Task: Complete the statement to include these box plots and in this order: ['hr', 'mnth', 'weathersit', 'isWorking', 'dayWeek', 'xformHr']
Hint
colstoBoxPlot = ['hr','mnth','weathersit','isWorking','dayWeek','xformHr']
for cols in colstoBoxPlot:
fig = plt.figure(figsize=(4,4))
fig.clf()
ax = fig.gca()
bikesData.boxplot(column=['cntDeTrended'], by = [cols], ax = ax)
plt.xlabel('Box Plot of bike detrended counts by '+str(cols))
plt.ylabel('Number of bikes')
plt.show()
Observations: From these plots, we can observe the likely predictive power of categorical features.
Significant and complex variation in hourly bike demand can be seen in hr feature and xformHr. (this behavior may prove difficult to model)
In contrast, it looks doubtful that weather situation (weathersit), mnth, dayWeek or isWorking is going to be very helpful in predicting bike demand, despite the relatively high correlation value observed previously.
The result shown in ‘dayWeek’ boxplot is surprising — we expected bike demand to depend on the day of the week.
Once again, the outliers at the low end of bike demand can be seen in the box plots.
Action:
Hr and xformHr can be a good datetime feature to train the model on
Outlier observation shall be acted upon during the fine-tuning of the model.
Note - Having trouble with the assessment engine? Follow the steps listed here
Loading comments...