Skip to content

See how complex ML models can learn naive strategies of forecasting stock price movements

Notifications You must be signed in to change notification settings

ruankie/naive-ml

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

37 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

The naivety of ML models and how to avoid the naive trap

Read the accompanying blog post for more details.

When forecasting stock price movements with fancy ML models, you want your models to learn something useful from the provided data that humans might not be aware of so you have a competitive edge. This repo shows examples of where those fancy ML models learn something dumb that looks intelligent and how to check if your models are doing the same.

How to use

  1. Clone this repo onto your machine
  2. Open development environment using one of the following methods:
    • Open in VS code using the Remote-Containers extension
    • Manually reproduce the development environment using the Dockerfile in .devcontainer/
    • Install the requiremetns your local environment by running pip install -r requirements.txt
  3. Navigate to notebooks/full-example.ipynb to see a step-by-step example or to reproduce the results
  4. (Optional) Browse through the code in src/
  5. (Optional) Add your own custom models to src/models.py

About

See how complex ML models can learn naive strategies of forecasting stock price movements

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published