Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Ml 10 develop a baseline model #4

Merged
merged 1 commit into from
Nov 25, 2024

Conversation

promatty
Copy link
Contributor

@promatty promatty commented Nov 6, 2024

No description provided.

@promatty promatty requested a review from justin-phxm November 6, 2024 02:01
@promatty promatty force-pushed the ML-10-Develop-a-baseline-model branch from 47d458d to aa97ade Compare November 6, 2024 02:58
Copy link
Contributor

@justin-phxm justin-phxm left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

  • Your Analysis looks good. I would throw it into ChatGPT to clean it up though.
    There's a few things we can do to improve it.
  1. Clean the data
  2. use different algorithms
  3. consider using a meta-model
  4. train on different subsets of the training data (cross validation)
  5. ask chatGPT lmao

@promatty promatty force-pushed the ML-10-Develop-a-baseline-model branch from aa97ade to 64da499 Compare November 6, 2024 03:12
@promatty promatty force-pushed the ML-10-Develop-a-baseline-model branch from 64da499 to 77eb9cd Compare November 20, 2024 02:22
@promatty promatty requested a review from justin-phxm November 20, 2024 02:24
@promatty promatty force-pushed the ML-10-Develop-a-baseline-model branch from 77eb9cd to 21c8d8e Compare November 20, 2024 02:32
Copy link
Contributor

@justin-phxm justin-phxm left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Even after reevaulation of performance, there should be no negative minimum distance:
image

Great summary on the various regression algorithms. My prediction is that Random Forest is going to perform the best.
image

Can you also include an interpretation of the data at the end... What does this data mean in the real world? How can we make data driven decisions from this? For example, maybe an explanation of how this model can predict the distance of a given set of data with +-10m accuracy.
image

You can honestly remove this one section as it is not very relevant. Instead, add a section at the top about how we want a regression algorithm instead of a classification model.
image

You can also turn this into a function as it is duplicated multiple times.
image

I believe it is standard practice to include all the imports at the beginning of the file. Please move this to the top.
image

Look into Hyperparameter tuning... We won't implement this in the current ticket but we will definitely use it in the future to evaluate our models. It's not nearly as important in LinearRegression models but will be huge when it comes to decisionTrees/Forests.

@promatty promatty force-pushed the ML-10-Develop-a-baseline-model branch from 21c8d8e to 01e606e Compare November 23, 2024 22:10
@promatty promatty requested a review from justin-phxm November 23, 2024 22:12
Copy link
Contributor

@justin-phxm justin-phxm left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I was hoping that removing the unrealistic negative values would decrease MAE but it's fine. This can happen sometimes

@justin-phxm justin-phxm merged commit bdcae23 into master Nov 25, 2024
1 check passed
@justin-phxm justin-phxm deleted the ML-10-Develop-a-baseline-model branch November 25, 2024 03:40
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants