Materials associated with the MLBoost YouTube channel: https://www.youtube.com/@MLBoost. The channel features concise videos aimed at presenting the core concepts in a focused manner. By delving into the code provided in this repository subsequent to watching the videos, you can gain a deeper understanding of the subject matter and gain hands-on experience. I have made an effort to implement the content from scratch, utilizing numpy exclusively, to facilitate a comprehensive grasp of the intricacies involved.
| No | Video | Title and What you Will Learn |
|---|---|---|
| 1 |  |
Reliable Forecasting: Evaluation - MAPE - First item - Second item |
| 2 |  |
Reliable Forecasting: Evaluation - adjustedMAPE - First item - Second item |
| 3 |  |
A Forecasting Competition - First item - Second item |
| 4 | |
Best Possible Model May Lose to a Naive One if Evaluation Metric is Not Consistent with - First item - Second item |
| 5 |  |
In ML Competitions, when the Error is MSE, Submit the Expected Value of Inferred Distribution. - First item - Second item |
| 6 |  |
In ML Competitions, when the Error is MAE, Submit the Median of Inferred Distribution. - First item - Second item |
| 7 |  |
Under Absolute Percentage Error loss, a Non-conventional Median is Optimal! - First item - Second item |
If you have any questions about the contents of this repo, feel free to post it as a video comment or a GitHub issue. I will make sure to reply asap.
- TODO
- At the begining of each notebook add a Table of Content (see the notebook of Episode 8 for an example.).
- Complete "What you will learn" section for each episode in the ReadMe.
- TORESEARCH
- Items with this tag are potential research opportunities.
conda env create -f MLBoost.yml