数据清洗

df = df.dropna() test = test.dropna()

数据处理

job_list = df['job'].unique().tolist() marital_list = df['marital'].unique().tolist()

job_marital = pd.DataFrame(columns=['job', 'marital', 'count'])

for job in job_list: for marital in marital_list: count = len(df[(df['job'] == job) & (df['marital'] == marital)]) job_marital = job_marital.append({'job': job, 'marital': marital, 'count': count}, ignore_index=True)

可视化

from bokeh.palettes import Spectral6 from bokeh.transform import factor_cmap

output_file('job_marital.html')

p = figure(x_range=job_marital['job'].unique().tolist(), plot_height=400, title='Job and Marital Status Relationship', toolbar_location=None, tools="")

p.vbar(x='job_marital', top='count', width=0.9, source=job_marital, line_color='white', fill_color=factor_cmap('job_marital', palette=Spectral6, factors=job_marital['marital'].unique().tolist()))

p.xgrid.grid_line_color = None p.y_range.start = 0 p.xaxis.axis_label = 'Job and Marital Status' p.xaxis.major_label_orientation = 1.2 p.outline_line_color = None

show(p)