Seasborn Distplot goes unresponsive

问题

I am trying to plot a simple Distplot using pandas and seaborn to understand the density of the datasets.

Input

#Car,45
#photo,4
#movie,6
#life,1
#Horse,14
#Pets,20
#run,67
#picture,89

The dataset has above 10K rows, no headers and I am trying to use col[1]

code

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns


df = pd.read_csv('keyword.csv', delimiter=',', header=None, usecols=[1])
#print df
sns.distplot(df)

plt.show()

No error as I can print the input column but the distplot is taking ages to compute and freezes my screen. Any suggestion to speed the process.

Edit1: As Suggested in the Comment Below I try to change from pandas.read_csv to np.loadtxt and now I get an error.

Code:

import numpy as np
from numpy import log as log
import matplotlib.pyplot as plt
import seaborn as sns
import pandas

df = np.loadtxt('keyword.csv', delimiter=',', usecols=(1), unpack=True)
sns.kdeplot(df)
sns.distplot(df)

plt.show()

Error:

Traceback (most recent call last):
  File "0_distplot_csv.py", line 7, in <module>
    df = np.loadtxt('keyword.csv', delimiter=',', usecols=(1), unpack=True)
  File "/usr/local/lib/python2.7/dist-packages/numpy/lib/npyio.py", line 726, in loadtxt
    usecols = list(usecols)
TypeError: 'int' object is not iterable

Edit 2: I did try the mentioned suggestions from the comment section

sns.distplot(df[1])

This does the same as mentioned initially. The screen is frozen for ages.

sns.distplot(df[1].values)

I see a strange behavior in this case.

When the input is

Car,45
photo,4
movie,6
life,1
Horse,14
Pets,20
run,67
picture,89

It does plot but when the input is below

#Car,45
#photo,4
#movie,6
#life,1
#Horse,14
#Pets,20
#run,67
#picture,89

It is again the same freezing entire screen and would do nothing.

I did try to put comments=None thinking it might be reading them as comments. But looks like comments isn't used in pandas.

Thank you

回答1:

After several trials and a lot of online search, I could finally get what I was looking for. The code allows to load data with column number when we do not have headers. This also reads the rows with # comments.

code:

import numpy as np
import matplotlib.pyplot as plt
from pylab import*
import math
from matplotlib.ticker import LogLocator
from scipy.stats.kde import gaussian_kde
import seaborn as sns

data = np.genfromtxt('keyword.csv', delimiter=',', comments=None)

d0=data[:,1]

#Plot a simple histogram with binsize determined automatically
sns.kdeplot(np.array(d0), color='b', bw=0.5, marker='o', label='keyword')

plt.legend(loc='upper right')
plt.xlabel('Freq(x)')
plt.ylabel('pdf(x)')
#plt.gca().set_xscale("log")
#plt.gca().set_yscale("log")
plt.show()

来源：https://stackoverflow.com/questions/36616118/seasborn-distplot-goes-unresponsive

标签

python

pandas

plot

data-visualization

seaborn