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

feat: add custom min/max to task-level plots #58

Open
wants to merge 13 commits into
base: main
Choose a base branch
from

Conversation

JemmaLDaniel
Copy link
Contributor

@JemmaLDaniel JemmaLDaniel commented Mar 4, 2024

What?

  1. I added an option for a custom minimum or maximum for each task and metric to data_process_pipeline. This allows us to normalise in a more meaningful way when we know the theoretical limits to a metric.
  2. I also enforced a y-axis from 0 to 1 in task-level plotting. I believe a reader can interpret performance differences faster when graph axes are on a common scale.
  3. Added breathing room on the top of the single task plots (included in the most recent commits).
  4. Some very small spelling standardisation in comments.

Why?

  1. This allows us to normalise in a more meaningful way when we know the theoretical limits to a metric.
  2. I believe a reader can interpret performance differences faster when graph axes are on a common scale. This also becomes more necessary now that our task-level graphs now have potential to not always sit between 0 and 1 if they do not hit the theoretical metric limits.
  3. Previously, a normalised return of 1 and its upper confidence interval would be cut off at the top of the plot. Now, we can easily read these values off the plot.
  4. N/A

How?

  1. I added a dictionary to pass in custom minimums and maximums for a specific task and metric, which, where relevant, data_process_pipeline uses to override the 'local' minimum and maximum it may find in each case.
  2. I added a boolean to plot_single_task_curve to check if normalisation is required and set axes to [0, 1] if True.
  3. I add a variable to track the current highest observed confidence interval value, and set the y-axis limit to this, instead of always using 1 for normalised plots.
  4. N/A

@JemmaLDaniel JemmaLDaniel self-assigned this May 2, 2024
@JemmaLDaniel JemmaLDaniel added the enhancement New feature or request label May 2, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants