# Additional Function Reference

## RoME

`RoME.getRangeKDEMax2D`

— Function```
getRangeKDEMax2D(fgl, vsym1, vsym2)
```

Calculate the cartesian distance between two vertices in the graph using their symbol name, and by maximum belief point.

`RoME.initFactorGraph!`

— Function```
initFactorGraph!(
fg;
P0,
init,
N,
lbl,
solvable,
firstPoseType,
labels
)
```

Initialize a factor graph object as Pose2, Pose3, or neither and returns variable and factor symbols as array.

`RoME.addOdoFG!`

— Function```
addOdoFG!(fg, n, DX, cov; N, solvable, labels)
```

Create a new variable node and insert odometry constraint factor between which will automatically increment latest pose symbol x<k+1> for new node new node and constraint factor are returned as a tuple.

```
addOdoFG!(fgl, odo; N, solvable, labels)
```

Create a new variable node and insert odometry constraint factor between which will automatically increment latest pose symbol x<k+1> for new node new node and constraint factor are returned as a tuple.

## IncrementalInference

`IncrementalInference.approxCliqMarginalUp!`

— Function```
approxCliqMarginalUp!(csmc; ...)
approxCliqMarginalUp!(
csmc,
childmsgs;
N,
dbg,
multiproc,
logger,
iters,
drawpdf
)
```

Approximate Chapman-Kolmogorov transit integral and return separator marginals as messages to pass up the Bayes (Junction) tree, along with additional clique operation values for debugging.

Notes

`onduplicate=true`

by default internally uses deepcopy of factor graph and Bayes tree, and does**not**update the given objects. Set false to update`fgl`

and`treel`

during compute.

Future

- TODO: internal function chain is too long and needs to be refactored for maintainability.

`IncrementalInference.areCliqVariablesAllMarginalized`

— Function```
areCliqVariablesAllMarginalized(subfg, cliq)
```

Return true if all variables in clique are considered marginalized (and initialized).

`IncrementalInference.attemptTreeSimilarClique`

— Function```
attemptTreeSimilarClique(othertree, seeksSimilar)
```

Special internal function to try return the clique data if succesfully identified in `othertree::AbstractBayesTree`

, based on contents of `seeksSimilar::BayesTreeNodeData`

.

Notes

- Used to identify and skip similar cliques (i.e. recycle computations)

`IncrementalInference.childCliqs`

— Function```
childCliqs(treel, cliq)
```

Return a vector of child cliques to `cliq`

.

`IncrementalInference.cliqHistFilterTransitions`

— Function```
cliqHistFilterTransitions(hist, nextfnc)
```

Return state machine transition steps from history such that the `nextfnc::Function`

.

Related:

printCliqHistorySummary, filterHistAllToArray, sandboxCliqResolveStep

`IncrementalInference.cycleInitByVarOrder!`

— Function```
cycleInitByVarOrder!(subfg, varorder; solveKey, logger)
```

Cycle through var order and initialize variables as possible in `subfg::AbstractDFG`

. Return true if something was updated.

Notes:

- assumed
`subfg`

is a subgraph containing only the factors that can be used.- including the required up or down messages

- intended for both up and down initialization operations.

Dev Notes

- Should monitor updates based on the number of inferred & solvable dimensions

`IncrementalInference.doautoinit!`

— Function```
doautoinit!(dfg, xi; solveKey, singles, N, logger)
```

EXPERIMENTAL: initialize target variable `xi`

based on connected factors in the factor graph `fgl`

. Possibly called from `addFactor!`

, or `doCliqAutoInitUp!`

(?).

Notes:

- Special carve out for multihypo cases, see issue 427.

Development Notes:

- Target factor is first (singletons) or second (dim 2 pairwise) variable vertex in
`xi`

. - TODO use DFG properly with local operations and DB update at end.
- TODO get faster version of
`isInitialized`

for database version. - TODO: Persist this back if we want to here.
- TODO: init from just partials

`IncrementalInference.drawCliqSubgraphUpMocking`

— Function```
drawCliqSubgraphUpMocking(
fgl,
treel,
frontalSym;
show,
filepath,
engine,
viewerapp
)
```

Construct (new) subgraph and draw the subgraph associated with clique `frontalSym::Symbol`

.

Notes

- See
`drawGraphCliq`

/`writeGraphPdf`

for details on keyword options.

Related

drawGraphCliq, spyCliqMat, drawTree, buildCliqSubgraphUp, buildSubgraphFromLabels!

`IncrementalInference.fifoFreeze!`

— Function```
fifoFreeze!(dfg)
```

Freeze nodes that are older than the quasi fixed-lag length defined by `fg.qfl`

, according to `fg.fifo`

ordering.

Future:

- Allow different freezing strategies beyond fifo.

`IncrementalInference.filterHistAllToArray`

— Function```
filterHistAllToArray(tree, hists, frontals, nextfnc)
```

Return state machine transition steps from all cliq histories with transition `nextfnc::Function`

.

Related:

printCliqHistorySummary, cliqHistFilterTransitions, sandboxCliqResolveStep

`IncrementalInference.fmcmc!`

— Function```
fmcmc!(fgl, cliq, fmsgs, lbls, solveKey, N, MCMCIter)
fmcmc!(fgl, cliq, fmsgs, lbls, solveKey, N, MCMCIter, dbg)
fmcmc!(
fgl,
cliq,
fmsgs,
lbls,
solveKey,
N,
MCMCIter,
dbg,
logger
)
fmcmc!(
fgl,
cliq,
fmsgs,
lbls,
solveKey,
N,
MCMCIter,
dbg,
logger,
multithreaded
)
```

Iterate successive approximations of clique marginal beliefs by means of the stipulated proposal convolutions and products of the functional objects for tree clique `cliq`

.

`IncrementalInference.getClique`

— Function```
getClique(tree, cId)
```

Return the TreeClique node object that represents a clique in the Bayes (Junction) tree, as defined by one of the frontal variables `frt<:AbstractString`

.

Notes

- Frontal variables only occur once in a clique per tree, therefore is a unique identifier.

Related:

getCliq, getTreeAllFrontalSyms

`IncrementalInference.getCliqAllVarIds`

— Function```
getCliqAllVarIds(cliq)
```

Get all `cliq`

variable ids`::Symbol`

.

Related

getCliqVarIdsAll, getCliqFactorIdsAll, getCliqVarsWithFrontalNeighbors

`IncrementalInference.getCliqAssocMat`

— Function```
getCliqAssocMat(cliq)
```

Return boolean matrix of factor by variable (row by column) associations within clique, corresponds to order presented by `getCliqFactorIds`

and `getCliqAllVarIds`

.

`IncrementalInference.getCliqDepth`

— Function```
getCliqDepth(tree, cliq)
```

Return depth in tree as `::Int`

, with root as depth=0.

Related

getCliq

`IncrementalInference.getCliqDownMsgsAfterDownSolve`

— Function```
getCliqDownMsgsAfterDownSolve(
subdfg,
cliq,
solveKey;
status,
sender
)
```

Return dictionary of down messages consisting of all frontal and separator beliefs of this clique.

Notes:

- Fetches numerical results from
`subdfg`

as dictated in`cliq`

. - return LikelihoodMessage

`IncrementalInference.getCliqFrontalVarIds`

— Function```
getCliqFrontalVarIds(cliqdata)
```

Get the frontal variable IDs `::Int`

for a given clique in a Bayes (Junction) tree.

`IncrementalInference.getCliqVarInitOrderUp`

— Function```
getCliqVarInitOrderUp(subfg)
```

Return the most likely ordering for initializing factor (assuming up solve sequence).

Notes:

- sorts id (label) for increasing number of connected factors using the clique subfg with messages already included.

`IncrementalInference.getCliqMat`

— Function```
getCliqMat(cliq; showmsg)
```

Return boolean matrix of factor variable associations for a clique, optionally including (`showmsg::Bool=true`

) the upward message singletons. Variable order corresponds to `getCliqAllVarIds`

.

`IncrementalInference.getCliqSeparatorVarIds`

— Function```
getCliqSeparatorVarIds(cliqdata)
```

Get `cliq`

separator (a.k.a. conditional) variable ids`::Symbol`

.

`IncrementalInference.getCliqSiblings`

— Function```
getCliqSiblings(treel, cliq)
getCliqSiblings(treel, cliq, inclusive)
```

Return a vector of all siblings to a clique, which defaults to not `inclusive`

the calling `cliq`

.

`IncrementalInference.getCliqVarIdsPriors`

— Function```
getCliqVarIdsPriors(cliq)
getCliqVarIdsPriors(cliq, allids)
getCliqVarIdsPriors(cliq, allids, partials)
```

Get variable ids`::Int`

with prior factors associated with this `cliq`

.

Notes:

- does not include any singleton messages from upward or downward message passing.

`IncrementalInference.getCliqVarSingletons`

— Function```
getCliqVarSingletons(cliq)
getCliqVarSingletons(cliq, allids)
getCliqVarSingletons(cliq, allids, partials)
```

Get `cliq`

variable IDs with singleton factors – i.e. both in clique priors and up messages.

`IncrementalInference.getParent`

— Function```
getParent(treel, afrontal)
```

Return `cliq`

's parent clique.

`IncrementalInference.getTreeAllFrontalSyms`

— Function```
getTreeAllFrontalSyms(_, tree)
```

Return one symbol (a frontal variable) from each clique in the `::BayesTree`

.

Notes

- Frontal variables only occur once in a clique per tree, therefore is a unique identifier.

Related:

whichCliq, printCliqHistorySummary

`IncrementalInference.hasClique`

— Function```
hasClique(bt, frt)
```

Return boolean on whether the frontal variable `frt::Symbol`

exists somewhere in the `::BayesTree`

.

`DistributedFactorGraphs.isInitialized`

— Function```
isInitialized(var)
isInitialized(var, key)
```

Returns state of variable data `.initialized`

flag.

Notes:

- used by both factor graph variable and Bayes tree clique logic.

```
isInitialized(cliq)
```

Returns state of Bayes tree clique `.initialized`

flag.

Notes:

- used by Bayes tree clique logic.
- similar method in DFG

`DistributedFactorGraphs.isMarginalized`

— Function```
isMarginalized(vert)
isMarginalized(vert, solveKey)
```

Return `::Bool`

on whether this variable has been marginalized.

Notes:

- VariableNodeData default
`solveKey=:default`

`IncrementalInference.isTreeSolved`

— Function```
isTreeSolved(treel; skipinitialized)
```

Return true or false depending on whether the tree has been fully initialized/solved/marginalized.

`ApproxManifoldProducts.isPartial`

— Function```
isPartial(fcf)
```

Return `::Bool`

on whether factor is a partial constraint.

`IncrementalInference.localProduct`

— Function```
localProduct(dfg, sym; solveKey, N, dbg, logger)
```

Using factor graph object `dfg`

, project belief through connected factors (convolution with likelihood) to variable `sym`

followed by a approximate functional product.

Return: product belief, full proposals, partial dimension proposals, labels

`IncrementalInference.makeCsmMovie`

— Function```
makeCsmMovie(fg, tree; ...)
makeCsmMovie(
fg,
tree,
cliqs;
assignhist,
show,
filename,
frames
)
```

Convenience function to assign and make video of CSM state machine for `cliqs`

.

Notes

- Probably several teething issues still (lower priority).
- Use
`assignhist`

if solver params async was true, or errored.

Related

csmAnimate, printCliqHistorySummary

`IncrementalInference.parentCliq`

— Function```
parentCliq(treel, cliq)
```

Return `cliq`

's parent clique.

`RoME.predictVariableByFactor`

— Function```
predictVariableByFactor(dfg, targetsym, fct, prevars)
```

Method to compare current and predicted estimate on a variable, developed for testing a new factor before adding to the factor graph.

Notes

`fct`

does not have to be in the factor graph – likely used to test beforehand.- function is useful for detecting if
`multihypo`

should be used. `approxConv`

will project the full belief estimate through some factor but must already be in factor graph.

Example

```
# fg already exists containing :x7 and :l3
pp = Pose2Point2BearingRange(Normal(0,0.1),Normal(10,1.0))
# possible new measurement from :x7 to :l3
curr, pred = predictVariableByFactor(fg, :l3, pp, [:x7; :l3])
# example of naive user defined test on fit score
fitscore = minkld(curr, pred)
# `multihypo` can be used as option between existing or new variables
```

Related

approxConv

`IncrementalInference.printCliqHistorySummary`

— Function```
printCliqHistorySummary(fid, hist)
printCliqHistorySummary(fid, hist, cliqid)
```

Print a short summary of state machine history for a clique solve.

Related:

getTreeAllFrontalSyms, animateCliqStateMachines, printHistoryLine, printCliqHistorySequential

`IncrementalInference.resetCliqSolve!`

— Function```
resetCliqSolve!(dfg, treel, cliq; solveKey)
```

Reset the state of all variables in a clique to not initialized.

Notes

- resets numberical values to zeros.

Dev Notes

- TODO not all kde manifolds will initialize to zero.
- FIXME channels need to be consolidated

`IncrementalInference.resetData!`

— Function```
resetData!(vdata)
```

Partial reset of basic data fields in `::VariableNodeData`

of `::FunctionNode`

structures.

`IncrementalInference.resetTreeCliquesForUpSolve!`

— Function```
resetTreeCliquesForUpSolve!(treel)
```

Reset the Bayes (Junction) tree so that a new upsolve can be performed.

Notes

- Will change previous clique status from
`DOWNSOLVED`

to`INITIALIZED`

only. - Sets the color of tree clique to
`lightgreen`

.

`IncrementalInference.resetVariable!`

— Function```
resetVariable!(varid; solveKey)
```

Reset the solve state of a variable to uninitialized/unsolved state.

`IncrementalInference.setfreeze!`

— Function```
setfreeze!(dfg, sym)
```

Set variable(s) `sym`

of factor graph to be marginalized – i.e. not be updated by inference computation.

`IncrementalInference.setValKDE!`

— Function```
setValKDE!(vd, pts, bws)
setValKDE!(vd, pts, bws, setinit)
setValKDE!(vd, pts, bws, setinit, ipc)
```

Set the point centers and bandwidth parameters of a variable node, also set `isInitialized=true`

if `setinit::Bool=true`

(as per default).

Notes

`initialized`

is used for initial solve of factor graph where variables are not yet initialized.`inferdim`

is used to identify if the initialized was only partial.

`IncrementalInference.setVariableInitialized!`

— Function```
setVariableInitialized!(varid, status)
```

Set variable initialized status.

`IncrementalInference.solveCliqWithStateMachine!`

— Function```
solveCliqWithStateMachine!(
dfg,
tree,
frontal;
iters,
downsolve,
recordhistory,
verbose,
nextfnc,
prevcsmc
)
```

Standalone state machine solution for a single clique.

Related:

initInferTreeUp!

`IncrementalInference.transferUpdateSubGraph!`

— Function```
transferUpdateSubGraph!(dest, src; ...)
transferUpdateSubGraph!(dest, src, syms; ...)
transferUpdateSubGraph!(
dest,
src,
syms,
logger;
updatePPE,
solveKey
)
```

Transfer contents of `src::AbstractDFG`

variables `syms::Vector{Symbol}`

to `dest::AbstractDFG`

. Notes

- Reads,
`dest`

:=`src`

, for all`syms`

`IncrementalInference.treeProductDwn`

— Function```
treeProductDwn(fg, tree, cliq, sym; N, dbg)
```

Calculate a fresh–-single step–-approximation to the variable `sym`

in clique `cliq`

as though during the downward message passing. The full inference algorithm may repeatedly calculate successive apprimxations to the variable based on the structure of variables, factors, and incoming messages to this clique. Which clique to be used is defined by frontal variable symbols (`cliq`

in this case) – see `getClique(...)`

for more details. The `sym`

symbol indicates which symbol of this clique to be calculated. **Note** that the `sym`

variable must appear in the clique where `cliq`

is a frontal variable.

`IncrementalInference.treeProductUp`

— Function```
treeProductUp(fg, tree, cliq, sym; N, dbg)
```

Calculate a fresh (single step) approximation to the variable `sym`

in clique `cliq`

as though during the upward message passing. The full inference algorithm may repeatedly calculate successive apprimxations to the variables based on the structure of the clique, factors, and incoming messages. Which clique to be used is defined by frontal variable symbols (`cliq`

in this case) – see `getClique(...)`

for more details. The `sym`

symbol indicates which symbol of this clique to be calculated. **Note** that the `sym`

variable must appear in the clique where `cliq`

is a frontal variable.

`IncrementalInference.unfreezeVariablesAll!`

— Function```
unfreezeVariablesAll!(fgl)
```

Free all variables from marginalization.

Related

dontMarginalizeVariablesAll!

`IncrementalInference.dontMarginalizeVariablesAll!`

— Function```
dontMarginalizeVariablesAll!(fgl)
```

Free all variables from marginalization.

`IncrementalInference.updateFGBT!`

— Function```
updateFGBT!(fg, cliq, IDvals; dbg, fillcolor, logger)
```

Update cliq `cliqID`

in Bayes (Juction) tree `bt`

according to contents of `urt`

. Intended use is to update main clique after a upward belief propagation computation has been completed per clique.

`IncrementalInference.upGibbsCliqueDensity`

— Function```
upGibbsCliqueDensity(dfg, cliq, solveKey, inmsgs)
upGibbsCliqueDensity(dfg, cliq, solveKey, inmsgs, N)
upGibbsCliqueDensity(dfg, cliq, solveKey, inmsgs, N, dbg)
upGibbsCliqueDensity(
dfg,
cliq,
solveKey,
inmsgs,
N,
dbg,
iters
)
upGibbsCliqueDensity(
dfg,
cliq,
solveKey,
inmsgs,
N,
dbg,
iters,
logger
)
```

Perform computations required for the upward message passing during belief propation on the Bayes (Junction) tree. This function is usually called as via remote_call for multiprocess dispatch.

Notes

`fg`

factor graph,`tree`

Bayes tree,`cliq`

which cliq to perform the computation on,`parent`

the parent clique to where the upward message will be sent,`childmsgs`

is for any incoming messages from child cliques.

DevNotes

- FIXME total rewrite with AMP #41 and RoME #244 in mind

`IncrementalInference.resetVariableAllInitializations!`

— Function```
resetVariableAllInitializations!(fgl)
```

Reset initialization flag on all variables in `::AbstractDFG`

.

Notes

- Numerical values remain, but inference will overwrite since init flags are now
`false`

.