diff --git a/build.py b/build.py index c7d540b..38327fa 100644 --- a/build.py +++ b/build.py @@ -1,22 +1,24 @@ -def get_categorical_variables(df): - return [] +import pandas as pd +df = pd.read_csv('data/employee_retention_data.csv') +def get_categorical_variables(df): + return df.select_dtypes(include=['object']).columns def get_numerical_variables(df): - return [] + return df.select_dtypes(exclude=['object']).columns def get_numerical_variables_percentile(df): - pass + return df.select_dtypes(exclude=['object']).describe() def get_categorical_variables_modes(df): - pass + return df.select_dtypes(include=['object']).mode() def get_missing_values_count(df): - pass + return df.isnull() def plot_histogram_with_numerical_values(df): - pass + return df.plot.hist(bins=20) diff --git a/build.pyc b/build.pyc new file mode 100644 index 0000000..e49c421 Binary files /dev/null and b/build.pyc differ diff --git a/tests/__init__.pyc b/tests/__init__.pyc new file mode 100644 index 0000000..62dec64 Binary files /dev/null and b/tests/__init__.pyc differ diff --git a/tests/test_get_categorical_variables.pyc b/tests/test_get_categorical_variables.pyc new file mode 100644 index 0000000..7fbda9d Binary files /dev/null and b/tests/test_get_categorical_variables.pyc differ