Comments (13)
HeavyMetalGeek
commented on May 30, 2024
1
Using the following within Cargo.toml:
[dependencies]
argmin = { version = "0.4.4", features = ["ndarrayl"] }
argmin_testfunctions = "*"
ndarray = { version = "0.14", features = ["serde"] }
and adding extern crate argmin; to the previous minimal example yields linker errors. Removal of extern crate argmin; yields a successful build.
from argmin.
HeavyMetalGeek
commented on May 30, 2024
1
Not at all. I just wanted to be clear about my inexperience so that you could factor that into your troubleshooting.
I added to my Cargo.toml:
ndarray-linalg = { version = "0.13", features = ["intel-mkl-static"] }
and that solved the linker issues, yielding a successful build. May I suggest adding something about this to the readme, under "Usage" for Windows users?
Thank you so much for your help and patience.
from argmin.
stefan-k
commented on May 30, 2024
Hi,
if you are not using nalgebra then I would suggest removing the nalgebral feature from your Cargo.toml. Your other issue seems to be coming from the wrong version of ndarray. argmin uses ndarray-linalg which so far does not support version 0.15 of ndarray. I think this problem will resolve when you switch to version 0.14 of ndarray.
from argmin.
HeavyMetalGeek
commented on May 30, 2024
[dependencies]
argmin = { version = "0.4.4", features = ["ctrlc", "ndarrayl"] }
argmin_testfunctions = "*"
ndarray = { version = "0.14", features = ["serde"] }
resulted in "error: linking with link.exe failed: exit code: 1120"
[dependencies]
argmin = { version = "0.4.4", features = ["ctrlc"] }
argmin_testfunctions = "*"
ndarray = { version = "0.14", features = ["serde"] }
resulted in the same errors I previously posted.
from argmin.
stefan-k
commented on May 30, 2024
The first one should be correct, the second one can't work. I have never compiled argmin on windows and I don't have a windows computer available, therefore I can't debug this unfortunately. But since it is a linking error it may be related to linking against a BLAS library or something. Can you compile your code without the argmin dependency/code, just with ndarray?
from argmin.
HeavyMetalGeek
commented on May 30, 2024
I successfully compiled and ran a minimal example:
extern crate ndarray;
use ndarray::{array, Array1, Array2};
fn main() {
let a = vec![array![1.0, 2.0], array![3.0, 4.0]];
for i in &a {
println!("{}", i);
}
}
without issue. Let me know if there is something else you would like me to try.
from argmin.
stefan-k
commented on May 30, 2024
I think you might be missing an external library, but I don't know which.
Could you try to compile without the ctrlc feature?
from argmin.
stefan-k
commented on May 30, 2024
Thanks a lot for reporting this and your fix! Rust removed the necessity to use extern crate a while ago, but if I remember correctly, it was still necessary to use it in examples. Apparently it is not necessary anymore, therefore I've removed it from the examples in PR #116.
from argmin.
HeavyMetalGeek
commented on May 30, 2024
I'm sorry if I was unclear. I did not find a fix. However, I did find that removing features=["ndarrayl"] from my argmin dependency took care of my linker errors. The following still produces errors with NelderMead::new() and sd_tolerance():
Cargo.toml
[dependencies]
argmin = { version = "0.4.4" }
argmin_testfunctions = "*"
ndarray = { version = "0.14", features = ["serde"] }
main.rs
use argmin::prelude::*;
use argmin::solver::neldermead::NelderMead;
use argmin_testfunctions::rosenbrock;
use ndarray::{array, Array1, Array2};
/// First, create a struct for your problem
struct Rosenbrock {
a: f64,
b: f64,
}
/// Implement `ArgminOp` for `Rosenbrock`
impl ArgminOp for Rosenbrock {
type Param = Array1<f64>;
type Output = f64;
type Hessian = Array2<f64>;
type Jacobian = ();
type Float = f64;
fn apply(&self, p: &Self::Param) -> Result<Self::Output, Error> {
Ok(rosenbrock(&p.to_vec(), self.a, self.b))
}
}
fn run() -> Result<(), Error> {
// Define cost function
let cost = Rosenbrock { a: 1.0, b: 100.0 };
// Set up solver -- note that the proper choice of the vertices is very important!
let solver = NelderMead::new()
.with_initial_params(vec![
// array![-2.0, 3.0],
// array![-2.0, -1.0],
// array![2.0, -1.0],
array![-1.0, 3.0],
array![2.0, 1.5],
array![2.0, -1.0],
])
.sd_tolerance(0.0001);
// Run solver
let res = Executor::new(cost, solver, array![])
.add_observer(ArgminSlogLogger::term(), ObserverMode::Always)
.max_iters(100)
.run()?;
// Wait a second (lets the logger flush everything before printing again)
std::thread::sleep(std::time::Duration::from_secs(1));
// Print result
println!("{}", res);
Ok(())
}
fn main() {
if let Err(ref e) = run() {
println!("{}", e);
std::process::exit(1);
}
}Produces the following errors
error[E0277]: the trait bound `ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>: ArgminAdd<ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>, ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>>` is not satisfied
--> src\main.rs:39:30
|
39 | .with_initial_params(vec![
| ______________________________^
40 | | // array![-2.0, 3.0],
41 | | // array![-2.0, -1.0],
42 | | // array![2.0, -1.0],
... |
45 | | array![2.0, -1.0],
46 | | ])
| |_________^ the trait `ArgminAdd<ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>, ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>>` is not implemented for `ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>`
|
= note: this error originates in a macro (in Nightly builds, run with -Z macro-backtrace for more info)
error[E0277]: the trait bound `ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>: ArgminSub<ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>, ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>>` is not satisfied
--> src\main.rs:39:30
|
39 | .with_initial_params(vec![
| ______________________________^
40 | | // array![-2.0, 3.0],
41 | | // array![-2.0, -1.0],
42 | | // array![2.0, -1.0],
... |
45 | | array![2.0, -1.0],
46 | | ])
| |_________^ the trait `ArgminSub<ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>, ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>>` is not implemented for `ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>`
|
= note: this error originates in a macro (in Nightly builds, run with -Z macro-backtrace for more info)
error[E0277]: the trait bound `ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>: ArgminMul<_, ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>>` is not satisfied
--> src\main.rs:39:30
|
39 | .with_initial_params(vec![
| ______________________________^
40 | | // array![-2.0, 3.0],
41 | | // array![-2.0, -1.0],
42 | | // array![2.0, -1.0],
... |
45 | | array![2.0, -1.0],
46 | | ])
| |_________^ the trait `ArgminMul<_, ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>>` is not implemented for `ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>`
|
= note: this error originates in a macro (in Nightly builds, run with -Z macro-backtrace for more info)
error[E0277]: the trait bound `ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>: ArgminAdd<ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>, ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>>` is not satisfied
--> src\main.rs:38:18
|
38 | let solver = NelderMead::new()
| ^^^^^^^^^^^^^^^ the trait `ArgminAdd<ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>, ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>>` is not implemented for `ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>`
|
= note: required by `NelderMead::<P, F>::new`
error[E0277]: the trait bound `ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>: ArgminSub<ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>, ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>>` is not satisfied
--> src\main.rs:38:18
|
38 | let solver = NelderMead::new()
| ^^^^^^^^^^^^^^^ the trait `ArgminSub<ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>, ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>>` is not implemented for `ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>`
|
= note: required by `NelderMead::<P, F>::new`
error[E0277]: the trait bound `ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>: ArgminMul<_, ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>>` is not satisfied
--> src\main.rs:38:18
|
38 | let solver = NelderMead::new()
| ^^^^^^^^^^^^^^^ the trait `ArgminMul<_, ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>>` is not implemented for `ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>`
|
= note: required by `NelderMead::<P, F>::new`
error[E0599]: no method named `sd_tolerance` found for struct `NelderMead<ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>, _>` in the current scope
--> src\main.rs:47:10
|
47 | .sd_tolerance(0.0001);
| ^^^^^^^^^^^^ private field, not a method
|
::: C:\Users\Nobody\.cargo\registry\src\github.com-1ecc6299db9ec823\ndarray-0.14.0\src\lib.rs:1228:1
|
1228 | / pub struct ArrayBase<S, D>
1229 | | where
1230 | | S: RawData,
1231 | | {
... |
1241 | | strides: D,
1242 | | }
| | -
| | |
| | doesn't satisfy `_: ArgminAdd<ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>, ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>>`
| |_doesn't satisfy `_: ArgminMul<_, ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>>`
| doesn't satisfy `_: ArgminSub<ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>, ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>>`
|
= note: the method `sd_tolerance` exists but the following trait bounds were not satisfied:
`ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>: ArgminAdd<ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>, ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>>`
`ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>: ArgminSub<ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>, ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>>`
`ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>: ArgminMul<_, ArrayBase<OwnedRepr<{float}>, Dim<[usize; 1]>>>`
error: aborting due to 7 previous errors; 1 warning emitted
from argmin.
stefan-k
commented on May 30, 2024
You need the ndarrayl feature when you want to use ndarray arrays with argmin, otherwise the corresponding math traits aren't implemented for the ndarray types.
I'm a bit confused, if removing extern crate argmin; didn't solve this problem, what did it solve then?
from argmin.
HeavyMetalGeek
commented on May 30, 2024
I'm sorry. I'm fairly new to Rust and am probably not conveying details in an efficient manner. I was simply stating that, while still having the ndarrayl feature in the argmin dependencies, the argmin packages (or its dependencies) were the source of the linker errors, even after removing the ctrlc feature.
After adding the ndarrayl feature back into the argmin dependency, I copied everything to Replit to test in a linux environment and it builds fine. On Windows, however, I get linker errors and I can't seem to figure out what could be the cause. I thought maybe rust-lang/rustup#1455 would be a good reference, but I have only run into linker errors when using the argmin crate and I already had the Windows SDK installed from my Visual Studio installation and adding the path to link.exe to my PATH environment variable made no impact.
If there is anything I can do to weed out the issue and convey the results more efficiently, I am certainly open to criticism.
from argmin.
stefan-k
commented on May 30, 2024
I apologize if I sounded rude, that was not my intention.
Thank you for clarifying, I think I understand now. Just to summarize: You will need the ndarrayl feature, therefore the linking issue is the one that needs fixing (not the trait not implemented errors). ndarray-linalg, which is also used by argmin, links against a BLAS backend. According to their readme, the only BLAS backend which works on Windows is Intel-MKL. Do you have this installed? If not, I suggest you try to install it. You might even have to add ndarray-linalg to your Cargo.toml:
ndarray-linalg = { version = "0.13", features = ["intel-mkl-static"] } # apparently downloads Intel MKL itself, links staticallyndarray-linalg = { version = "0.13", features = ["intel-mkl-system"] } # Uses system Intel MKL, links dynamicallyThe best way to check if ndarray-linalg is causing this issue would probably be to try to compile one of their examples
from argmin.
stefan-k
commented on May 30, 2024
Glad that this solved the problem! Thanks for giving all the various ideas a try. I will expand the readme in the near future to avoid this problem in the future. Thanks!
from argmin.
Related Issues (20)
- Missing proper documentation on how to run examples
- Create inplace variants of math traits
- ParticleSwarm is not reproducible HOT 3
- More-Thuente Linesearch Algorithm Bug: "Search direction must be a descent direction" HOT 7
- BrentRoot is missing tests
- Add `faer-rs` as math backend
- Add stopping criteria based on the number of function evaluations
- Add LaTeX support to docs
- Add derivatives and Hessians to test functions in argmin_testfunctions
- Add more test functions to `argmin_testfunctions`
- Create visualization of all test functions in `argmin_testfunctions`
- `owl-qn` example does not converge HOT 4
- Observers and checkpointing are missing tests
- Enable observers to create final report
- Thoughts on future state handling
- Various non productive overhead HOT 1
- Example (lbfgs) cannot be compiled after upgrade to version 0.10 HOT 2
- rss/atom feed of the blog HOT 1
- newton and newton cg freeze rather than stopping HOT 1
- Difference between bfgs.rs and bfgs of optimtool.unconstrain.newton_quasi HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from argmin.