How can I solve " module 'pandas' has no attribute 'scatter_matrix' " error?

Question

I'm trying to run pd.scatter_matrix() function in Jupyter Notebook with my code below:

import matplotlib.pyplot as plt
import pandas as pd

# Load some data
iris = datasets.load_iris()
iris_df = pd.DataFrame(iris['data'], columns=iris['feature_names'])
iris_df['species'] = iris['target']

pd.scatter_matrix(iris_df, alpha=0.2, figsize=(10, 10))
plt.show()

But I'm getting AttributeError: module 'pandas' has no attribute 'scatter_matrix'. Even after executing conda update pandas and conda update matplotlib commands in Terminal, this is still occurring.

I executed pd.__version__ command to check my pandas version and it's '0.24.2'. What could be the problem?

Answer 1

Another option is keeping only pandas import and rewriting the command scatter_matrix, like in the example below:

import pandas as pd

pd.plotting.scatter_matrix(iris_df, alpha=0.2, figsize=(10, 10))

Answer 2

Using

from pandas.plotting._misc import scatter_matrix

don't use pd.scatter_matrix or pandas.scatter_matrix you can directly call scatter_matrix

e.g.

cmap = cm.get_cmap('gnuplot')
scatter = scatter_matrix(X, c = y, marker = 'o', s=40, hist_kwds={'bins':15}, 
    figsize=(9,9), cmap = cmap)
plt.suptitle('Scatter-matrix for each input variable')
plt.savefig('fruits_scatter_matrix')
plt.show()

Answer 3

Use:

from pandas.plotting import scatter_matrix

The code becomes:

import matplotlib.pyplot as plt
from pandas.plotting import scatter_matrix

iris = datasets.load_iris()
iris_df = pd.DataFrame(iris['data'], columns=iris['feature_names'])
iris_df['species'] = iris['target']

scatter_matrix(iris_df, alpha=0.2, figsize=(10, 10))
plt.show()

Answer 4

I used

from pandas.plotting import scatter_matrix

and called scatter_matrix directly worked like charm.

Answer 5

In our case we were executing below code "axs = pd.scatter_matrix(sampled_data, figsize=(10, 10)) "

so the error clearly says scatter_matrix is not available in pandas

Solution: A bit of google and we found scatter_matrix is available in pandas.plotting

So correct code is "axs = pd.plotting.scatter_matrix(sampled_data, figsize=(10, 10)) "