Example discrete markov chain weather#702
Conversation
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #702 +/- ##
===========================================
+ Coverage 51.60% 91.11% +39.50%
===========================================
Files 73 90 +17
Lines 8003 8418 +415
===========================================
+ Hits 4130 7670 +3540
+ Misses 3873 748 -3125 🚀 New features to boost your workflow:
|
|
Looks pretty good, left just a suggestion to make it more atemporal |
|
Thanks @ricardoV94 , I removed "old" and "new" mentions. I hope I did not miss anything |
|
sorry @ricardoV94 I had forgot to push the "now" removals |
| @@ -0,0 +1,1223 @@ | |||
| { | |||
There was a problem hiding this comment.
|
Looks great, just needs provenance. I would consider using a PyMC model to generate the data (from a prior draw) then use pm.observe to estimate it, since that's the workflow we want to promote. Not a deal-breaker for me but I'd like to stop suggesting users need to write their models twice, once in numpy then again in pymc. |
|
Would be cool to use real weather data! |
Do you have some? @juanitorduz this should be in pymc-examples, sorry if it wasn't clear. Notebooks here are invisible |
ok (for now I will keep synthetic data). Do we wanna have a copy here ? I will open a PR on pymc-examples 👍 |
|
nah no copy here |
|
@juanitorduz this looks good, wanna move it to pymc-examples? |
yes! I will so it sometime this week 🙏 |
|
Closing in favor of pymc-devs/pymc-examples#889 |
5 participants
Example of the feature added in #693
Example provided by @ricardoV94