Plotting #2338
Draft
Plotting #2338
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
codingWhale13
added
enhancement
lostella
reviewed
Oct 4, 2022
|
another potential use case is plotting some of the variates in the multivariate forecasting case... can we get a helper for that too? |
|
Thanks for the input, @kashif. 👍 The |
jaheba
reviewed
Oct 6, 2022
|
Can you check if this will also work for Quantile forecasts? |
It does not. 😄 PR #2352 should fix this. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description of changes:
This PR addresses #2333 and proposes a
plotfunction that is given a forecast and optionally a time series for reference as well as a bunch of options about how to plot things.The most interesting part is plotting the different prediction intervals. This now works pretty much as in
Forecast.plot. The only major difference is that the lower and upper parts of each prediction interval are created separately to keep the colors true (in Forecast.plot, when plotting a lot of prediction intervals, the colors in the middle get very dark because thealphaof the color gets less with each overlap).The
plot_prob_forecastsused in this tutorial can be replaced by a call tolab.plot. Also, the vertical red line in the tutorial can be created using thetrain_test_separatorparameter.It might make sense to return the Matplotlib
figandaxso that the user can draw more custom things on top of it (like highlighting the train vs. test data in different colors as is done here).This proposal would pretty much replace the
plotmethod ofForecast. I think this makes sense because plotting is something separate and the Forecasts would be more lightweight this way.Biggest caveat at the moment: 150 lines is a lot of code to just plot a forecast. Just providing all the different knobs for the user to choose (like color, legend position etc.) makes things a bit messy but not providing them would render the
plotfunction pretty useless, I think. So, maybe there's an easier way to do this.By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.