Thanks!

[anderpham]

• Your notebook file is good arrangement and provide enough content and code to the requirement.
• Solutions are correct
  Good job!

[anderpham]

Assignment Week 2 Review:

What was the average (mean) BasePay of all employees per year? (2011-2014) ?

Your code: groupby "BasePay" is not correct, you have to groupby Year

A correct solution:

sal.groupby("Year")["BasePay"].mean()

How many unique job titles are there?

To count the number of unique value, you can use the method nunique() in pandas, or simply use len() to count total elements of your output array

sal["JobTitle"].nunique()

How many Job Titles were represented by only one person in 2013? (e.g. Job Titles with only one occurence in 2013?)

Your solution is not correct because you counted the number appearance of "EmployeeName"
A correct solution:

sum(sal[sal["Year"] == 2013]["JobTitle"].value_counts() == 1)

Distplot for BasePay of Year 2012-2014

You can draw a single chart including multiple year lines

# Some BasePay values are missing, fill it with 0 instead
sal['BasePay'].fillna(0, inplace=True) 

for year in range(2012, 2015):
    sns.distplot(sal[sal['Year']==year]['BasePay'], hist=False, label=str(year))

Is there a correlation between length of the Job Title string and Salary?

The corr and heatmap show that there's almost not any correlation between these two labels. (-0.03 is far less than 1)

Recap: you did good job in overall ! Most of the answers are correct and you understand the subject.

[anderpham]

Assignment week 3 review

The customer with the highest money spent on purchases comes from Netherlands

The correct CustomerID is 14646. The problem with your answer is that you sorted by highest Purchases (Unit * Quantity) but forgot to groupby CustomerID, because a unique customer can have many transactions (Invoices) and the sort_values have to be made on sum of all transactions, you can do this way:

data['TotalPrice'] = data['Quantity'] * data['UnitPrice']
data_customer_Ne = data[data['Country'] == 'Netherlands'].groupby('CustomerID').sum().sort_values(by='TotalPrice', ascending=False)

data_customer_Ne.head(1)

On which year had the highest sales?

Your solution of using native Python way (define function to extract Year) is quite okay, but you can do it faster and write less code, by using a pandas function to_datetime

data['InvoiceDate'] = pd.to_datetime(data['InvoiceDate']) 
data['InvoiceYear'] = data['InvoiceDate'].dt.year
data.groupby('InvoiceYear')['Purchase'].sum().sort_values(ascending=False)

For the rest (chart on how many order per hours/month, 10 countries that have the most 'Quantity' ordered, etc )

I found an interesting notebook resource which is well explained very similar to this assignment, let check:
https://www.kaggle.com/kernels/scriptcontent/5897749/download