Skip to content

v0.8.0

Compare
Choose a tag to compare
@julia-tagbot julia-tagbot released this 05 Jan 06:10
· 668 commits to master since this release
v0.8.0

v0.8.0 (2020-01-04)

Diff since v0.7.4

Closed issues:

  • Recommend memoization in docs for expensive functions in models when Gibbs sampling (#1024)
  • Missing Pkg in dependencies (#1020)
  • Numerical errors on Turing v0.7.4 (#1017)
  • Problem with random switch (#1013)
  • Error with Gibbs compositional sampler and an array (#1004)
  • Elliptical slice sampling (#991)
  • Simple syntax for querying the model and posterior sample (#989)
  • Unifying assume and observe and handling missing data and user-input variable names (#972)
  • Issues in Intro (#942)
  • Cannot compile Turing on fresh Julia install (#936)
  • Keyword arugments in @model are not treated as data (#911)
  • Sampling from an InverseWishart with HMC (#883)
  • Too much macro dependence? (#820)
  • Log probability of an observation (#817)
  • Model syntax: provide automatically all random variables as argument? (#792)
  • Use ModelingToolkit for frontend or IR? (#789)
  • Observe vectorisation issue. (#761)
  • Assume in vectorisation of HMC bug. (#760)
  • Function to add to logp within @model Turing model (#754)
  • Unifying error messages. (#747)
  • Option to reject samples that cause errors (#703)
  • Why does @model build a declaration instead of a value? (#667)
  • array comprehension for random variables (#604)
  • Compiler generated aliases (#533)
  • Compiler does not pass on the sampler (#532)
  • @varname can be replaced with a function (#514)
  • Refactoring VarName: new indexing scheme (#484)
  • Change broadcasting for ~ macro (#476)
  • Evaluating the posterior (#400)
  • Specialised inference (#313)

Merged pull requests: