stripe / rainier

Bayesian inference in Scala.
https://rainier.fit
Apache License 2.0
432 stars 51 forks source link

be explicit about parameters vs placeholders #426

Closed avibryant closed 4 years ago

avibryant commented 4 years ago

This just makes it clear where we're using Parameter vs Placeholder vs the generic Variable.

It also deletes some of the Viz stuff to be reintroduced later with support for placeholders and bounds.

This is setting things up for some larger changes around Placeholder.

codecov-io commented 4 years ago

Codecov Report

Merging #426 into 0.3-dev will increase coverage by 1.69%. The diff coverage is 70.37%.

Impacted file tree graph

@@             Coverage Diff             @@
##           0.3-dev     #426      +/-   ##
===========================================
+ Coverage    56.47%   58.16%   +1.69%     
===========================================
  Files           69       68       -1     
  Lines         2727     2644      -83     
  Branches       144      144              
===========================================
- Hits          1540     1538       -2     
+ Misses        1187     1106      -81
Impacted Files Coverage Δ
.../scala/com/stripe/rainier/bench/SBCBenchmark.scala 0% <ø> (ø) :arrow_up:
...com/stripe/rainier/bench/RegressionBenchmark.scala 0% <0%> (ø) :arrow_up:
.../main/scala/com/stripe/rainier/core/Estimate.scala 0% <0%> (ø) :arrow_up:
...c/main/scala/com/stripe/rainier/plot/Jupyter.scala 0% <0%> (ø) :arrow_up:
...in/scala/com/stripe/rainier/core/Categorical.scala 44.44% <0%> (ø) :arrow_up:
...in/scala/com/stripe/rainier/compute/Compiler.scala 92.59% <100%> (ø) :arrow_up:
...in/scala/com/stripe/rainier/compute/Gradient.scala 93.22% <100%> (-0.23%) :arrow_down:
...main/scala/com/stripe/rainier/core/Generator.scala 68.18% <100%> (ø) :arrow_up:
...src/main/scala/com/stripe/rainier/core/Model.scala 75.86% <100%> (+11.15%) :arrow_up:
...c/main/scala/com/stripe/rainier/compute/Real.scala 83.33% <100%> (+4.32%) :arrow_up:
... and 4 more

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 0bf50bd...b6a15a5. Read the comment docs.

avibryant commented 4 years ago

Nothing here that I think needs review in detail; this is more just for your awareness. The next PR will be more interesting.