Could not find jnitorch in class, module, and library paths.

[8bitreid]

I'm on a M1 Mac.
Exception in thread "main" java.lang.UnsatisfiedLinkError: no jnitorch in java.library.path: /Users/USER/Library/Java/Extensions:/Library/Java/Extensions:/Network/Library/Java/Extensions:/System/Library/Java/Extensions:/usr/lib/java:

Python

➜ python3
Python 3.11.6 (main, Oct  2 2023, 13:45:54) [Clang 15.0.0 (clang-1500.0.40.1)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import platform
>>> platform.platform()
'macOS-14.1-arm64-arm-64bit'

Java

~/code/storch-project took 3m 15s
➜ java -version
openjdk version "20.0.2" 2023-07-18
OpenJDK Runtime Environment (build 20.0.2+9-78)
OpenJDK 64-Bit Server VM (build 20.0.2+9-78, mixed mode, sharing)

[sbrunk]

Just to be sure it is really due to the M1 arch. How did you add the native libraries? See https://storch.dev/installation.html#via-pytorch-platform-1 for examples (best try with pytorch-platform if you haven't done already).

[8bitreid]

this compiles. My project is still not running though. I'm going change some code and see if this resolved.

[8bitreid]

@sbrunk

resolvers ++= Resolver.sonatypeOssRepos("snapshots")
lazy val root = (project in file("."))
  .settings(
    name := "storch-project",
    libraryDependencies ++= Seq(
      "dev.storch" %% "core" % "0.0-05078a8-SNAPSHOT",
      "org.bytedeco" % "pytorch-platform" % "2.0.1-1.5.10-SNAPSHOT"
    )
  )

fork := true

this is what I have.

[8bitreid]

I'm using IntelliJ CE.

[8bitreid]

seeing this

[error] Exception in thread "main" java.lang.UnsatisfiedLinkError: no jniopenblas_nolapack in java.library.path: /Users/USERNAME/Library/Java/Extensions:/Library/Java/Extensions:/Network/Library/Java/Extensions:/System/Library/Java/Extensions:/usr/lib/java:.
[error] 	at java.base/java.lang.ClassLoader.loadLibrary(ClassLoader.java:2434)
[error] 	at java.base/java.lang.Runtime.loadLibrary0(Runtime.java:818)
[error] 	at java.base/java.lang.System.loadLibrary(System.java:1989)
[error] 	at org.bytedeco.javacpp.Loader.loadLibrary(Loader.java:1832)
[error] 	at org.bytedeco.javacpp.Loader.load(Loader.java:1423)
[error] 	at org.bytedeco.javacpp.Loader.load(Loader.java:1234)
[error] 	at org.bytedeco.javacpp.Loader.load(Loader.java:1210)
[error] 	at org.bytedeco.openblas.global.openblas_nolapack.<clinit>(openblas_nolapack.java:12)
[error] 	at java.base/java.lang.Class.forName0(Native Method)
[error] 	at java.base/java.lang.Class.forName(Class.java:467)
[error] 	at org.bytedeco.javacpp.Loader.load(Loader.java:1289)
[error] 	at org.bytedeco.javacpp.Loader.load(Loader.java:1234)
[error] 	at org.bytedeco.javacpp.Loader.load(Loader.java:1210)
[error] 	at org.bytedeco.pytorch.TensorOptions.<clinit>(TensorOptions.java:105)
[error] 	at torch.internal.NativeConverters$.tensorOptions(NativeConverters.scala:111)
[error] 	at torch.Tensor$.apply(Tensor.scala:995)
[error] 	at devcake.ExampleStorch$.<clinit>(ExampleStorch.scala:9)
[error] 	at devcake.ExampleStorch.main(ExampleStorch.scala)
[error] Caused by: java.lang.UnsatisfiedLinkError: Could not find jniopenblas_nolapack in class, module, and library paths.
[error] 	at org.bytedeco.javacpp.Loader.loadLibrary(Loader.java:1799)
[error] 	... 14 more
[error] Nonzero exit code returned from runner: 1
[error] (Compile / run) Nonzero exit code returned from runner: 1
[error] Total time: 2 s, completed Oct 31, 2023, 10:53:43 PM

[sbrunk]

We don't have native ARM builds yet unfortunately, but I'm hoping that Apple's compat layer will make it work with the x86_64 native binaries nevertheless until then. Could you try setting the classifier explicitly:

libraryDependencies += Seq(
  "dev.storch" %% "core" % "0.0-05078a8-SNAPSHOT",
  "org.bytedeco" % "pytorch" % "2.0.1-1.5.10-SNAPSHOT",
  "org.bytedeco" % "pytorch" % "2.0.1-1.5.10-SNAPSHOT" classifier "macosx-x86_64",
  "org.bytedeco" % "openblas" % "0.3.23-1.5.10-SNAPSHOT" classifier "macosx-x86_64"
)

[8bitreid]

I have tried this step, but was unsuccessful. I can check the error tonight - but IIRC it was not the same error. However, it failed. Also, I saw a javacpp bug around M1 using a snapshot. I have not tried to do this but here is the link bytedeco/javacpp-presets#1069

…

On Thu, Nov 2, 2023 at 12:13 PM Sören Brunk ***@***.***> wrote: We don't have native ARM builds yet unfortunately, but I'm hoping that Apple's compat layer will make it work with the x86_64 native binaries nevertheless until then. Could you try setting the classifier explicitly: libraryDependencies += Seq( "dev.storch" %% "core" % "0.0-05078a8-SNAPSHOT", "org.bytedeco" % "pytorch" % "2.0.1-1.5.10-SNAPSHOT", "org.bytedeco" % "pytorch" % "2.0.1-1.5.10-SNAPSHOT" classifier "macosx-x86_64", "org.bytedeco" % "openblas" % "0.3.23-1.5.10-SNAPSHOT" classifier "macosx-x86_64" ) — Reply to this email directly, view it on GitHub <#65 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACFQR4ASHPTHRNH4PTYBNRDYCPIDJAVCNFSM6AAAAAA6XA3SPOVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM3TINJZGMZDO> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[sbrunk]

Yes please try to run it again and post the error message.