Merge pull request #26 from UBC-MDS/imputer

add test and implementation for Imputer function

Author: fatse

eda_utils_py/eda_utils_py.py

@@ -1,34 +1,36 @@
 import pandas as pd
 import altair as alt
 from pandas.api.types import is_numeric_dtype
+import numbers
 
-def imputer(dataframe, strategy="mean", fill_value=None):
+
+def imputer(df, strategy="mean", fill_value=None):
     """
     A function to implement imputation functionality for completing missing values.
 
     Parameters
     ----------
-    dataframe : pandas.core.frame.DataFrame
+    df : pandas.core.frame.DataFrame
         a dataframe that might contain missing data
     strategy : string, default="mean"
         The imputation strategy.
-            - If “mean”, then replace missing values using the mean along each column. Can only be used with numeric data.
-            - If “median”, then replace missing values using the median along each column. Can only be used with numeric data.
-            - If “most_frequent”, then replace missing using the most frequent value along each column. Can be used with strings or numeric data. If there is more than one such value, only the smallest is returned.
-            - If “constant”, then replace missing values with fill_value. Can be used with strings or numeric data.
-    fill_value : string or numerical value, default=None
-        When strategy == “constant”, fill_value is used to replace all occurrences of missing_values. If left to the default, fill_value will be 0 when imputing numerical data and “missing_value” for strings or object data types.
-        
+            - If "mean", then replace missing values using the mean along each column. Can only be used with numeric data.
+            - If "median", then replace missing values using the median along each column. Can only be used with numeric data.
+            - If "most_frequent", then replace missing using the most frequent value along each column. Can be used with strings or numeric data. If there is more than one such value, only the smallest is returned.
+            - If "constant", then replace missing values with fill_value. Can be used with strings or numeric data.
+    fill_value : numerical value, default=None
+        When strategy == "constant", fill_value is used to replace all occurrences of missing_values. If left to the default, fill_value will be 0 when imputing numerical data.
+
     Returns
     -------
     pandas.core.frame.DataFrame
         a dataframe that contains no missing data
-        
+
     Examples
     ---------
     >>> import pandas as pd
     >>> from eda_utils_py import cor_map
-    
+
     >>> data = pd.DataFrame({
     >>>     'SepalLengthCm':[5.1, 4.9, 4.7],
     >>>     'SepalWidthCm':[1.4, 1.4, 1.3],
@@ -41,24 +43,62 @@ def imputer(dataframe, strategy="mean", fill_value=None):
     1            4.9           1.4           0.2
     2            4.7           1.3           0.2
     """
-    pass
+
+    # Tests whether input data is of pd.DataFrame type
+    if not isinstance(df, pd.DataFrame):
+        raise TypeError("The input dataframe must be of pd.DataFrame type")
+
+    # Tests whether input strategy is of type str
+    if not isinstance(strategy, str):
+        raise TypeError("strategy must be of type str")
+
+    # Tests whether input fill_value is of type numbers or None
+    if not isinstance(fill_value, type(None)) and not isinstance(
+        fill_value, numbers.Number
+    ):
+        raise TypeError("fill_value must be of type None or numeric type")
+
+    # Tests whether the inputs for strategy and fill_value are consistent
+    if isinstance(fill_value, numbers.Number) and strategy != "constant":
+        raise Exception("fill_value can be a number only if strategy is 'constant'")
+
+    # Tests whether the inputs for strategy and fill_value are consistent
+    if isinstance(fill_value, type(None)) and strategy == "constant":
+        raise Exception("fill_value should be a number when strategy is 'constant'")
+
+    result = pd.DataFrame()
+    if strategy == "mean":
+        result = df.apply(lambda x: x.fillna(x.mean()), axis=0)
+    elif strategy == "median":
+        result = df.apply(lambda x: x.fillna(x.median()), axis=0)
+    elif strategy == "most_frequent":
+        result = df.apply(lambda x: x.fillna(x.value_counts().index[0]), axis=0)
+    elif strategy == "constant":
+        result = df.apply(lambda x: x.fillna(fill_value))
+    else:
+        raise Exception(
+            "strategy should be one of 'mean', 'median', 'most_frequent' and 'constant'"
+        )
+
+    return result
 
 
-def cor_map(dataframe, num_col, col_scheme = 'purpleorange'):
-    
+
+def cor_map(dataframe, num_col, col_scheme="purpleorange"):
+
     """
     A function to implement a correlation heatmap including coefficients based on given numeric columns of a data frame.
 
     Parameters
     ----------
     dataframe : pandas.core.frame.DataFrame
         The data frame to be used for EDA.
-    num_col : list  
+    num_col : list
         A list of string of column names with numeric data from the data frame.
-    col_scheme : str, default = 'purpleorange'  
+    col_scheme : str, default = 'purpleorange'
         The color scheme of the heatmap desired, can only be one of the following;
             - 'purpleorange'
-            - 'blueorange' 
+            - 'blueorange'
             - 'redblue'
 
     Returns
@@ -70,73 +110,73 @@ def cor_map(dataframe, num_col, col_scheme = 'purpleorange'):
     ---------
     >>> import pandas as pd
     >>> from eda_utils_py import cor_map
-    
+
     >>> data = pd.DataFrame({
     >>>     'SepalLengthCm':[5.1, 4.9, 4.7],
     >>>     'SepalWidthCm':[1.4, 1.4, 1.3],
     >>>     'PetalWidthCm':[0.2, 0.2, 0.2],
     >>>     'Species':['Iris-setosa','Iris-virginica', 'Iris-germanica']
     >>> })
 
-    >>> numerical_columns = ['SepalLengthCm','SepalWidthCm','PetalWidthCm']    
+    >>> numerical_columns = ['SepalLengthCm','SepalWidthCm','PetalWidthCm']
     >>> cor_map(data, numerical_columns, col_scheme = 'purpleorange')
-        
+
     """
-    
+
     # Tests whether input data is of pd.DataFrame type
     if not isinstance(dataframe, pd.DataFrame):
         raise TypeError("The input dataframe must be of pd.DataFrame type")
 
     # Tests whether input num_col is of type list
     if not isinstance(num_col, list):
         raise TypeError("The input num_col must be of type list")
-    
-    # Tests whether values of num_col is of type str 
+
+    # Tests whether values of num_col is of type str
     for x in num_col:
         if not isinstance(x, str):
             raise TypeError("The type of values in num_col must all be str")
 
-    # Tests whether input col_scheme is of type str    
+    # Tests whether input col_scheme is of type str
     if not isinstance(col_scheme, str):
         raise TypeError("col_scheme must be of type str")
 
-    # Tests whether col_scheme is one of three possible options    
-    if col_scheme not in ('purpleorange', 'blueorange', 'redblue'):
-        raise Exception("This color scheme is not available, please use either 'purpleorange', 'blueorange' or 'redblue'")
+    # Tests whether col_scheme is one of three possible options
+    if col_scheme not in ("purpleorange", "blueorange", "redblue"):
+        raise Exception(
+            "This color scheme is not available, please use either 'purpleorange', 'blueorange' or 'redblue'"
+        )
 
     # Tests whether all input columns exist in the input data
     for x in num_col:
         if x not in list(dataframe.columns):
-            raise Exception("The given column names must exist in the given dataframe.")    
-        
+            raise Exception("The given column names must exist in the given dataframe.")
+
     # Tests whether all input columns in num_col are numeric columns
     for x in num_col:
         if not is_numeric_dtype(dataframe[x]):
             raise Exception("The given numerical columns must all be numeric.")
-    
-    
-    corr_matrix = dataframe[num_col].corr().reset_index().melt('index')
-    corr_matrix.columns = ['var1', 'var2', 'cor']
-
-    plot = alt.Chart(corr_matrix).mark_rect().encode(
-        x=alt.X('var1', title=None),
-        y=alt.Y('var2', title=None),
-        color=alt.Color('cor',legend=None,
-                       scale = alt.Scale(scheme = col_scheme)),
-    ).properties(
-        title = 'Correlation Matrix',
-        width=400, height=400
+
+    corr_matrix = dataframe[num_col].corr().reset_index().melt("index")
+    corr_matrix.columns = ["var1", "var2", "cor"]
+
+    plot = (
+        alt.Chart(corr_matrix)
+        .mark_rect()
+        .encode(
+            x=alt.X("var1", title=None),
+            y=alt.Y("var2", title=None),
+            color=alt.Color("cor", legend=None, scale=alt.Scale(scheme=col_scheme)),
+        )
+        .properties(title="Correlation Matrix", width=400, height=400)
     )
 
     text = plot.mark_text(size=15).encode(
-        text=alt.Text('cor', format=".2f"),
+        text=alt.Text("cor", format=".2f"),
         color=alt.condition(
-            "datum.cor > 0.5 | datum.cor < -0.3",
-            alt.value('white'),
-            alt.value('black')
-        )
+            "datum.cor > 0.5 | datum.cor < -0.3", alt.value("white"), alt.value("black")
+        ),
     )
-    
+
     cor_heatmap = plot + text
 
     return cor_heatmap
@@ -147,26 +187,26 @@ def outlier_identifier(dataframe, columns=None, method="trim"):
     A function that identify and deal with outliers based on the method the user choose
 
     Parameters
-    ---------- 
+    ----------
     dataframe : pandas.core.frame.DataFrame
         The target dataframe where the function is performed.
     columns : list, default=None
         The target columns where the function needed to be performed. Defualt is None, the function will check all columns
     method : string
-        The method of dealing with outliers. 
+        The method of dealing with outliers.
             - if "trim" : we completely remove data points that are outliers.
             - if "median" : we replace outliers with median values
-        
+
     Returns
     -------
     pandas.core.frame.DataFrame
         a dataframe which the outlier has already process by the chosen method
-    
+
     Examples
     --------
     >>> import pandas as pd
     >>> from eda_utils_py import cor_map
-        
+
     >>> data = pd.DataFrame({
     >>>    'SepalLengthCm':[5.1, 4.9, 4.7],
     >>>    'SepalWidthCm':[1.4, 1.4, 9999999.99],
@@ -200,7 +240,7 @@ def scale(dataframe, columns=None):
     --------
     >>> import pandas as pd
     >>> from eda_utils_py import scale
-    
+
     >>> data = pd.DataFrame({
     >>>     'SepalLengthCm':[5.1, 4.9, 4.7],
     >>>     'SepalWidthCm':[1.4, 1.4, 1.3],
@@ -209,7 +249,7 @@ def scale(dataframe, columns=None):
     >>> })
 
     >>> numerical_columns = ['SepalLengthCm','SepalWidthCm','PetalWidthCm']
-    
+
     >>> scale(data, numerical_columns)
     """
     pass

poetry.lock

@@ -8,6 +8,7 @@ license = "MIT"
 [tool.poetry.dependencies]
 python = "^3.8"
 pandas = "^1.2.2"
+altair = "^4.1.0"
 
 [tool.poetry.dev-dependencies]
 Sphinx = "^3.5.1"