Energy product in elliptic discretizer (Continuation) #1008

TiKeil · 2020-07-09T19:07:02Z

In this PR I continue #875 . I added the requested changes and changed a demo very briefly. Let me know if there is additional work to do for this.

TiKeil · 2020-07-09T19:36:02Z

@renefritze, As it shows in https://dev.azure.com/pymor/pymor/_build/results?buildId=2757&view=logs&j=b2cbc7ea-7585-56c2-8847-7376d77354b6&t=1f352729-2dd2-595d-2a18-f5a7564f328f, it seems that af2c20c was not enough for changing the args in the elliptic2 demo. Could you help out where else I need to edit it?

renefritze · 2020-07-10T06:55:06Z

The problem isn't in the commit you linked, I think. But in the changing of the docopt string for the demo. To get more insight you can locally execute just that demo with -h and you'll probably see a similar error. My guess would be that the colons confuse docopt's parsing.

sdrave · 2020-07-16T09:18:53Z

@TiKeil, can you rebase on master?

TiKeil · 2020-07-16T14:52:51Z

Does anybody understand whats wrong with the tests again?

renefritze · 2020-07-16T15:24:45Z

Did you see https://zivgitlab.uni-muenster.de/pymor/pymor/-/jobs/373734#L1079?

TiKeil · 2020-07-17T09:49:28Z

Thanks a lot. Could have found this by myself : /

src/pymordemos/elliptic2.py

TiKeil · 2020-07-17T11:14:43Z

To get more insight you can locally execute just that demo with -h and you'll probably see a similar error.

This is not showing an error.

My guess would be that the colons confuse docopt's parsing.

Do you see any difference to the docstring in https://github.com/pymor/pymor/blob/master/src/pymordemos/elliptic.py ?

renefritze · 2020-07-17T11:35:28Z

To get more insight you can locally execute just that demo with -h and you'll probably see a similar error.

This is not showing an error.

My guess would be that the colons confuse docopt's parsing.

Do you see any difference to the docstring in https://github.com/pymor/pymor/blob/master/src/pymordemos/elliptic.py ?

Yes: e699c01 😉

If this works, the commit should be squashed before the merge.

codecov · 2020-07-17T12:34:34Z

Codecov Report

Merging #1008 into master will decrease coverage by 0.03%.
The diff coverage is 65.51%.

Impacted Files	Coverage Δ
src/pymor/discretizers/builtin/cg.py	`89.93% <65.51%> (-1.32%)`	⬇️
src/pymor/bindings/ngsolve.py	`84.55% <0.00%> (-0.74%)`	⬇️
src/pymor/vectorarrays/list.py	`83.77% <0.00%> (-0.57%)`	⬇️
src/pymor/vectorarrays/numpy.py	`84.55% <0.00%> (+0.24%)`	⬆️

TiKeil · 2020-07-17T12:35:18Z

Thanks @renefritze for fixing.
I think this should be ready now.

AUTHORS.md

src/pymor/discretizers/builtin/cg.py

src/pymordemos/elliptic2.py

TiKeil · 2020-08-03T17:54:11Z

tried to address your points as much as possible.

src/pymor/discretizers/builtin/cg.py

src/pymordemos/elliptic2.py

TiKeil · 2020-08-18T12:29:50Z

Thanks @sdrave, I think I addressed all your points.

Co-authored-by: Stephan Rave <stephanrave@uni-muenster.de>

sdrave · 2020-08-25T15:11:49Z

src/pymordemos/elliptic2.py

+                    k: compute the energy norm of the last snapshot, where the energy-product is constructed
+                    with a randomly generated parameter instance with random seed integer k.


Sorry, but why not just let the user specify the actual value? Specifying the seed seems weird, when it's only a single number we are talking about.

You are absolutely right. I thought that problem 0 and 1 differ in terms of parameter space. I changed that now.

src/pymordemos/elliptic2.py

src/pymor/discretizers/builtin/cg.py

Co-authored-by: Stephan Rave <stephanrave@uni-muenster.de>

renefritze added the pr:new-feature Introduces a new feature label Jul 10, 2020

sdrave added this to the 2020.2 milestone Jul 16, 2020

sdrave mentioned this pull request Jul 16, 2020

Energy product in elliptic discretizer #875

Closed

TiKeil force-pushed the energy-product-in-elliptic-discretizer branch from 99f6e3f to a1e4d62 Compare July 16, 2020 10:27

TiKeil force-pushed the energy-product-in-elliptic-discretizer branch from a1e4d62 to c3ed6bc Compare July 17, 2020 10:09

HenKlei reviewed Jul 17, 2020

View reviewed changes

src/pymordemos/elliptic2.py Outdated Show resolved Hide resolved

TiKeil force-pushed the energy-product-in-elliptic-discretizer branch from e699c01 to fa4b578 Compare July 17, 2020 12:33

sdrave requested changes Jul 24, 2020

View reviewed changes

TiKeil force-pushed the energy-product-in-elliptic-discretizer branch 2 times, most recently from 3a691bd to 80285ba Compare August 3, 2020 17:47

TiKeil force-pushed the energy-product-in-elliptic-discretizer branch from 80285ba to 55d2192 Compare August 4, 2020 09:14

TiKeil force-pushed the energy-product-in-elliptic-discretizer branch from 55d2192 to bc32ab6 Compare August 12, 2020 13:53

sdrave requested changes Aug 17, 2020

View reviewed changes

TiKeil force-pushed the energy-product-in-elliptic-discretizer branch from f44847d to cf7ef1e Compare August 18, 2020 12:28

ftschindler and others added 4 commits August 24, 2020 17:24

[discretizers.cg] allow to assemble energy product for a fixed parameter

539fe69

[discretizers.builtin.cg] fix energy product assembly

a3bc754

[discretizers.builtin.cg] some fixes in energy product assembly

59c6a91

[discretizers.cg] account for pending requestes changes in pymor#875

5b6eef4

Tim Keil and others added 17 commits August 24, 2020 17:24

[discretizers.cg] small change for new parameter handling

b1539ca

[pymordemos] change elliptic2 to also use an energy product

19f3c45

[pymordemos] change the args in elliptic2

6dcbb08

[pymordemos] fix wrong call of --fv demo

e1c5ad4

[pymordemos] try to fix wrong indention in docstring

d016084

add AUTHORS entry

a28a453

[discretizers.cg] small change in description of arguments

ff85d13

[discretizers.cg] energy_product -> mu_energy_product

a015b67

[discretizers.cg] put correct robin data to the energy product

4156f70

[discretizers.cg] first add energy_product, then build the model

590cb5b

[discretizers.cg] assemble lincomb operator

3da1fe0

[pymordemos] update the elliptic2 demo for better explanations

5b0c437

Update src/pymordemos/elliptic2.py

b0fbd75

Co-authored-by: Stephan Rave <stephanrave@uni-muenster.de>

Update src/pymordemos/elliptic2.py

65de646

Co-authored-by: Stephan Rave <stephanrave@uni-muenster.de>

[discretizers.cg] try to be more precise for 'mu_energy_product'

399b965

[discretizers.cg] this will never happen

9a35718

change arguments of the demos

c0a8984

TiKeil force-pushed the energy-product-in-elliptic-discretizer branch from cf7ef1e to c0a8984 Compare August 24, 2020 15:25

sdrave reviewed Aug 25, 2020

View reviewed changes

Tim Keil and others added 3 commits August 25, 2020 17:35

Update src/pymordemos/elliptic2.py

3229c4c

Co-authored-by: Stephan Rave <stephanrave@uni-muenster.de>

Update src/pymor/discretizers/builtin/cg.py

c9848e4

Co-authored-by: Stephan Rave <stephanrave@uni-muenster.de>

[pymordemos] do not give a seed but the actual parameter

0b18cd3

sdrave approved these changes Aug 25, 2020

View reviewed changes

sdrave merged commit 289e7dc into pymor:master Aug 25, 2020

TiKeil deleted the energy-product-in-elliptic-discretizer branch August 26, 2020 10:07

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Energy product in elliptic discretizer (Continuation) #1008

Energy product in elliptic discretizer (Continuation) #1008

TiKeil commented Jul 9, 2020

TiKeil commented Jul 9, 2020

renefritze commented Jul 10, 2020

sdrave commented Jul 16, 2020

TiKeil commented Jul 16, 2020

renefritze commented Jul 16, 2020

TiKeil commented Jul 17, 2020

TiKeil commented Jul 17, 2020

renefritze commented Jul 17, 2020

codecov bot commented Jul 17, 2020 •

edited

Loading

TiKeil commented Jul 17, 2020

TiKeil commented Aug 3, 2020

TiKeil commented Aug 18, 2020

sdrave Aug 25, 2020

TiKeil Aug 25, 2020

		k: compute the energy norm of the last snapshot, where the energy-product is constructed
		with a randomly generated parameter instance with random seed integer k.

Energy product in elliptic discretizer (Continuation) #1008

Energy product in elliptic discretizer (Continuation) #1008

Conversation

TiKeil commented Jul 9, 2020

TiKeil commented Jul 9, 2020

renefritze commented Jul 10, 2020

sdrave commented Jul 16, 2020

TiKeil commented Jul 16, 2020

renefritze commented Jul 16, 2020

TiKeil commented Jul 17, 2020

TiKeil commented Jul 17, 2020

renefritze commented Jul 17, 2020

codecov bot commented Jul 17, 2020 • edited Loading

Codecov Report

TiKeil commented Jul 17, 2020

TiKeil commented Aug 3, 2020

TiKeil commented Aug 18, 2020

sdrave Aug 25, 2020

Choose a reason for hiding this comment

TiKeil Aug 25, 2020

Choose a reason for hiding this comment

codecov bot commented Jul 17, 2020 •

edited

Loading