Soss v0.20.3 Diff since v0.20.2 <strong>Closed issues:</strong> No more Soss.predict(... more Soss v0.20.3 Diff since v0.20.2 <strong>Closed issues:</strong> No more Soss.predict() methods for densities from Distributions.jl (#294) <strong>Merged pull requests:</strong> <code>predict</code> for Disitributions (#295) (@cscherrer)
Soss v0.18.3 Diff since v0.18.2 <strong>Closed issues:</strong> DAG construction with... more Soss v0.18.3 Diff since v0.18.2 <strong>Closed issues:</strong> DAG construction with local variables (#253) <strong>Merged pull requests:</strong> Scoping for local variables (#260) (@cscherrer)
Soss v0.17.0 Diff since v0.16.2 <strong>Closed issues:</strong> Leaky <code>M&l... more Soss v0.17.0 Diff since v0.16.2 <strong>Closed issues:</strong> Leaky <code>M</code> (#67) Double citation for SymPy (#101) Nested models (#130) Clean up SymPy output (#138) Move all the Python stuff to the new <code>SossSymPy.jl</code> package (#170) Nested traces (#227) <strong>Merged pull requests:</strong> CompatHelper: bump compat for "SpecialFunctions" to "1.0" (#224) (@github-actions[bot]) Dev (#233) (@cscherrer) WIP: add basemeasure for models (#239) (@mschauer) Dev (#244) (@cscherrer) fix leaky <code>M</code> (#249) (@thautwarm) Dev (#254) (@cscherrer)
Soss v0.14.0 Diff since v0.13.1 <strong>Closed issues:</strong> use comprehensions in... more Soss v0.14.0 Diff since v0.13.1 <strong>Closed issues:</strong> use comprehensions instead of <code>For(1:N)</code>? (#14) What am I doing wrong? (#166) TODO: Switch from Travis CI to GitHub Actions CI (#172) Add function to rewrite Sample statements to first assign rhs to a variable (#175) <strong>Merged pull requests:</strong> CompatHelper: bump compat for "FillArrays" to "0.9" (#160) (@github-actions[bot]) CompatHelper: bump compat for "LazyArrays" to "0.17" (#163) (@github-actions[bot]) README: Mention the SossMLJ work (#173) (@DilumAluthge) withdistributions (#176) (@cscherrer) add .~ notation (#177) (@cscherrer) Update toposort.jl (#179) (@millerjoey) CompatHelper: add new compat entry for "CategoricalArrays" at version "0.8" (#180) (@github-actions[bot]) Add and update multiple <code>[compat]</code> entries (#181) (@github-actions[bot]) CompatHelper: add new compat entr...
0 We present Soss, a declarative probabilistic programming language embedded in the Julia languag... more 0 We present Soss, a declarative probabilistic programming language embedded in the Julia language. Soss represents statistical models in terms of abstract syntax trees, and uses staged compilation for on-demand generation of ``inference primitives'' (random sampling, log-density, etc) without requiring casual users to worry about such details. The approach taken by Soss makes it easy to extend to take advantage of other packages in the rapidly-growing Julia ecosystem. At the time of this writing, Soss users can choose from several inference back-ends and connect easily with larger systems SymPy and Gen.
Soss v0.18.2 Diff since v0.18.1 <strong>Closed issues:</strong> Model composition (#2... more Soss v0.18.2 Diff since v0.18.1 <strong>Closed issues:</strong> Model composition (#258) Update <code>testvalue</code> (#259)
AdvancedHMC v0.3.0 Diff since v0.2.27 <strong>Closed issues:</strong> Modular design ... more AdvancedHMC v0.3.0 Diff since v0.2.27 <strong>Closed issues:</strong> Modular design for refreshing momentum variables (#13) Unify <code>Trajectory</code> type for static, dynamic HMC samplers. (#103) First-class support of sampling on a struct (#159) Return adapted mass matrix (#230) HMC gives GPU compilation error (#235) Error gradient (#243) Support for Complex and Matrix Parameters (#251) StanHMC-adaptor (#252) Compatibility with ComponentArrays.jl (#253) Support for complex numbers (#262) the example in the doc does not run (#264) Minimal example from README.md gives error with Beta distribution (#266) <strong>Merged pull requests:</strong> Added note to documentation of sample function (#236) (@trappmartin) Update for MCMCDebugging (#238) (@xukai92) GitHub Actions workflow to pull changes to dev (#239) (@xukai92) Refactoring termination criterion (#240) (@xukai92) Update for MCMCDebugging (#241) (@xukai92) Fixed default type for NesterovDual...
We present a generic framework for par-allel coordinate descent (CD) algorithms that includes, as... more We present a generic framework for par-allel coordinate descent (CD) algorithms that includes, as special cases, the orig-inal sequential algorithms Cyclic CD and Stochastic CD, as well as the recent paral-lel Shotgun algorithm. We introduce two novel parallel algorithms that are also spe-cial cases—Thread-Greedy CD and Coloring-Based CD—and give performance measure-ments for an OpenMP implementation of these. 1.
Soss v0.20.3 Diff since v0.20.2 <strong>Closed issues:</strong> No more Soss.predict(... more Soss v0.20.3 Diff since v0.20.2 <strong>Closed issues:</strong> No more Soss.predict() methods for densities from Distributions.jl (#294) <strong>Merged pull requests:</strong> <code>predict</code> for Disitributions (#295) (@cscherrer)
Soss v0.18.3 Diff since v0.18.2 <strong>Closed issues:</strong> DAG construction with... more Soss v0.18.3 Diff since v0.18.2 <strong>Closed issues:</strong> DAG construction with local variables (#253) <strong>Merged pull requests:</strong> Scoping for local variables (#260) (@cscherrer)
Soss v0.17.0 Diff since v0.16.2 <strong>Closed issues:</strong> Leaky <code>M&l... more Soss v0.17.0 Diff since v0.16.2 <strong>Closed issues:</strong> Leaky <code>M</code> (#67) Double citation for SymPy (#101) Nested models (#130) Clean up SymPy output (#138) Move all the Python stuff to the new <code>SossSymPy.jl</code> package (#170) Nested traces (#227) <strong>Merged pull requests:</strong> CompatHelper: bump compat for "SpecialFunctions" to "1.0" (#224) (@github-actions[bot]) Dev (#233) (@cscherrer) WIP: add basemeasure for models (#239) (@mschauer) Dev (#244) (@cscherrer) fix leaky <code>M</code> (#249) (@thautwarm) Dev (#254) (@cscherrer)
Soss v0.14.0 Diff since v0.13.1 <strong>Closed issues:</strong> use comprehensions in... more Soss v0.14.0 Diff since v0.13.1 <strong>Closed issues:</strong> use comprehensions instead of <code>For(1:N)</code>? (#14) What am I doing wrong? (#166) TODO: Switch from Travis CI to GitHub Actions CI (#172) Add function to rewrite Sample statements to first assign rhs to a variable (#175) <strong>Merged pull requests:</strong> CompatHelper: bump compat for "FillArrays" to "0.9" (#160) (@github-actions[bot]) CompatHelper: bump compat for "LazyArrays" to "0.17" (#163) (@github-actions[bot]) README: Mention the SossMLJ work (#173) (@DilumAluthge) withdistributions (#176) (@cscherrer) add .~ notation (#177) (@cscherrer) Update toposort.jl (#179) (@millerjoey) CompatHelper: add new compat entry for "CategoricalArrays" at version "0.8" (#180) (@github-actions[bot]) Add and update multiple <code>[compat]</code> entries (#181) (@github-actions[bot]) CompatHelper: add new compat entr...
0 We present Soss, a declarative probabilistic programming language embedded in the Julia languag... more 0 We present Soss, a declarative probabilistic programming language embedded in the Julia language. Soss represents statistical models in terms of abstract syntax trees, and uses staged compilation for on-demand generation of ``inference primitives'' (random sampling, log-density, etc) without requiring casual users to worry about such details. The approach taken by Soss makes it easy to extend to take advantage of other packages in the rapidly-growing Julia ecosystem. At the time of this writing, Soss users can choose from several inference back-ends and connect easily with larger systems SymPy and Gen.
Soss v0.18.2 Diff since v0.18.1 <strong>Closed issues:</strong> Model composition (#2... more Soss v0.18.2 Diff since v0.18.1 <strong>Closed issues:</strong> Model composition (#258) Update <code>testvalue</code> (#259)
AdvancedHMC v0.3.0 Diff since v0.2.27 <strong>Closed issues:</strong> Modular design ... more AdvancedHMC v0.3.0 Diff since v0.2.27 <strong>Closed issues:</strong> Modular design for refreshing momentum variables (#13) Unify <code>Trajectory</code> type for static, dynamic HMC samplers. (#103) First-class support of sampling on a struct (#159) Return adapted mass matrix (#230) HMC gives GPU compilation error (#235) Error gradient (#243) Support for Complex and Matrix Parameters (#251) StanHMC-adaptor (#252) Compatibility with ComponentArrays.jl (#253) Support for complex numbers (#262) the example in the doc does not run (#264) Minimal example from README.md gives error with Beta distribution (#266) <strong>Merged pull requests:</strong> Added note to documentation of sample function (#236) (@trappmartin) Update for MCMCDebugging (#238) (@xukai92) GitHub Actions workflow to pull changes to dev (#239) (@xukai92) Refactoring termination criterion (#240) (@xukai92) Update for MCMCDebugging (#241) (@xukai92) Fixed default type for NesterovDual...
We present a generic framework for par-allel coordinate descent (CD) algorithms that includes, as... more We present a generic framework for par-allel coordinate descent (CD) algorithms that includes, as special cases, the orig-inal sequential algorithms Cyclic CD and Stochastic CD, as well as the recent paral-lel Shotgun algorithm. We introduce two novel parallel algorithms that are also spe-cial cases—Thread-Greedy CD and Coloring-Based CD—and give performance measure-ments for an OpenMP implementation of these. 1.
Uploads
Papers by Chad Scherrer