COM6018 Data Science with Python

Week 10 - More assignment feedback

Jon Barker

Overview

• Group Feedback from Assignment 2 - Q3 and Q4
• Review of Lab Class 8 (Curve Fitting with Scikit-Learn)
• Bits and Bobs
• Assignment 2

Group Feedback from Assignment 1

Things to consider

• Is the data correct?
• Do the plots allow easy comparison?
• Are the fonts large enough?
• Is the title clear?
• Have the axes been labeled?
• Are the axes units clear?
• Have good line styles been used?
• Are the lines labeled clearly?

Q3

• Plot the per capita CO₂ emissions against the GDP per capita for each country.

• Design your plot so that the size of the marker is proportional to the population.

• Only consider countries with a population of at leat 5 million people.

• Comment on the relationship between the two variables and how it has changed over time.

Q3 plots

Q3 plots

Q3 plots

Q3 plots

Q3 plots

Q3 plots

Q3 plots

Q3 plots

Q3 plots

Q3 plots

Q4

• Make a plot that compares the distribution of GDP per capita across the countries
  in the world at 10-yearly intervals from 1950 to 2020.
• Comment on how the distribution has changed over time.
• In particular, does it appear that wealth inequality has increased or decreased over time

Q4 plots

Q4 plots

Q4 plots

Q4 plots

Q4 plots

Q4 plots

Q4 plots

Review of Lab Class 8

Review of Lab Class 8

Using Scikit-Learn to fit the CO₂ atmospheric concentration data.

We will then use our fitting model to predict future CO₂ atmospheric concentration.

On what date will the CO₂ atmospheric concentration reach 450 ppm?

The Task

The stages of the task are as follows,

• Loading the CO₂ data.
• Fitting a polynomial curve to describe the growth trend.
• Fitting a periodic function to describe the seasonal variation.
• Tuning the model hyperparameters (order of polynomial, etc).
• Evaluating the model.
• Using the model to make a prediction.

The Solution Notebook

The solutions to the lab have been released.

Open the Solution Notebook

Bits and Bobs

Persisting Models in Scikit-Learn

• Once trained a model can be saved to disk and then loaded again later.

• We can use this to distrubute a trained model to other people.

• Two approaches to saving a model in Scikit-Learn,

  • Pickle
  • Joblib

• The preferred approach is Joblib.

Saving and loading a model with pickle

pickle is a standard Python library for saving and loading Python objects to disk.

If we have a model called model we can save it to disk with,

import pickle

pickle.dump(model, open('model.pkl', 'wb'))

Then we can load the model later with,

import pickle

# Read the pickle file from disk
model = pickle.load(open('model.pkl', 'rb'))

Saving and loading a model with joblib

If we have a model called model we can save it to disk with,

import joblib

joblib.dump(model, 'model.joblib')

Then we can load the model later with,

import joblib

model = joblib.load('model.joblib')

# or using a file handle
model = joblib.load(open('model.joblib', 'rb'))

Next Steps

• Complete/Review Lab Class 8:

  • Complete the Curve Fitting lab.

• Lab Class on Wednesday:

  • Will provide support for getting started with the assignment.