Visualization Plots

This Jupyter notebook provide some examples of how to use the Visualization module from mdatagen. The first step is get Iris dataset from Scikit-Learn. Afterward, we will generate missing values as following:

In [31]:

# Get iris dataset and insert MV in 2 columns
import pandas as pd
from sklearn.datasets import load_iris
import numpy as np
from typing import Optional
from mdatagen.plots import PlotMissingData

def create_iris_missing(mrate, col_1:str, col_2:Optional[str]=None):
    iris = load_iris()
    iris_df = pd.DataFrame(data=iris.data, columns=iris.feature_names)
    #iris_df['target'] = iris.target

    iris_miss = iris_df.copy()
    num_missing = int(mrate * len(iris_miss))
    missing_indices = np.random.choice(iris_miss.index, size=num_missing, replace=False)
    iris_miss.loc[missing_indices, col_1] = np.nan

    if col_2:
        missing_indices = np.random.choice(iris_miss.index, size=num_missing, replace=False)
        iris_miss.loc[missing_indices, col_2] = np.nan

    return iris_df, iris_miss

# Get iris dataset and insert MV in 2 columns import pandas as pd from sklearn.datasets import load_iris import numpy as np from typing import Optional from mdatagen.plots import PlotMissingData def create_iris_missing(mrate, col_1:str, col_2:Optional[str]=None): iris = load_iris() iris_df = pd.DataFrame(data=iris.data, columns=iris.feature_names) #iris_df['target'] = iris.target iris_miss = iris_df.copy() num_missing = int(mrate * len(iris_miss)) missing_indices = np.random.choice(iris_miss.index, size=num_missing, replace=False) iris_miss.loc[missing_indices, col_1] = np.nan if col_2: missing_indices = np.random.choice(iris_miss.index, size=num_missing, replace=False) iris_miss.loc[missing_indices, col_2] = np.nan return iris_df, iris_miss

Only 1 feature with missing values¶

In [32]:

mr = 0.4
col_1 = "petal length (cm)"
col_2 = "sepal width (cm)"
cols = (col_1, col_2)
iris_df, iris_miss = create_iris_missing(mr, col_1="petal length (cm)")

mr = 0.4 col_1 = "petal length (cm)" col_2 = "sepal width (cm)" cols = (col_1, col_2) iris_df, iris_miss = create_iris_missing(mr, col_1="petal length (cm)")

In [4]:

iris_df

iris_df

Out[4]:

	sepal length (cm)	sepal width (cm)	petal length (cm)	petal width (cm)
0	5.1	3.5	1.4	0.2
1	4.9	3.0	1.4	0.2
2	4.7	3.2	1.3	0.2
3	4.6	3.1	1.5	0.2
4	5.0	3.6	1.4	0.2
...	...	...	...	...
145	6.7	3.0	5.2	2.3
146	6.3	2.5	5.0	1.9
147	6.5	3.0	5.2	2.0
148	6.2	3.4	5.4	2.3
149	5.9	3.0	5.1	1.8

150 rows × 4 columns

In [5]:

iris_miss

iris_miss

Out[5]:

	sepal length (cm)	sepal width (cm)	petal length (cm)	petal width (cm)
0	5.1	3.5	1.4	0.2
1	4.9	3.0	NaN	0.2
2	4.7	3.2	1.3	0.2
3	4.6	3.1	1.5	0.2
4	5.0	3.6	NaN	0.2
...	...	...	...	...
145	6.7	3.0	NaN	2.3
146	6.3	2.5	5.0	1.9
147	6.5	3.0	NaN	2.0
148	6.2	3.4	5.4	2.3
149	5.9	3.0	5.1	1.8

150 rows × 4 columns

In [ ]:

miss_plot = PlotMissingData(data_missing=iris_miss, data_original=iris_df)

miss_plot = PlotMissingData(data_missing=iris_miss, data_original=iris_df)

In [7]:

miss_plot.visualize_miss("normal")

miss_plot.visualize_miss("normal")

In [8]:

miss_plot.visualize_miss("bar")

miss_plot.visualize_miss("bar")

In [9]:

miss_plot.visualize_miss("dendrogram")

miss_plot.visualize_miss("dendrogram")

In [10]:

miss_plot.visualize_miss("heatmap")

miss_plot.visualize_miss("heatmap")

In [11]:

miss_plot.visualize_miss(visualization_type="histogram", col_missing=col_1, num_bins=5)

miss_plot.visualize_miss(visualization_type="histogram", col_missing=col_1, num_bins=5)

In [12]:

miss_plot.visualize_miss(visualization_type="boxplot", col_missing=col_1)

miss_plot.visualize_miss(visualization_type="boxplot", col_missing=col_1)

In [13]:

miss_plot.visualize_miss(visualization_type="scatterplot", cols=(col_1, col_2))

miss_plot.visualize_miss(visualization_type="scatterplot", cols=(col_1, col_2))

We can also change the axis of the plot (x or y for missing):

In [14]:

miss_plot.visualize_miss(visualization_type="scatterplot", cols=(col_2, col_1))

miss_plot.visualize_miss(visualization_type="scatterplot", cols=(col_2, col_1))

Two features with missing values¶

In [33]:

_, iris_miss_2d = create_iris_missing(mr, col_1, col_2)

_, iris_miss_2d = create_iris_missing(mr, col_1, col_2)

In [34]:

iris_miss_2d

iris_miss_2d

Out[34]:

	sepal length (cm)	sepal width (cm)	petal length (cm)	petal width (cm)
0	5.1	3.5	1.4	0.2
1	4.9	NaN	1.4	0.2
2	4.7	3.2	1.3	0.2
3	4.6	NaN	1.5	0.2
4	5.0	NaN	1.4	0.2
...	...	...	...	...
145	6.7	NaN	5.2	2.3
146	6.3	2.5	NaN	1.9
147	6.5	3.0	5.2	2.0
148	6.2	3.4	NaN	2.3
149	5.9	NaN	5.1	1.8

150 rows × 4 columns

In [35]:

miss_plot_2d = PlotMissingData(data_missing=iris_miss_2d, data_original=iris_df)

miss_plot_2d = PlotMissingData(data_missing=iris_miss_2d, data_original=iris_df)

In [36]:

miss_plot_2d.visualize_miss("normal")

miss_plot_2d.visualize_miss("normal")

In [37]:

miss_plot_2d.visualize_miss("bar")

miss_plot_2d.visualize_miss("bar")

In [38]:

miss_plot_2d.visualize_miss("dendrogram")

miss_plot_2d.visualize_miss("dendrogram")

In [39]:

miss_plot_2d.visualize_miss("heatmap")

miss_plot_2d.visualize_miss("heatmap")

In [40]:

miss_plot_2d.visualize_miss(visualization_type="histogram", col_missing=col_1, num_bins=5)

miss_plot_2d.visualize_miss(visualization_type="histogram", col_missing=col_1, num_bins=5)

In [41]:

miss_plot_2d.visualize_miss(visualization_type="histogram", col_missing=col_2, num_bins=5)

miss_plot_2d.visualize_miss(visualization_type="histogram", col_missing=col_2, num_bins=5)

In [42]:

miss_plot_2d.visualize_miss(visualization_type="boxplot", col_missing=col_1)

miss_plot_2d.visualize_miss(visualization_type="boxplot", col_missing=col_1)

In [43]:

miss_plot_2d.visualize_miss(visualization_type="boxplot", col_missing=col_2)

miss_plot_2d.visualize_miss(visualization_type="boxplot", col_missing=col_2)

In [44]:

miss_plot_2d.visualize_miss(visualization_type="scatterplot", cols=(col_1, col_2))

miss_plot_2d.visualize_miss(visualization_type="scatterplot", cols=(col_1, col_2))

In [201]:

iris_miss

iris_miss

Out[201]:

	sepal length (cm)	sepal width (cm)	petal length (cm)	petal width (cm)
0	5.1	3.5	NaN	0.2
1	4.9	3.0	1.4	0.2
2	4.7	3.2	NaN	0.2
3	4.6	3.1	1.5	0.2
4	5.0	3.6	1.4	0.2
...	...	...	...	...
145	6.7	3.0	NaN	2.3
146	6.3	2.5	5.0	1.9
147	6.5	3.0	5.2	2.0
148	6.2	3.4	NaN	2.3
149	5.9	3.0	NaN	1.8

150 rows × 4 columns