Please enable JavaScript.
Coggle requires JavaScript to display documents.
Applied Machine Learning This checklist can guide you through your…
Applied Machine Learning
This checklist can guide you through your Machine Learning projects.
8
2
DATA COLLECTION
Get the Data
Note:
Automate as much as possible so you can easily get fresh data.
Find and document where you can get the data.
Check legal obligations and get authorisation if necessary. : : :
Get access authorisations.
Create a workspace (with enough storage space).
Get the data.
List the data that you need and how much you need.
Check how much space it will take.
Convert the data to a format you can easily manipulate (without changing the data itself).
Ensure sensitive information is deleted or protected (e.g. anonymised).
Check the size and type of data (time series, sample, geographical, etc.).
Sample a test set, put it aside, and never look at it.
3
DATA ANALYSIS
Explore the Data
Note:
Try to get insights from a field expert for these steps.
Create a copy of the data for exploration (sampling it down to a manageable size if necessary)
7
1
PROBLEM DEFINITION
Frame the Problem - Look at the Big Picture
What are the current solutions / workaround (if any)?
How should you frame this problem (supervised/unsupervised, online/offline, etc.)?
How should performance be measured?
How will your solution be used?
Is the performance measure aligned with the business objective?
Define the objective in business terms.
What would be the minimum performance needed to reach the business objective?
What are comparable problems? Can you reuse experience or tools?
Is human expertise available?
How would you solve the problem manually?
List the assumptions you (or others) have made so far.
Verify assumptions if possible.
4
5
6
ML is the science and the art of programming computers so they can learn from data. Machine learning algorithm use vectors to make predictions. We can understand ML as a suite of statistical algorithms and the study of them. The purpose of these algorithms is to learn from previous data by decreasing the error of their predictions. In ML we often refer to the result of our algorithm as a model.