Hi,
I am trying to build a proportion table of late vendors based off orders. I built this basic code back in october through Jupyter Notebook and it still works fine there. I am trying to implement it on power bi now and I am confused whether it is necessary to declare the variable I am using, even though I have selected the column in Power BI. I get a name error as follows: LateDays is not defined. LateDays is the field that I get on power bi by using Odata, a service my company uses to fetch live data.
this is the code:
df = pandas.DataFrame(LateDays)
df = dataset.drop_duplicates()
import pandas as pd
import numpy as np
import matplotlib
import matplotlib.pyplot as plt
matplotlib.style.use('seaborn')
import seaborn as sns
sns.set(font_scale = 2)
# Paste or type your script code here:
def is_late(x):
return x > 0
df['Late'] = df['LateDays'].apply(is_late)
group_by_vendor = df.groupby(['CompanyName', 'Late'])
delayed_by_vendor = df.groupby(['CompanyName', 'Late']).size().unstack().reset_index()
delayed_by_vendor['orders_count'] = (delayed_by_vendor[False] + delayed_by_vendor[True])
delayed_by_vendor['Proportion_Late'] = delayed_by_vendor[True] / delayed_by_vendor['orders_count']
delayed_by_vendor.sort_values('proportion_late', ascending=False)
Right at the first string of code I get the name error. How can I define the variable in the code if the variable is fetched off a live service with huge amounts of data? thanks.
