IP timeseries matplotlib

ip timeseries matplotlib

user_name = '[email protected]'
################################################################################
# Data Processing
import pandas as pd
dataset = df[df['UserIds']==user_name][['CreationDate','ClientIP']]
dataset['ClientIP'] = dataset['ClientIP'].replace('','0.0.0.0').replace('<null>','0.0.0.0')
dataset['ip'] = dataset['ClientIP'].astype(str).apply(lambda x: x.split(':')[0] if len(x.split('.'))==4 and len(x.split(':'))==2 else x)
dataset = dataset.set_index('CreationDate')
dataset['ip_id'] = dataset['ip'].astype('category').cat.codes
categories = dataset.groupby(['ip','ip_id']).size().reset_index()
categories.sort_values(by=0,ascending=False).head()
################################################################################
# Plotting
import matplotlib.pyplot as plt
%matplotlib inline 
plt.style.use('ggplot')
import matplotlib.dates as mdates

fig = plt.figure(figsize=(15, 5))
ax = plt.subplot2grid((1,1), (0,0))
dataset['ip_id'].plot(ax=ax, y='ip_id',style=".")
ax.set_yticklabels(categories['ip'], fontsize=8)
ax.set_yticks(np.arange(min(dataset['ip_id']), max(dataset['ip_id']+1), 1.0))
ax.set_ylim(min(dataset['ip_id']), max(dataset['ip_id']))
ax.xaxis.set_major_locator(mdates.DayLocator(interval = 1))
ax.xaxis.set_major_formatter(mdates.DateFormatter('%d-%m'))
ax.tick_params(axis='x', rotation=90, labelsize=8)
plt.axvline(dt.datetime(2020, 2, 7),linestyle='--',color='r')
fig.show()