Closed Shreyanand closed 3 years ago
Check out this pull request on
See visual diffs & provide feedback on Jupyter Notebooks.
Powered by ReviewNB
/approve
[APPROVALNOTIFIER] This PR is APPROVED
Approval requirements bypassed by manually added approval.
This pull-request has been approved by: MichaelClifford
The full list of commands accepted by this bot can be found here.
The pull request process is described here
View / edit / reply to this conversation on ReviewNB
chauhankaranraj commented on 2021-04-07T14:27:27Z ----------------------------------------------------------------
Is there a particular reason for using these set of params i.e. (3, 1, 2)
and (3, 1, 3)
for this particular dataset, or is it more like trial and error? In either case, should we mention somewhere why these were chosen?
Shreyanand commented on 2021-04-07T14:49:07Z ----------------------------------------------------------------
Yup, so in the last notebook, we looked at how to find these parameters and we found these ones to have better AICs than the rest. Maybe I could add that in the markdown somewhere.
View / edit / reply to this conversation on ReviewNB
chauhankaranraj commented on 2021-04-07T14:27:28Z ----------------------------------------------------------------
Should this cell have been a markdown-type cell?
Shreyanand commented on 2021-04-07T14:50:21Z ----------------------------------------------------------------
Yes, sloppy :D
View / edit / reply to this conversation on ReviewNB
chauhankaranraj commented on 2021-04-07T14:27:28Z ----------------------------------------------------------------
Same thing here, should the reasoning for selecting (3, 1, 3)
and (3, 1, 3, 4)
be mentioned here?
@Shreyanand looks like the PR got merged before I could submit review :cry:
View / edit / reply to this conversation on ReviewNB
chauhankaranraj commented on 2021-04-07T14:27:28Z
Same thing here, should the reasoning for selecting
(3, 1, 3)
and(3, 1, 3, 4)
be mentioned here?
+1 I think it will be valuable to add reasoning since these parameters might change once we change the dataset, say next month.
Yup, so in the last notebook, we looked at how to find these parameters and we found these ones to have better AICs than the rest.
View entire conversation on ReviewNB
Related Issues and Dependencies
9
This introduces a breaking change
Description
In this notebook, I compare several forecasting techniques for an example dataset of a cluster metric.