The blangsdk from ubc-stat-ml

Improve slice sampler

Use doubling instead of linear expansion procedure, re-read Neal's paper and re-write the implementation.

Fix issues encountered while testing MixtureModel

Bug: posteriors clearly incorrect
Performance: seems very slow at 1000 datapoint, encouragingly, get much more than 2x faster at 500 datapoints, so some factor dependencies must be wrong
Marking fields with transient seems to cause undesirable side effects, need better way to signal graph inference to skip some [document adding transient to K value changes behavior]
Style: problematic that loops cannot be shared [split to separate issue]
Style: autodeboxing is needed [see separate issue]

More advanced tempering features

See md document in /notes/

Table's should not be @Immutable

Add back slice sampler and/or non-reversible stuff

Check cases where a complex data structure is partially observed

E.g. are cases like Simplex correctly handled?

Fix registration of names in GraphAnalysis

The annotation @DeboxedName should be used instead of the raw field name. Wait until variable printing architecture is setup.

Need to add zero and one-string constructors to built-in datatypes

Basic HMM model and associated sampler

Demonstrate at same time reuse of models.

Add checks that all variables are covered by samplers

Command-line blang has a 5--10 sec latency

@justinHume : I'll look into that

Multivariate normal distribution, associated sampler

Add checks that static time constraints are constant

Exported maven artefacts too big

Currently almost 100MB per version. Will break artefact git repo soon if not fixed

Reporting acceptance rates, etc

Simplify Factor class hierarchy

Get rid of LogScaleFactor, move its method to Factor. Generalize when (if) the need arises.

Move runtime from bayonet to blangSDK

This is required since these depend on interfaces such as Model that are needed in blangDSL (and it's non-trivial because of plug-in architecture and seems suboptimal to have blangDSL import bayonet).

Note: important to also erase from bayonet to avoid having two versions of same thing.

Light weight data frame structure needed for plated random variables

Working on Plated<T>. More details soon.

!! Create compute-aware annealing methods

E.g. for trees or exchangeable datasets, use only subset of sites in the first iterations

Some basic performance benchmarks

Standalone runner should handle file names with non standard characters gracefully

E.g., dashes

Implement parallel tempering

Simplex type, Dirichlet distribution and associated sampler

Demonstrate at same time that the type engineering architecture works

Add support for distributed SMC/PT

Need annealing to take care of initialization; and parallel tempering at same time for perf?

Issues:

how to do the factor transformation
transform them into what (what exponents?)
how to mask bad energies from processors?

Move syntax tests from DSL to SDK

Already deleted from DSL as they were not catching basic syntax errors.

Implement "primitive" types

Real
Int
Bool

Associated:

basic sampler
command line parser

Basic set of discrete distributions, associated samplers

Integer valued:

Multinomial
Poisson

Consider a new type of support factor with an explicit finite list. Allows better samplers.

Sampler exclusions

Reorganize examples

Remove blang.examples from SDK
Create project template
Create some script to upgrade everything or via snapshots
Write examples via an instantiation of the template

Add static time conditionals?

Integrate xlinear

Most of the work will be with integration of graph introspection and views.

Use xtend's range operators

E.g., in src/main/java/blang/examples/MixtureModel.bl, instead of StaticUtils::range(observations.size()), we should be able to use 0 ..< observations.size(). (Xtend reference)

Add support for enums

Do not do via java enums! Because the values cannot be read from command line then. Would also make a mess everywhere. Instead, use some kind of FiniteSupport

Improve performance of cloning

Some immutable stuff to mark for shallow copy, e.g. arrays of integers for sampler cache
Datasets
Plates, etc [mark as immutable as well]
Parallelize!

Basic HMM model and associated sampler - DUPLICATED

Demonstrate at same time reuse of models.

Further improvements on data frames

Inverted control with plates
Make Table interface, concrete types are ScalarTable and MatrixTable
- ScalarTable configured by matrix.inPlates(..) only
- MatrixTable configured by matrix.inPlates(..); matrix.rowsIndexedBy(...); matrix.colsIndexedBy(...)
- In both cases, plate indices are modelled by sets (no duplicates)
Keep the same idea in ScalarTable
Sub-frames

Improve data frame

Most important: make it so that loops can be combined,
Conversion into lists/matrices

Normal
Gamma
Beta
...

Upgrade inits and improve command line option reporting

Fix printing

Make it work on SMC samplers
Use hierarchical dir? Field names? ExperimentResults?

Add model integrity checks

DAG property [implemented, still need to hook up?]
all declared variables used [done? check]
..

Adaptation

Need to at least set univariate scale.

Improve publishing architecture - current artifact/git hack not scalable

Problems with current solution:

Limitations of github pages: https://help.github.com/articles/what-is-github-pages/
- max of 1GB
- max of 10 builds per day
Limitations of git: push gets slower