SILVA DOS SANTOS 2022 75.00
3 346550120 07/2022 Cartão de Pagamento do Governo Federal - CPGF Ministério do Planejamento e Orçamento EMIVAL LUDOVINO DE SANTANA 2022 200.00
4 346550124 05/2022 Cartão de Pagamento do Governo Federal - CPGF Ministério da Fazenda ANTONIO BARBOSA ABREU JUNIOR 2022 202.50
... ... ... ... ... ... ... ...
52288 375255596 03/2023 Cartão de Pagamento do Governo Federal - CPGF Ministério do Planejamento e Orçamento OBEDE EDON DA SILVA 2023 106.27
52289 375255601 03/2023 Cartão de Pagamento do Governo Federal - CPGF Ministério da Justiça e Segurança Pública LUCIARA NEVES DE SOUZA 2023 2147.47
52290 375255602 03/2023 Cartão de Pagamento do Governo Federal - CPGF Ministério da Justiça e Segurança Pública LUCIARA NEVES DE SOUZA 2023 2231.76
52291 375255603 03/2023 Cartão de Pagamento do Governo Federal - CPGF Ministério da Justiça e Segurança Pública LUCIARA NEVES DE SOUZA 2023 1112.73
52292 375255604 03/2023 Cartão de Pagamento do Governo Federal - CPGF Ministério da Justiça e Segurança Pública LUCIARA NEVES DE SOUZA 2023 3158.83

52293 rows × 7 columns

Changing column names to english

In [ ]:
new_df = new_df.rename(columns={'mesExtrato':'month','tipoCartao.descricao':'description','valorTransacao_y':'trans_value','unidadeGestora.orgaoMaximo.nome':'gov_name','portador.nome':'person_name'})
new_df
Out[ ]:
id month description gov_name person_name Year trans_value
0 346550115 09/2022 Cartão de Pagamento do Governo Federal - CPGF Ministério de Minas e Energia GABRIELA COSTA STOLL 2022 139.90
1 346550116 04/2022 Cartão de Pagamento do Governo Federal - CPGF Ministério do Planejamento e Orçamento ELIAS ANTONIO DOS SANTOS 2022 1040.00
2 346550117 06/2022 Cartão de Pagamento do Governo Federal - CPGF Ministério da Defesa DJALMA SILVA DOS SANTOS 2022 75.00
3 346550120 07/2022 Cartão de Pagamento do Governo Federal - CPGF Ministério do Planejamento e Orçamento EMIVAL LUDOVINO DE SANTANA 2022 200.00
4 346550124 05/2022 Cartão de Pagamento do Governo Federal - CPGF Ministério da Fazenda ANTONIO BARBOSA ABREU JUNIOR 2022 202.50
... ... ... ... ... ... ... ...
52288 375255596 03/2023 Cartão de Pagamento do Governo Federal - CPGF Ministério do Planejamento e Orçamento OBEDE EDON DA SILVA 2023 106.27
52289 375255601 03/2023 Cartão de Pagamento do Governo Federal - CPGF Ministério da Justiça e Segurança Pública LUCIARA NEVES DE SOUZA 2023 2147.47
52290 375255602 03/2023 Cartão de Pagamento do Governo Federal - CPGF Ministério da Justiça e Segurança Pública LUCIARA NEVES DE SOUZA 2023 2231.76
52291 375255603 03/2023 Cartão de Pagamento do Governo Federal - CPGF Ministério da Justiça e Segurança Pública LUCIARA NEVES DE SOUZA 2023 1112.73
52292 375255604 03/2023 Cartão de Pagamento do Governo Federal - CPGF Ministério da Justiça e Segurança Pública LUCIARA NEVES DE SOUZA 2023 3158.83

52293 rows × 7 columns

Exploratory Data Analysis (IN PROGRESS)¶

This phase involves an in-depth exploration of the data to answer intriguing questions about government spending. Our analysis is focused on a few key aspects:

Government Comparison; Monthly Spending Patterns; Top Spending Public Agencies; Top Spender; Outliers Detection; Highest Spending Establishments; etc

Identifying/Handling Outliers¶

  1. We'll use histogram to see how the data is distributed.
In [ ]:
fig_22 = px.histogram(df_2022, x='valorTransacao')
fig_22.show()
fig_23 = px.histogram(df_2023, x='valorTransacao')
fig_23.show()

Since the data is skewed, we'll not use a z-score. We'll use interquartile range (IQR) to determine the outliers.

  1. We'll create a box splot to visualy show outliers
In [ ]:
fig_22 = px.box(df_2022, y='valorTransacao')
fig_22.show()
  1. And, now, we'll use interquartile range (IQR) to determine the outliers.
In [ ]:
def find_outliers_IQR(df):

   q1=df.quantile(0.25)

   q3=df.quantile(0.75)

   IQR=q3-q1

   outliers = df[((df<(q1-1.5*IQR)) | (df>(q3+1.5*IQR)))]

   return outliers
In [ ]:
df = df_2022
outliers = find_outliers_IQR(df['valorTransacao'])
print('number of outliers: '+ str(len(outliers)))
print('max outlier value: '+ str(outliers.max()))
print('min outlier value: '+ str(outliers.min()))

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
Cell In[62], line 2
      1 df = df_2022
----> 2 outliers = find_outliers_IQR(df['valorTransacao'])
      3 print('number of outliers: '+ str(len(outliers)))
      4 print('max outlier value: '+ str(outliers.max()))

Cell In[61], line 3, in find_outliers_IQR(df)
      1 def find_outliers_IQR(df):
----> 3    q1=df.quantile(0.25)
      5    q3=df.quantile(0.75)
      7    IQR=q3-q1

File /opt/homebrew/lib/python3.11/site-packages/pandas/core/series.py:2650, in Series.quantile(self, q, interpolation)
   2646 # We dispatch to DataFrame so that core.internals only has to worry
   2647 #  about 2D cases.
   2648 df = self.to_frame()
-> 2650 result = df.quantile(q=q, interpolation=interpolation, numeric_only=False)
   2651 if result.ndim == 2:
   2652     result = result.iloc[:, 0]

File /opt/homebrew/lib/python3.11/site-packages/pandas/core/frame.py:10882, in DataFrame.quantile(self, q, axis, numeric_only, interpolation, method)
  10875 axis = self._get_axis_number(axis)
  10877 if not is_list_like(q):
  10878     # BlockManager.quantile expects listlike, so we wrap and unwrap here
  10879     # error: List item 0 has incompatible type "Union[float, Union[Union[
  10880     # ExtensionArray, ndarray[Any, Any]], Index, Series], Sequence[float]]";
  10881     # expected "float"
> 10882     res_df = self.quantile(  # type: ignore[call-overload]
  10883         [q],
  10884         axis=axis,
  10885         numeric_only=numeric_only,
  10886         interpolation=interpolation,
  10887         method=method,
  10888     )
  10889     if method == "single":
  10890         res = res_df.iloc[0]

File /opt/homebrew/lib/python3.11/site-packages/pandas/core/frame.py:10927, in DataFrame.quantile(self, q, axis, numeric_only, interpolation, method)
  10923     raise ValueError(
  10924         f"Invalid method: {method}. Method must be in {valid_method}."
  10925     )
  10926 if method == "single":
> 10927     res = data._mgr.quantile(qs=q, axis=1, interpolation=interpolation)
  10928 elif method == "table":
  10929     valid_interpolation = {"nearest", "lower", "higher"}

File /opt/homebrew/lib/python3.11/site-packages/pandas/core/internals/managers.py:1587, in BlockManager.quantile(self, qs, axis, interpolation)
   1584 new_axes = list(self.axes)
   1585 new_axes[1] = Index(qs, dtype=np.float64)
-> 1587 blocks = [
   1588     blk.quantile(axis=axis, qs=qs, interpolation=interpolation)
   1589     for blk in self.blocks
   1590 ]
   1592 return type(self)(blocks, new_axes)

File /opt/homebrew/lib/python3.11/site-packages/pandas/core/internals/managers.py:1588, in <listcomp>(.0)
   1584 new_axes = list(self.axes)
   1585 new_axes[1] = Index(qs, dtype=np.float64)
   1587 blocks = [
-> 1588     blk.quantile(axis=axis, qs=qs, interpolation=interpolation)
   1589     for blk in self.blocks
   1590 ]
   1592 return type(self)(blocks, new_axes)

File /opt/homebrew/lib/python3.11/site-packages/pandas/core/internals/blocks.py:1463, in Block.quantile(self, qs, interpolation, axis)
   1460 assert axis == 1  # only ever called this way
   1461 assert is_list_like(qs)  # caller is responsible for this
-> 1463 result = quantile_compat(self.values, np.asarray(qs._values), interpolation)
   1464 # ensure_block_shape needed for cases where we start with EA and result
   1465 #  is ndarray, e.g. IntegerArray, SparseArray
   1466 result = ensure_block_shape(result, ndim=2)

File /opt/homebrew/lib/python3.11/site-packages/pandas/core/array_algos/quantile.py:37, in quantile_compat(values, qs, interpolation)
     35     fill_value = na_value_for_dtype(values.dtype, compat=False)
     36     mask = isna(values)
---> 37     return quantile_with_mask(values, mask, fill_value, qs, interpolation)
     38 else:
     39     return values._quantile(qs, interpolation)

File /opt/homebrew/lib/python3.11/site-packages/pandas/core/array_algos/quantile.py:95, in quantile_with_mask(values, mask, fill_value, qs, interpolation)
     93     result = np.repeat(flat, len(values)).reshape(len(values), len(qs))
     94 else:
---> 95     result = _nanpercentile(
     96         values,
     97         qs * 100.0,
     98         na_value=fill_value,
     99         mask=mask,
    100         interpolation=interpolation,
    101     )
    103     result = np.array(result, copy=False)
    104     result = result.T

File /opt/homebrew/lib/python3.11/site-packages/pandas/core/array_algos/quantile.py:216, in _nanpercentile(values, qs, na_value, mask, interpolation)
    214     return result
    215 else:
--> 216     return np.percentile(
    217         values,
    218         qs,
    219         axis=1,
    220         # error: No overload variant of "percentile" matches argument types
    221         # "ndarray[Any, Any]", "ndarray[Any, dtype[floating[_64Bit]]]",
    222         # "int", "Dict[str, str]"  [call-overload]
    223         **{np_percentile_argname: interpolation},  # type: ignore[call-overload]
    224     )

File /opt/homebrew/lib/python3.11/site-packages/numpy/lib/function_base.py:4283, in percentile(a, q, axis, out, overwrite_input, method, keepdims, interpolation)
   4281 if not _quantile_is_valid(q):
   4282     raise ValueError("Percentiles must be in the range [0, 100]")
-> 4283 return _quantile_unchecked(
   4284     a, q, axis, out, overwrite_input, method, keepdims)

File /opt/homebrew/lib/python3.11/site-packages/numpy/lib/function_base.py:4555, in _quantile_unchecked(a, q, axis, out, overwrite_input, method, keepdims)
   4547 def _quantile_unchecked(a,
   4548                         q,
   4549                         axis=None,
   (...)
   4552                         method="linear",
   4553                         keepdims=False):
   4554     """Assumes that q is in [0, 1], and is an ndarray"""
-> 4555     return _ureduce(a,
   4556                     func=_quantile_ureduce_func,
   4557                     q=q,
   4558                     keepdims=keepdims,
   4559                     axis=axis,
   4560                     out=out,
   4561                     overwrite_input=overwrite_input,
   4562                     method=method)

File /opt/homebrew/lib/python3.11/site-packages/numpy/lib/function_base.py:3823, in _ureduce(a, func, keepdims, **kwargs)
   3820             index_out = (0, ) * nd
   3821             kwargs['out'] = out[(Ellipsis, ) + index_out]
-> 3823 r = func(a, **kwargs)
   3825 if out is not None:
   3826     return out

File /opt/homebrew/lib/python3.11/site-packages/numpy/lib/function_base.py:4721, in _quantile_ureduce_func(a, q, axis, out, overwrite_input, method)
   4719     else:
   4720         arr = a.copy()
-> 4721 result = _quantile(arr,
   4722                    quantiles=q,
   4723                    axis=axis,
   4724                    method=method,
   4725                    out=out)
   4726 return result

File /opt/homebrew/lib/python3.11/site-packages/numpy/lib/function_base.py:4840, in _quantile(arr, quantiles, axis, method, out)
   4838     result_shape = virtual_indexes.shape + (1,) * (arr.ndim - 1)
   4839     gamma = gamma.reshape(result_shape)
-> 4840     result = _lerp(previous,
   4841                    next,
   4842                    gamma,
   4843                    out=out)
   4844 if np.any(slices_having_nans):
   4845     if result.ndim == 0 and out is None:
   4846         # can't write to a scalar, but indexing will be correct

File /opt/homebrew/lib/python3.11/site-packages/numpy/lib/function_base.py:4655, in _lerp(a, b, t, out)
   4641 def _lerp(a, b, t, out=None):
   4642     """
   4643     Compute the linear interpolation weighted by gamma on each point of
   4644     two same shape array.
   (...)
   4653         Output array.
   4654     """
-> 4655     diff_b_a = subtract(b, a)
   4656     # asanyarray is a stop-gap until gh-13105
   4657     lerp_interpolation = asanyarray(add(a, diff_b_a * t, out=out))

TypeError: unsupported operand type(s) for -: 'str' and 'str'