Mixture is failing to add in Python tutorial 2 #25

GearsAD · 2022-03-23T12:49:06Z

Payload is failing to add, but the content looks to be correct. We should investigate unpacking and resolve.

This is the python packed factor payload:

{"label": "x1x2f1", "_version": "0.18.1", "_variableOrderSymbols": ["x1", "x2"], "data": {"eliminated": false, "potentialused": false, "edgeIDs": [], "fnc": {"N": 2, "F_": "PackedLinearRelative", "S": ["hypo1", "hypo2"], "components": [{"_type": "IncrementalInference.PackedRayleigh", "sigma": 3.0}, {"_type": "IncrementalInference.PackedUniform", "a": 30.0, "b": 55.0, "PackedSamplableTypeJSON": "IncrementalInference.PackedUniform"}], "diversity": {"_type": "IncrementalInference.PackedCategorical", "p": [0.4, 0.6]}}, "multihypo": [], "certainhypo": [1, 2], "nullhypo": 0.0, "solveInProgress": 0, "inflation": 5.0}, "tags": ["FACTOR"], "timestamp": "2022-03-23T12:21:24.554Z", "nstime": "0", "fnctype": "Mixture", "solvable": 1}

The text was updated successfully, but these errors were encountered:

jim-hill-r · 2022-03-23T13:17:25Z

Could you drop the couple of lines of Caesar code that produces this factor in a comment? I can write a unit test against the cloud and see what happens.

GearsAD · 2022-03-23T13:44:21Z

Yep I checked the payloads but can do.

GearsAD · 2022-03-23T13:46:55Z

using IncrementalInference
using DistributedFactorGraphs

fg = initfg()

# add the first node
addVariable!(fg, :x0, ContinuousScalar)

addFactor!(fg, [:x0], Prior(Normal(0,1)))
addVariable!(fg, :x1, ContinuousScalar)
# P(Z | :x1 - :x0 ) where Z ~ Normal(10,1)
addFactor!(fg, [:x0, :x1], LinearRelative(Normal(10.0,1)))

addVariable!(fg, :x2, ContinuousScalar)
mmo = Mixture(LinearRelative, 
              (hypo1=Rayleigh(3), hypo2=Uniform(30,55)), 
              [0.4; 0.6])
addFactor!(fg, [:x1, :x2], mmo)

# This factor - x1x2f1 is the one that is not loading.
f = getFactor(fg, :x1x2f1)

jim-hill-r · 2022-03-23T13:48:03Z

I'll take a look tonight.

GearsAD · 2022-03-23T14:14:48Z

Thanks! I see a decoding issue in the worker logs, but the payloads look the same.

# Julia
{"label":"x1x2f1","_version":"0.18.1","_variableOrderSymbols":["x1","x2"],"data":{"eliminated":false,"potentialused":false,"edgeIDs":[],"fnc":{"N":2,"F_":"PackedLinearRelative","S":["hypo1","hypo2"],"components":[{"_type":"IncrementalInference.PackedRayleigh","sigma":3.0},{"_type":"IncrementalInference.PackedUniform","a":30.0,"b":55.0,"PackedSamplableTypeJSON":"IncrementalInference.PackedUniform"}],"diversity":{"_type":"IncrementalInference.PackedCategorical","p":[0.4,0.6]}},"multihypo":[],"certainhypo":[1,2],"nullhypo":0.0,"solveInProgress":0,"inflation":5.0},"tags":["FACTOR"],"timestamp":"2022-03-23T07:09:55.020-05:00","nstime":"0","fnctype":"Mixture","solvable":1}
# Python
{"label":"x1x2f1","_version":"0.18.1","_variableOrderSymbols":["x1","x2"],"data":{"eliminated":false,"potentialused":false,"edgeIDs":[],"fnc":{"N":2,"F_":"PackedLinearRelative","S":["hypo1","hypo2"],"components":[{"_type":"IncrementalInference.PackedRayleigh","sigma":3.0},{"_type":"IncrementalInference.PackedUniform","a":30.0,"b":55.0,"PackedSamplableTypeJSON":"IncrementalInference.PackedUniform"}],"diversity":{"_type":"IncrementalInference.PackedCategorical","p":[0.4,0.6]}},"multihypo":[],"certainhypo":[1,2],"nullhypo":0.0,"solveInProgress":0,"inflation":5.0},"tags":["FACTOR"],"timestamp":"2022-03-23T12:21:24.554Z","nstime":"0","fnctype":"Mixture","solvable":1}

jim-hill-r · 2022-03-23T14:17:44Z

TIMESTAMPS!!!

GearsAD · 2022-03-23T15:05:35Z

Really?

jim-hill-r · 2022-03-23T15:37:19Z

'Z' is the pattern we have used in the SDK, but maybe the Mixtures code didn't get upgraded to Timezones.jl along with the rest. Not sure the best way to fix this. @Affie Advise?

dehann · 2022-03-23T20:52:34Z

but maybe the Mixtures code didn't get upgraded to Timezones.jl along with the rest.

The odds of that are low, but I cannot say with certainty 'no'. An error stack would be quick to resolve. By looking blind at the code, I'd be curious if the problem happens above, below, or inside this line:
https://github.com/JuliaRobotics/IncrementalInference.jl/blob/0f63b500191e35b3ec34f86bbe1185fce756ee80/src/Factors/Mixture.jl#L143

Nothing here suggests that the Timestamp would give an error.

Longer term though, it does speak to need for resolving:

Standardize `IIF.Mixture` JuliaRobotics/Caesar.jl#808

Tried Two things (FileDFG and Copy-Paste Bug into...)

So I also tried Sam's example above via FileDFG, and that seemed to work fine.
Then I modified the FileDFG tar.gz by copying in the Z timestamp string above (into factors/x1x2f1.json), but loads fine too. Here is the "test_bug" FileDFG:
- test_bug.tar.gz
- test.tar.gz

Also, I think the Z timestamp is generated by SDK.py. I think we should support both timestamp types in DFG/IIF?

xref Support both Timestamp formats "-00:00" and "Z"? JuliaRobotics/DistributedFactorGraphs.jl#868

json string with "-07:00" test.tar.gz

julia> fg
GraphsDFG{SolverParams, DFGVariable, DFGFactor}
  UserId: DefaultUser
  RobotId: DefaultRobot
  SessionId: Session_7c4118
  Description: Graphs.jl implementation
  Nr variables: 3
  Nr factors: 3
  User Data: Symbol[]
  Robot Data: Symbol[]
  Session Data: Symbol[]


julia> saveDFG("test.tar.gz", fg)

julia> fg_ = loadDFG("test.tar.gz")
sfolder = split(dstname, '.') = SubString{String}["test", "tar", "gz"]
[ Info: loadDFG! detected a gzip test.tar.gz -- unpacking via /tmp/caesar/random/e88cb92c now...
[ Info: Loaded 3 variables - [:x0, :x1, :x2]
[ Info: Inserting variables into graph...
┌ Warning: Unable to create measurements and gradients for LinearRelative{1, Normal{Float64}}(Normal{Float64}(μ=10.0, σ=1.0)) during prep of CCW, falling back on no-partial information assumption.  Enable ENV["JULIA_DEBUG"] = "IncrementalInference" for @debug printing to see the error.
└ @ IncrementalInference ~/.julia/dev/IncrementalInference/src/services/CalcFactor.jl:405
[ Info: Loaded 3 factors - [:x0f1, :x0x1f1, :x1x2f1]
[ Info: Inserting factors into graph...
[ Info: Rebuilding CCW's for the factors...
┌ Warning: Unable to create measurements and gradients for LinearRelative{1, Normal{Float64}}(Normal{Float64}(μ=10.0, σ=1.0)) during prep of CCW, falling back on no-partial information assumption.  Enable ENV["JULIA_DEBUG"] = "IncrementalInference" for @debug printing to see the error.
└ @ IncrementalInference ~/.julia/dev/IncrementalInference/src/services/CalcFactor.jl:405
[ Info: DFG.loadDFG! is deleting a temp folder created during unzip, /tmp/caesar/random/e88cb92c
GraphsDFG{SolverParams, DFGVariable, DFGFactor}
  UserId: DefaultUser
  RobotId: DefaultRobot
  SessionId: Session_763cdb
  Description: Graphs.jl implementation
  Nr variables: 3
  Nr factors: 3
  User Data: Symbol[]
  Robot Data: Symbol[]
  Session Data: Symbol[]

json string with `Z` test_bug.tar.gz

julia> fg_ = loadDFG("test_bug.tar.gz")
sfolder = split(dstname, '.') = SubString{String}["test_bug", "tar", "gz"]
[ Info: loadDFG! detected a gzip test_bug.tar.gz -- unpacking via /tmp/caesar/random/0057b7c2 now...
[ Info: Loaded 3 variables - [:x0, :x1, :x2]
[ Info: Inserting variables into graph...
┌ Warning: Unable to create measurements and gradients for LinearRelative{1, Normal{Float64}}(Normal{Float64}(μ=10.0, σ=1.0)) during prep of CCW, falling back on no-partial information assumption.  Enable ENV["JULIA_DEBUG"] = "IncrementalInference" for @debug printing to see the error.
└ @ IncrementalInference ~/.julia/dev/IncrementalInference/src/services/CalcFactor.jl:405
[ Info: Loaded 3 factors - [:x0f1, :x0x1f1, :x1x2f1]
[ Info: Inserting factors into graph...
[ Info: Rebuilding CCW's for the factors...
┌ Warning: Unable to create measurements and gradients for LinearRelative{1, Normal{Float64}}(Normal{Float64}(μ=10.0, σ=1.0)) during prep of CCW, falling back on no-partial information assumption.  Enable ENV["JULIA_DEBUG"] = "IncrementalInference" for @debug printing to see the error.
└ @ IncrementalInference ~/.julia/dev/IncrementalInference/src/services/CalcFactor.jl:405
[ Info: DFG.loadDFG! is deleting a temp folder created during unzip, /tmp/caesar/random/0057b7c2
GraphsDFG{SolverParams, DFGVariable, DFGFactor}
  UserId: DefaultUser
  RobotId: DefaultRobot
  SessionId: Session_1f664a
  Description: Graphs.jl implementation
  Nr variables: 3
  Nr factors: 3
  User Data: Symbol[]
  Robot Data: Symbol[]
  Session Data: Symbol[]

jim-hill-r · 2022-03-23T21:07:11Z

Let's write the test and validate the "red herring" assumption that timestamps are the issue.

GearsAD · 2022-03-24T03:38:11Z

No I believe this is a serialization issue unrelated to timestamps. More to follow in a few moments.

GearsAD · 2022-03-24T04:33:18Z

It looks to be related to the way the Dict{string, Any} are converted to NamedTuples that are used to create the distributions and the PackedMixture model. We can discuss in morning.

dehann · 2022-03-24T17:53:32Z

Right, there were two bugs.

DFG is using JSON2 to unpack the json strings.

That is why FileDFG was able to unpack the bug string above. Sam came up against differences between JSON and JSON2 when trying this yesterday. DFG is using JSON2.

The right way to unpack factors currently is:
https://github.com/JuliaRobotics/DistributedFactorGraphs.jl/blob/24ef7c067ff2de4a069070a43dab939f8bbe5d30/src/FileDFG/services/FileDFG.jl#L145-L146

Cloud was not yet on IIF v0.27.3

Aside, I have a good idea of how to support both JSON and JSON2. Thats happening here:

Serialization and Marshaling (again) JuliaRobotics/DistributedFactorGraphs.jl#590 (comment)

dehann · 2022-03-24T19:32:31Z

Think we can close this now.

GearsAD transferred this issue from NavAbility/NavAbilitySDK.py Mar 23, 2022

GearsAD mentioned this issue Mar 23, 2022

Fixing up tutorial 2 in Python #26

Merged

jim-hill-r self-assigned this Mar 23, 2022

dehann mentioned this issue Mar 23, 2022

Support both Timestamp formats "-00:00" and "Z"? JuliaRobotics/DistributedFactorGraphs.jl#868

Open

dehann self-assigned this Mar 24, 2022

dehann closed this as completed Mar 24, 2022

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Mixture is failing to add in Python tutorial 2 #25

Mixture is failing to add in Python tutorial 2 #25

GearsAD commented Mar 23, 2022 •

edited

Loading

jim-hill-r commented Mar 23, 2022

GearsAD commented Mar 23, 2022

GearsAD commented Mar 23, 2022 •

edited by dehann

Loading

jim-hill-r commented Mar 23, 2022

GearsAD commented Mar 23, 2022

jim-hill-r commented Mar 23, 2022

GearsAD commented Mar 23, 2022

jim-hill-r commented Mar 23, 2022

dehann commented Mar 23, 2022 •

edited

Loading

jim-hill-r commented Mar 23, 2022

GearsAD commented Mar 24, 2022 •

edited

Loading

GearsAD commented Mar 24, 2022

dehann commented Mar 24, 2022 •

edited

Loading

dehann commented Mar 24, 2022

Mixture is failing to add in Python tutorial 2 #25

Mixture is failing to add in Python tutorial 2 #25

Comments

GearsAD commented Mar 23, 2022 • edited Loading

jim-hill-r commented Mar 23, 2022

GearsAD commented Mar 23, 2022

GearsAD commented Mar 23, 2022 • edited by dehann Loading

jim-hill-r commented Mar 23, 2022

GearsAD commented Mar 23, 2022

jim-hill-r commented Mar 23, 2022

GearsAD commented Mar 23, 2022

jim-hill-r commented Mar 23, 2022

dehann commented Mar 23, 2022 • edited Loading

Tried Two things (FileDFG and Copy-Paste Bug into...)

json string with "-07:00" test.tar.gz

json string with Z test_bug.tar.gz

jim-hill-r commented Mar 23, 2022

GearsAD commented Mar 24, 2022 • edited Loading

GearsAD commented Mar 24, 2022

dehann commented Mar 24, 2022 • edited Loading

DFG is using JSON2 to unpack the json strings.

Cloud was not yet on IIF v0.27.3

dehann commented Mar 24, 2022

GearsAD commented Mar 23, 2022 •

edited

Loading

GearsAD commented Mar 23, 2022 •

edited by dehann

Loading

dehann commented Mar 23, 2022 •

edited

Loading

json string with `Z` test_bug.tar.gz

GearsAD commented Mar 24, 2022 •

edited

Loading

dehann commented Mar 24, 2022 •

edited

Loading