| 13 Nov 2023 |
|  @ericjmorey:matrix.org changed their profile picture. | 16:33:01 |
|  @tuxtuxtuxtuxtux:opensuse.org joined the room. | 19:15:02 |
| 14 Nov 2023 |
| @tuxtuxtuxtuxtux:opensuse.org | Hi everyone | 07:52:18 |
 Hannes | Hi Tux! | 09:02:09 |
| 15 Nov 2023 |
|  rollskatflat joined the room. | 10:36:56 |
| rollskatflat | Hi, I am new to Julia, but have done plenty of MATLAB and scientific Python in my time.
I am just trying to understand how to vectorise an execution.
a .* b works just fine.
But now I have created a Vector{SVector} from the Rotations.jl package
I have used
push!(myVec, mySVec)
To build the data structure | 10:41:03 |
| rollskatflat | I then create a rotation matrix using QuatRotation, to create the quaternion.
myQuat * myVec[1]
Returns a result as expected
As do all subsequent indices
But, when I go
myVec .* myQuat | 10:42:39 |
| rollskatflat | DimensionMismatch: arrays could not be broadcast to a common size | 10:43:21 |
| rollskatflat | My guess is SVector is actually a static Vector and the .* Is trying to do an element wise multiply across the two, rather than treating myVec as a single value. | 10:53:33 |
 qwjyh | .* treated Vector{SVector} as you want.
In this case, myQuat is also broadcasted and cause dimension mismatch.
To handle this correctly, you need to wrap myQuat with Ref.
julia> Ref(my_quat) .* my_vec
5-element Vector{SVector{3, Float64}}:
For more information, see .* (and broadcast) in the help mode.
| 14:00:38 |
| qwjyh | * .* treated Vector{SVector} as you wanted.
In this case, myQuat is also broadcasted and cause dimension mismatch.
To handle this correctly, you need to wrap myQuat with Ref.
julia> Ref(my_quat) .* my_vec
5-element Vector{SVector{3, Float64}}:
For more information, see .* (and broadcast) in the help mode.
| 14:00:51 |
| qwjyh | * Indeed, .* treated Vector{SVector} as you wanted.
In this case, myQuat is also broadcasted and cause dimension mismatch.
To handle this correctly, you need to wrap myQuat with Ref.
julia> Ref(my_quat) .* my_vec
5-element Vector{SVector{3, Float64}}:
For more information, see .* (and broadcast) in the help mode.
| 15:52:44 |
| rollskatflat | Thank you so much! | 20:05:44 |
| 20 Nov 2023 |
|  Songtao Gui joined the room. | 07:11:30 |
| 28 Nov 2023 |
|  @norliana:matrix.org joined the room. | 18:50:55 |
| @norliana:matrix.org left the room. | 19:13:55 |
| 30 Nov 2023 |
|  @finlifin:matrix.org left the room. | 15:34:47 |
| 1 Dec 2023 |
|  @malte:maltee.de joined the room. | 10:30:28 |
| @malte:maltee.de | I'm trying to convert a system from discrete to continuous, not by using d2c but by replacing z^-1 by exp(-s*T) as this provides a more accurate representation. How would I do that (for a MIMO state space system)? | 10:41:38 |
| @malte:maltee.de | The best way I can think of is to separate the system in its matrices, assemble everything with a continuous instead of discrete delay and re-assign the input/output names. That's just very clunky for what should be a fairly simple operation. | 10:51:38 |
| @malte:maltee.de | can I even assign input and output names to delay systems? | 10:54:54 |
| @malte:maltee.de | If I could get an equivalent of setDelayModel from MATLAB, that would also suffice | 11:32:55 |
| 8 Dec 2023 |
|  csantosb joined the room. | 21:58:16 |
| 10 Dec 2023 |
| @malte:maltee.de | is there an easy way to use use a negative sign in connections of the connect function of RobustAndOptimalControl? | 19:29:37 |
| 14 Dec 2023 |
|  benjamin changed their profile picture. | 13:36:58 |
| 26 Dec 2023 |
| @malte:maltee.de | Is this room still alive or is this the widow of a libera.chat bridged room with all the action happening on the IRC side? | 08:52:15 |
| rollskatflat | People still use IRC a lot? | 09:51:26 |
| @malte:maltee.de | ¯\_(ツ)_/¯ | 09:52:46 |
| @malte:maltee.de | I noticed julialang.org mentions zulip, slack and discord, so this room is unofficial? | 09:53:22 |
| 27 Dec 2023 |
 vchuravy | it's not yet active enough to be official | 14:22:32 |
