-
Notifications
You must be signed in to change notification settings - Fork 28
Runs: update semantic colors (now supports config key-value pairs) #1420
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
Conversation
Deploying docs with
|
Latest commit: |
0b06fb5
|
Status: | ✅ Deploy successful! |
Preview URL: | https://2c392340.docodile.pages.dev |
Branch Preview URL: | https://semantic-color-config-update.docodile.pages.dev |
Images automagically compressed by Calibre's image-actions ✨ Compression reduced images by 44.7%, saving 152.34 KB.
448 images did not require optimisation. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Some small copy suggestions, which you should ignore if they are incorrect. :)
@@ -7,23 +7,23 @@ menu: | |||
title: Semantic run plot legends | |||
--- | |||
|
|||
Create meaningful legends for run line plots to visually distinguish them based on metrics you log to W&B. Identify trends and patterns based on the highest, lowest, or latest values of a metric across your training and evaluation runs. Runs are bucketed and colored in plots based on the plot's configured metric, Y value, and number of buckets. | |||
Create meaningful legends for line plots to visually distinguish them based on a metric or run configuration you log to W&B. Identify trends and patterns based on the highest, lowest, or latest values of a metric key-value or run configuration's key-value pairs across your training and evaluation runs. Runs are bucketed and colored based on the plot's defined metric key-value or run configuration key-value pair, Y value, and number of buckets. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This could use some copy massaging: "...values of a metric key-value or run configuration's key-value pairs..."
… into semantic_color_config_updates
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
One nitty nit nit :)
@@ -96,7 +96,7 @@ Consider the following: | |||
- **Key** is set to `"Accuracy"` (abbreviated as `"acc"`). | |||
- **Y value** is set to `"Max"` | |||
|
|||
With this set up, W&B colors run based on their accuracy values, using the maximum accuracy value logged for each run. | |||
With this configuration, W&B colors each run based on their accuracy values. The colors vary from a light yellow color to a deep color. Lighter colors represent lower accuracy values, while deeper colors represent higher accuracy values. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
With this configuration, W&B colors each run based on their accuracy values. The colors vary from a light yellow color to a deep color. Lighter colors represent lower accuracy values, while deeper colors represent higher accuracy values. | |
With this configuration, W&B colors each run based on its accuracy values. The colors vary from a light yellow color to a deep color. Lighter colors represent lower accuracy values, while deeper colors represent higher accuracy values. |
Migrating from Metric-based colors and Metric to Key-based colors and Key to support metrics + config values. Config values refering to dictionary we pass as an argument to config parameter in
wandb.init()
. e.g.Jira ticket: https://wandb.atlassian.net/browse/DOCS-1598
Status: Waiting for green light to merge.