Docs modifications to use QuadratureTraining instead of GridTraining #729

sdesai1287 · 2023-09-03T22:57:30Z

solved #634

docs/src/tutorials/low_level.md

ChrisRackauckas · 2023-09-03T23:04:00Z

docs/src/tutorials/param_estim.md

@@ -113,7 +113,7 @@ Then finally defining and optimizing using the `PhysicsInformedNN` interface.

 ```@example param_estim
 discretization = NeuralPDE.PhysicsInformedNN([chain1, chain2, chain3],
-                                             NeuralPDE.GridTraining(dt), param_estim = true,


dt is still defined?

It seems to be used in other places in this file, thats why I didnt remove it

yes, dt is used for evenly spaced time grid for the data

but it's not required when changing to QuadratureTraining.

yes, it can be removed

ok I think I have fixed this one

ChrisRackauckas · 2023-09-03T23:04:14Z

docs/src/tutorials/constraints.md

@@ -65,7 +65,7 @@ function norm_loss_function(phi, θ, p)
 end

 discretization = PhysicsInformedNN(chain,
-                                   GridTraining(dx),
+                                   QuadratureTraining(),


dx is still defined.

same as dt from the other file, it seems to be used in other places in this file, thats why I didnt remove it

But none of this file should be using it anymore?

Ok I removed the instances of dx but it seems like the program is still creating a grid of some kind (see line 61 and below), maybe only for evaluation. I am assuming I don't want to delete all that stuff?

sathvikbhagavan · 2023-09-06T03:55:49Z

Doc build is failing here and here

sdesai1287 · 2023-09-06T18:17:26Z

Doc build is failing here and here

Ok I fixed the issue for the second build fail, but I am not sure what the problem is for the first one

sathvikbhagavan · 2023-09-15T05:50:27Z

Docs is still erroring at https://github.com/SciML/NeuralPDE.jl/actions/runs/6185760782/job/16791965357?pr=729#step:5:86
The doc build is using [email protected]
Any idea why it is failing? Does it need [email protected]?

ChrisRackauckas · 2023-09-15T12:46:22Z

It at least wouldn't need a bump. What was the last version it was using wherei t was passing?

sdesai1287 · 2023-09-25T06:50:03Z

This is passing for me locally without bringing up the ComponentArrays issue. @ChrisRackauckas Can we rebuild the CI or something? The the changes I made should not affect ComponentArrays

ChrisRackauckas · 2023-09-25T10:23:38Z

I rebased onto master. Let's see how that goes.

ChrisRackauckas · 2023-09-26T02:54:26Z

Just please set the ComponentArrays bound to what it needs to be so we can merge.

sathvikbhagavan · 2023-09-26T05:00:20Z

I am not sure whats happening but I tried building the docs locally and it fails with the same error. But if I tried reducing the maxiters to 1200 from 2000 in https://github.com/SciML/NeuralPDE.jl/blob/master/docs/src/tutorials/constraints.md?plain=1#L91, it surprisingly passes.

… docs_update

sathvikbhagavan · 2023-09-28T07:45:44Z

The docs build is now failing because of IntegralsCubature 0.2.3 [also reproduced it locally] but it is passing with 0.2.2 locally. I haven't looked into the root cause.

The compat for ComponentArrays introduced in 87433bc isn't really affecting as it still uses 0.15.3 - https://github.com/SciML/NeuralPDE.jl/actions/runs/6323742800/job/17171915807?pr=729#step:4:96 but the other error we were seeing before is now passing.

sathvikbhagavan · 2023-09-30T03:57:13Z

@ChrisRackauckas any idea why the doc build is erroring with https://github.com/SciML/NeuralPDE.jl/actions/runs/6323742800/job/17171915807?pr=729#step:5:43

I traced it to https://github.com/SciML/Integrals.jl/blob/master/lib/IntegralsCubature/src/IntegralsCubature.jl#L171 locally but I am not what is the exact reason.

sathvikbhagavan · 2023-10-02T05:03:15Z

@sdesai1287 can you rebase your branch on master? Compat bounds were introduced for IntegralsCubature in #743 where docs passed. It should pass here as well and then this can be merged.

…ce for special likelihood term

Need to update docs for diff training strategies,etc

…ion 1, (keep existing compat)

… docs_update

ChrisRackauckas · 2023-10-06T14:26:48Z

The docs successfully built on master, so I put this on top of a successful master.

sathvikbhagavan

@ChrisRackauckas docs build is passing. I think we can merge this.

ChrisRackauckas · 2023-10-07T06:25:39Z

🎉

Docs modifications

b179412

ChrisRackauckas reviewed Sep 3, 2023

View reviewed changes

docs/src/tutorials/low_level.md Outdated Show resolved Hide resolved

ChrisRackauckas reviewed Sep 3, 2023

View reviewed changes

Samedh Desai and others added 2 commits September 3, 2023 23:15

Removed last instances of GridTraining in tutorials

70609cc

Removed dx and dt

e1996c6

docs build fail cleanup

b41c094

ChrisRackauckas closed this Sep 14, 2023

ChrisRackauckas reopened this Sep 14, 2023

Samedh Desai and others added 4 commits September 25, 2023 06:23

Docs modifications

98638a5

Removed last instances of GridTraining in tutorials

d2bd11a

Removed dx and dt

136d52f

docs build fail cleanup

aa87003

ChrisRackauckas force-pushed the docs_update branch from b41c094 to aa87003 Compare September 25, 2023 10:23

Samedh Desai added 2 commits September 27, 2023 02:00

tried adding a compat version to project.toml

87433bc

Merge branch 'docs_update' of github.com:sdesai1287/NeuralPDE.jl into…

60218bc

… docs_update

sdesai1287 and others added 3 commits October 1, 2023 22:15

Merge branch 'SciML:master' into docs_update

ce0bd7d

update Project.toml

7837d10

New PR

5e20ed1

AstitvaAggarwal and others added 20 commits October 1, 2023 22:36

less std for weights

f22481d

Cleaner Tests, all pass, handled edge cases

e735c84

minor changes

63de20d

Julia versions affect accuracy

9a94743

Added my suggested missing Loss function part, adjusted tests

d4786c7

minor changes

9c38687

added example, fixed multi dependant variable case, verfied performan…

ba58a2a

…ce for special likelihood term

fixed tests

70c7175

float 64 flux layers

345067a

now uses diff training strategies,

ab5700f

Need to update docs for diff training strategies,etc

tests

37d5e34

relaxed tests

da7f26f

relaxed tests

324ae46

added docs

5ca7d77

CompatHelper: add new compat entry for MonteCarloMeasurements at vers…

7831268

…ion 1, (keep existing compat)

CompatHelper: bump compat for SciMLBase to 2, (keep existing compat)

790ac82

Docs modifications

e6f588e

Removed last instances of GridTraining in tutorials

aa39260

update Project.toml

61a3c6e

Merge branch 'docs_update' of github.com:sdesai1287/NeuralPDE.jl into…

c9ab921

… docs_update

ChrisRackauckas closed this Oct 5, 2023

ChrisRackauckas reopened this Oct 5, 2023

Merge branch 'master' into docs_update

3560a97

sathvikbhagavan mentioned this pull request Oct 6, 2023

Investigating doc failure #751

Closed

Merge branch 'master' into docs_update

0528d75

sathvikbhagavan approved these changes Oct 7, 2023

View reviewed changes

ChrisRackauckas merged commit 4c79038 into SciML:master Oct 7, 2023
1 of 10 checks passed

sathvikbhagavan mentioned this pull request Oct 8, 2023

NNODE fails on simple logistic curve #634

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Docs modifications to use QuadratureTraining instead of GridTraining #729

Docs modifications to use QuadratureTraining instead of GridTraining #729

sdesai1287 commented Sep 3, 2023 •

edited

Loading

ChrisRackauckas Sep 3, 2023

sdesai1287 Sep 4, 2023

sathvikbhagavan Sep 5, 2023

ChrisRackauckas Sep 5, 2023

sathvikbhagavan Sep 5, 2023

sdesai1287 Sep 5, 2023

ChrisRackauckas Sep 3, 2023

sdesai1287 Sep 4, 2023

ChrisRackauckas Sep 5, 2023

sdesai1287 Sep 5, 2023 •

edited

Loading

sathvikbhagavan commented Sep 6, 2023

sdesai1287 commented Sep 6, 2023

sathvikbhagavan commented Sep 15, 2023

ChrisRackauckas commented Sep 15, 2023

sdesai1287 commented Sep 25, 2023 •

edited

Loading

ChrisRackauckas commented Sep 25, 2023

ChrisRackauckas commented Sep 26, 2023

sathvikbhagavan commented Sep 26, 2023

sathvikbhagavan commented Sep 28, 2023 •

edited

Loading

sathvikbhagavan commented Sep 30, 2023

sathvikbhagavan commented Oct 2, 2023

ChrisRackauckas commented Oct 6, 2023

sathvikbhagavan left a comment

ChrisRackauckas commented Oct 7, 2023

Docs modifications to use QuadratureTraining instead of GridTraining #729

Docs modifications to use QuadratureTraining instead of GridTraining #729

Conversation

sdesai1287 commented Sep 3, 2023 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

sdesai1287 Sep 5, 2023 • edited Loading

Choose a reason for hiding this comment

sathvikbhagavan commented Sep 6, 2023

sdesai1287 commented Sep 6, 2023

sathvikbhagavan commented Sep 15, 2023

ChrisRackauckas commented Sep 15, 2023

sdesai1287 commented Sep 25, 2023 • edited Loading

ChrisRackauckas commented Sep 25, 2023

ChrisRackauckas commented Sep 26, 2023

sathvikbhagavan commented Sep 26, 2023

sathvikbhagavan commented Sep 28, 2023 • edited Loading

sathvikbhagavan commented Sep 30, 2023

sathvikbhagavan commented Oct 2, 2023

ChrisRackauckas commented Oct 6, 2023

sathvikbhagavan left a comment

Choose a reason for hiding this comment

ChrisRackauckas commented Oct 7, 2023

sdesai1287 commented Sep 3, 2023 •

edited

Loading

sdesai1287 Sep 5, 2023 •

edited

Loading

sdesai1287 commented Sep 25, 2023 •

edited

Loading

sathvikbhagavan commented Sep 28, 2023 •

edited

Loading