从 csv 数据绘制频率直方图时出错

Question

I am working with a csv file with pandas module on python3. Csv file consists of 5 columns: job, company's name, description of the job, amount of reviews, location of the job; and i want to plot a frequency histogram , where i pick only the jobs containing the words "mechanical engineer" and find the frequencies of the 5 most frequent locations for the "mechanical engineer" job.

So,i defined a variable engloc which stores all the "mechanical engineer" jobs.

engloc=df[df.position.str.contains('mechanical engineer|mechanical engineering', flags=re.IGNORECASE, regex=True)].location

and did a histogram plot with matplotlib with code i found online

 x = np.random.normal(size = 1000)
 plt.hist(engloc, bins=50)
 plt.gca().set(title='Frequency Histogram ', ylabel='Frequency');

but it printed like this

How can i plot a proper frequency histogram where it plots using only 5 of the most frequent locations for jobs containing "mechanical engineer" words, instead of putting all of the locations in the graph?

This is a sample from the csv file

Answer 1

Something along the following lines should help you with numerical data:

import numpy as np
counts_, bins_ = np.histogram(englog.values)
filtered = [(c,b) for (c,b) in zip(counts_,bins_) if counts_>=5]
counts, bins = list(zip(*filtered))
plt.hist(bins[:-1], bins, weights=counts)

For a string type try:

from collections import Counter 
coords, counts = list(zip(*Counter(englog.values).most_common(5)))
plt.bar(coords, counts)