Employer Set Project

Pre-release material comes out before the tasks start. It introduces the employer's scenario, so read it carefully and make sure you understand the business, its users and the problem.

Key dates

From the course key dates (provisional, so check with your teacher).

The ESP tasks cover planning, fixing defects, design, development and reflective evaluation. This table shows which parts of the guide help most with each. It's revision guidance, not the official mark scheme.

The spec lists these for the Employer Set Project only: pandas (import CSV, data frames, totals, averages, counts, trends) and matplotlib (graphs with labels and legends). Press ▶ Run on each example to try it: the first run loads pandas, which takes a little longer. Run the examples in order, because example 1 creates the file the others read.

1. Import a CSV into a data frame

A data frame is a table: rows are records, columns are fields. pd.read_csv() turns a CSV file into one, and works out the data types for you.

import pandas as pd

with open("sales.csv", "w") as f:
    f.write("date,branch,item,quantity,price\n")
    f.write("2027-05-03,Leeds,Latte,12,3.20\n")
    f.write("2027-05-03,York,Tea,8,1.80\n")
    f.write("2027-05-04,Leeds,Scone,5,2.50\n")
    f.write("2027-05-04,York,Latte,9,3.20\n")
    f.write("2027-05-05,Leeds,Latte,15,3.20\n")
    f.write("2027-05-05,York,Scone,11,2.50\n")

sales = pd.read_csv("sales.csv")
print(sales.head(3))
print(len(sales), "rows,", len(sales.columns), "columns")

         date branch   item  quantity  price
0  2027-05-03  Leeds  Latte        12    3.2
1  2027-05-03   York    Tea         8    1.8
2  2027-05-04  Leeds  Scone         5    2.5
6 rows, 5 columns

head(3) shows the first 3 rows. The numbers down the left are the index. Unlike the plain csv module (6.5), pandas reads numbers as numbers, not strings.

2. Totals, averages and counts

import pandas as pd

sales = pd.read_csv("sales.csv")
sales["total"] = sales["quantity"] * sales["price"]

print("Total takings: £" + format(sales["total"].sum(), ".2f"))
print("Average quantity:", round(sales["quantity"].mean(), 1))
print("Biggest sale: £" + format(sales["total"].max(), ".2f"))
print(sales["item"].value_counts())

Total takings: £169.60
Average quantity: 10.0
Biggest sale: £48.00
item
Latte    3
Scone    2
Tea      1
Name: count, dtype: int64

3. Group, filter and sort

import pandas as pd

sales = pd.read_csv("sales.csv")
sales["total"] = sales["quantity"] * sales["price"]

print(sales.groupby("branch")["total"].sum())
print()
lattes = sales[sales["item"] == "Latte"]
print(lattes[["date", "branch", "quantity"]])
print()
print(sales.sort_values("total", ascending=False).head(2))

branch
Leeds    98.9
York     70.7
Name: total, dtype: float64

         date branch  quantity
0  2027-05-03  Leeds        12
3  2027-05-04   York         9
4  2027-05-05  Leeds        15

         date branch   item  quantity  price  total
4  2027-05-05  Leeds  Latte        15    3.2   48.0
0  2027-05-03  Leeds  Latte        12    3.2   38.4

4. Show a trend with a labelled graph

import pandas as pd
import matplotlib.pyplot as plt

sales = pd.read_csv("sales.csv")
sales["total"] = sales["quantity"] * sales["price"]
daily = sales.groupby(["date", "branch"])["total"].sum().unstack()
print(daily)

daily.plot(kind="line", marker="o")
plt.title("Daily takings by branch")
plt.xlabel("Date")
plt.ylabel("Takings (£)")
plt.legend(title="Branch")
plt.tight_layout()
plt.savefig("takings.png")
print("Chart saved as takings.png")

branch      Leeds  York
date
2027-05-03   38.4  14.4
2027-05-04   12.5  28.8
2027-05-05   48.0  27.5
Chart saved as takings.png

When you run it here, the chart appears under the output. On your own computer, plt.show() opens it in a window instead of saving it.

Each program has defects. For each one, give the line number, the problem, the type of error (syntax, runtime or logic) and the corrected line. Press ▶ Run to watch the faulty code fail, then fix it in the editor until it works.

scores = [72, 45, 50, 88, 39]
total = 0
passes = 0
for score in scores:
    total = total + score
    if score > 50:
        passes = passes + 1
average = total / len(score)
print("Average: " + average)
print("Scores of 50 or more:", passes)

scores = [72, 45, 50, 88, 39]
total = 0
passes = 0
for score in scores:
    total = total + score
    if score >= 50:
        passes = passes + 1
average = total / len(scores)
print("Average: " + str(average))
print("Scores of 50 or more:", passes)

Average: 58.8
Scores of 50 or more: 3

colours = ["red", "blue", "red", "green", "blue", "red"]
counts = {}
for colour in colours:
    if colour in counts:
        counts[colour] = 1
    else:
        counts[colour] = counts[colour] + 1
for colour, number in counts.items:
    print(colour, number)

colours = ["red", "blue", "red", "green", "blue", "red"]
counts = {}
for colour in colours:
    if colour in counts:
        counts[colour] = counts[colour] + 1
    else:
        counts[colour] = 1
for colour, number in counts.items():
    print(colour, number)

red 3
blue 2
green 1

age = int(input("Age: "))
while age < 16 or age > 99:
    print("Age must be 16 to 99")
print("Accepted")

while True:
    entry = input("Age: ")
    if not entry.isdigit():
        print("Enter a whole number")
    elif not 16 <= int(entry) <= 99:
        print("Age must be 16 to 99")
    else:
        break
print("Accepted:", entry)

Age: ten
Enter a whole number
Age: 12
Age must be 16 to 99
Age: 17
Accepted: 17

An evaluation judges how well your solution meets the brief and what you'd do differently. Like an 'evaluate' exam question, it needs evidence, both strengths and weaknesses, and a supported judgement.

• I can use the Appendix 2 Python commands without looking them up
• I can read an error message and find the faulty line quickly
• I can read a CSV with pandas and work out totals, averages and counts
• I can group, filter and sort a data frame
• I can draw a graph with a title, axis labels and a legend
• I can write a test plan with valid, boundary, invalid and erroneous data
• I can draw a flowchart and a data model for a scenario
• I can write a balanced evaluation backed up by evidence
• I save my work often, with clear file names and comments in my code