Code
# Import libraries
import polars as pl
import duckdb as db
import matplotlib.pyplot as plt
import seaborn as sns
import json
plt.style.use('ggplot')
Introduction
LEGO is a Danish toy production company. The company is best known for its colorful interlocking plastic bricks, and the vast possibilities of what can be built with them. The Lego Group also produces a variety of other toys, including board games, video games, and clothing.
In addition to its traditional brick-based toys, Lego also produces a variety of other products, including:
- Lego Technic: A line of more complex, motorized sets that allow users to build working vehicles and machines.
- Lego Creator: A line of sets that allow users to build a variety of different models, from buildings to vehicles to animals.
- Lego City: A line of sets that allow users to build cityscapes and play out scenes from everyday life.
- Lego Friends: A line of sets that are specifically designed for girls, and feature female characters and settings.
- Lego Ninjago: A line of sets that are based on the popular Lego Ninjago: Masters of Spinjitzu television series.
- The Lego Movie: A computer-animated comedy film that was released in 2014, and is based on the Lego brand.
Lego products are popular with children of all ages, and they are also enjoyed by adults. Lego bricks are a great way to encourage creativity and problem-solving skills, and they can also be used to build models of just about anything you can imagine.
Lego was derived from the Danish phrase leg godt, which means “play well”.
In this project, we will show a summary of lego blocks.
Reading datasets
A comprehensive database of lego blocks is provided by Rebrickable.
The data is available as csv file and the schema is shown below
Database schema
Let us start by reading in the colors data to get a sense of the diversity of lego sets!
Environment settings
Code
# get token
filename = 'credentials.json'
# read json file
with open(filename) as f:
keys = json.load(f)
# read credentials
token = keys['md_token']Code
# connect to motherduck cloud
conn = db.connect(f'md:?motherduck_token={token}')Code
# retrieve dataframe
colors = conn.execute('select * from projects.colors').pl()Code
colors.head()
shape: (5, 4)
| id | name | rgb | is_trans |
|---|---|---|---|
| i64 | str | str | bool |
| -1 | "Unknown" | "0033B2" | false |
| 0 | "Black" | "05131D" | false |
| 1 | "Blue" | "0055BF" | false |
| 2 | "Green" | "237841" | false |
| 3 | "Dark Turquoise" | "008F9B" | false |
Exploring the Colors dataset
Now that we have read the colors dataset, we can start exploring it! Let us start by understanding the number of colors available.
Code
# How many distinct colors are available?
colors.shape(135, 4)Transparent Colors in Lego Sets
The colors data has a column named is_trans that indicates whether a color is transparent or not.
We shall to explore the distribution of transparent vs. non-transparent colors.
Code
# Distribution of colors based on transparency
colors_summary = colors.group_by('is_trans').agg(pl.len())
colors_summary
shape: (2, 2)
| is_trans | len |
|---|---|
| bool | u32 |
| true | 28 |
| false | 107 |
Exploring Lego Sets
Another interesting dataset available in this database is the sets data.
It contains a comprehensive list of sets over the years and the number of parts that each of these sets contained.
Sets data
Let us use this data to explore how the average number of parts in Lego sets has varied over the years.
Code
# Read sets data
sets = conn.sql('select * from projects.sets').pl()
sets.head()
shape: (5, 5)
| set_num | name | year | theme_id | num_parts |
|---|---|---|---|---|
| str | str | i64 | i64 | i64 |
| "00-1" | "Weetabix Castle" | 1970 | 414 | 471 |
| "0011-2" | "Town Mini-Figures" | 1978 | 84 | 12 |
| "0011-3" | "Castle 2 for 1 Bonus Offer" | 1987 | 199 | 2 |
| "0012-1" | "Space Mini-Figures" | 1979 | 143 | 12 |
| "0013-1" | "Space Mini-Figures" | 1979 | 143 | 12 |
Code
# Create a summary of average number of parts by year
parts_by_year = sets.group_by('year').agg(pl.col('num_parts').mean())
parts_by_year.head()
shape: (5, 2)
| year | num_parts |
|---|---|
| i64 | f64 |
| 1983 | 116.105263 |
| 2001 | 104.365782 |
| 2013 | 181.359191 |
| 1995 | 179.039062 |
| 1977 | 91.5 |
Lego Themes Over Years
Lego blocks ship under multiple themes.
Let us try to get a sense of how the number of themes shipped has varied over the years.
Code
# Number of themes shipped by year
sets.head()
shape: (5, 5)
| set_num | name | year | theme_id | num_parts |
|---|---|---|---|---|
| str | str | i64 | i64 | i64 |
| "00-1" | "Weetabix Castle" | 1970 | 414 | 471 |
| "0011-2" | "Town Mini-Figures" | 1978 | 84 | 12 |
| "0011-3" | "Castle 2 for 1 Bonus Offer" | 1987 | 199 | 2 |
| "0012-1" | "Space Mini-Figures" | 1979 | 143 | 12 |
| "0013-1" | "Space Mini-Figures" | 1979 | 143 | 12 |
Code
themes_by_year = sets.group_by('year', maintain_order=True).agg(pl.col('theme_id').len()).sort('year')
themes_by_year.tail()
shape: (5, 2)
| year | theme_id |
|---|---|
| i64 | u32 |
| 2013 | 593 |
| 2014 | 715 |
| 2015 | 670 |
| 2016 | 609 |
| 2017 | 470 |
Code
# Plot trends in average number of parts by year
plt.figure(figsize=(11, 7))
plt.plot(themes_by_year['year'], themes_by_year['theme_id'], marker='o', linestyle='-')
plt.title('Lego - Evolution of average number of parts (1950-2017)', fontsize = 18)
plt.xlabel('')
plt.ylabel('Themes')
plt.legend('',frameon=False)
plt.show()
Code
# close connection
conn.close()Contact
Jesus LM
Economist & Data Scientist