Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[PyTorch] Update to 2.1.2 #1444

Merged
merged 3 commits into from
Dec 17, 2023
Merged

[PyTorch] Update to 2.1.2 #1444

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
5156aba
Add missing mapping for `std::array<Stat>``
HGuillemet Dec 2, 2023
25d3c8b
Update PyTorch to 2.1.2
HGuillemet Dec 15, 2023
c5f2ccd
Update CHANGELOG.md and fix nits
saudet Dec 16, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@
* Refactor and improve presets for PyTorch ([pull #1360](https://github.com/bytedeco/javacpp-presets/pull/1360))
* Include `mkl_lapack.h` header file in presets for MKL ([issue #1388](https://github.com/bytedeco/javacpp-presets/issues/1388))
* Map new higher-level C++ API of Triton Inference Server ([pull #1361](https://github.com/bytedeco/javacpp-presets/pull/1361))
* Upgrade presets for OpenCV 4.8.1, FFmpeg 6.1, HDF5 1.14.3, DNNL 3.3.2, OpenBLAS 0.3.25, ARPACK-NG 3.9.1, CPython 3.12.0, NumPy 1.26.2, SciPy 1.11.4, LLVM 17.0.6, Leptonica 1.83.1, Tesseract 5.3.3, CUDA 12.3.0, cuDNN 8.9.5, NCCL 2.18.5, PyTorch 2.1.1 ([pull #1426](https://github.com/bytedeco/javacpp-presets/pull/1426)), TensorFlow Lite 2.15.0, Triton Inference Server 2.38.0, DepthAI 2.23.0, ONNX 1.15.0, ONNX Runtime 1.16.3, TVM 0.14.0, and their dependencies
* Upgrade presets for OpenCV 4.8.1, FFmpeg 6.1, HDF5 1.14.3, DNNL 3.3.2, OpenBLAS 0.3.25, ARPACK-NG 3.9.1, CPython 3.12.0, NumPy 1.26.2, SciPy 1.11.4, LLVM 17.0.6, Leptonica 1.83.1, Tesseract 5.3.3, CUDA 12.3.0, cuDNN 8.9.5, NCCL 2.18.5, PyTorch 2.1.2 ([pull #1426](https://github.com/bytedeco/javacpp-presets/pull/1426)), TensorFlow Lite 2.15.0, Triton Inference Server 2.38.0, DepthAI 2.23.0, ONNX 1.15.0, ONNX Runtime 1.16.3, TVM 0.14.0, and their dependencies

### June 6, 2023 version 1.5.9
* Virtualize `nvinfer1::IGpuAllocator` from TensorRT to allow customization ([pull #1367](https://github.com/bytedeco/javacpp-presets/pull/1367))
Expand Down
2 changes: 1 addition & 1 deletion platform/pom.xml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -292,7 +292,7 @@
<dependency>
<groupId>org.bytedeco</groupId>
<artifactId>pytorch-platform</artifactId>
<version>2.1.1-${project.version}</version>
<version>2.1.2-${project.version}</version>
</dependency>
<dependency>
<groupId>org.bytedeco</groupId>
Expand Down
6 changes: 3 additions & 3 deletions pytorch/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,7 +9,7 @@ Introduction
------------
This directory contains the JavaCPP Presets module for:

* PyTorch 2.1.1 https://pytorch.org/
* PyTorch 2.1.2 https://pytorch.org/

Please refer to the parent README.md file for more detailed information about the JavaCPP Presets.

Expand Down Expand Up @@ -48,14 +48,14 @@ We can use [Maven 3](http://maven.apache.org/) to download and install automatic
<dependency>
<groupId>org.bytedeco</groupId>
<artifactId>pytorch-platform</artifactId>
<version>2.1.1-1.5.10-SNAPSHOT</version>
<version>2.1.2-1.5.10-SNAPSHOT</version>
</dependency>

<!-- Additional dependencies required to use CUDA, cuDNN, and NCCL -->
<dependency>
<groupId>org.bytedeco</groupId>
<artifactId>pytorch-platform-gpu</artifactId>
<version>2.1.1-1.5.10-SNAPSHOT</version>
<version>2.1.2-1.5.10-SNAPSHOT</version>
</dependency>

<!-- Additional dependencies to use bundled CUDA, cuDNN, and NCCL -->
Expand Down
2 changes: 1 addition & 1 deletion pytorch/cppbuild.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,7 +35,7 @@ if [[ $PLATFORM == windows* ]]; then
export PYTHON_BIN_PATH=$(which python.exe)
fi

PYTORCH_VERSION=2.1.1
PYTORCH_VERSION=2.1.2

export PYTORCH_BUILD_VERSION="$PYTORCH_VERSION"
export PYTORCH_BUILD_NUMBER=1
Expand Down
2 changes: 1 addition & 1 deletion pytorch/platform/gpu/pom.xml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@

<groupId>org.bytedeco</groupId>
<artifactId>pytorch-platform-gpu</artifactId>
<version>2.1.1-${project.parent.version}</version>
<version>2.1.2-${project.parent.version}</version>
<name>JavaCPP Presets Platform GPU for PyTorch</name>

<properties>
Expand Down
2 changes: 1 addition & 1 deletion pytorch/platform/pom.xml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@

<groupId>org.bytedeco</groupId>
<artifactId>pytorch-platform</artifactId>
<version>2.1.1-${project.parent.version}</version>
<version>2.1.2-${project.parent.version}</version>
<name>JavaCPP Presets Platform for PyTorch</name>

<properties>
Expand Down
2 changes: 1 addition & 1 deletion pytorch/pom.xml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,7 +11,7 @@

<groupId>org.bytedeco</groupId>
<artifactId>pytorch</artifactId>
<version>2.1.1-${project.parent.version}</version>
<version>2.1.2-${project.parent.version}</version>
<name>JavaCPP Presets for PyTorch</name>

<dependencies>
Expand Down
4 changes: 2 additions & 2 deletions pytorch/samples/pom.xml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,14 +12,14 @@
<dependency>
<groupId>org.bytedeco</groupId>
<artifactId>pytorch-platform</artifactId>
<version>2.1.1-1.5.10-SNAPSHOT</version>
<version>2.1.2-1.5.10-SNAPSHOT</version>
</dependency>

<!-- Additional dependencies required to use CUDA, cuDNN, and NCCL -->
<dependency>
<groupId>org.bytedeco</groupId>
<artifactId>pytorch-platform-gpu</artifactId>
<version>2.1.1-1.5.10-SNAPSHOT</version>
<version>2.1.2-1.5.10-SNAPSHOT</version>
</dependency>

<!-- Additional dependencies to use bundled CUDA, cuDNN, and NCCL -->
Expand Down
18 changes: 9 additions & 9 deletions pytorch/src/gen/java/org/bytedeco/pytorch/cuda/DeviceStats.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,25 +42,25 @@ public class DeviceStats extends Pointer {
}

// COUNT: allocations requested by client code
public native @ByRef @Cast("c10::cuda::CUDACachingAllocator::StatArray*") BoolPointer allocation(); public native DeviceStats allocation(BoolPointer setter);
public native @ByRef @Cast("c10::cuda::CUDACachingAllocator::StatArray*") Stat allocation(); public native DeviceStats allocation(Stat setter);
// COUNT: number of allocated segments from cudaMalloc().
public native @ByRef @Cast("c10::cuda::CUDACachingAllocator::StatArray*") BoolPointer segment(); public native DeviceStats segment(BoolPointer setter);
public native @ByRef @Cast("c10::cuda::CUDACachingAllocator::StatArray*") Stat segment(); public native DeviceStats segment(Stat setter);
// COUNT: number of active memory blocks (allocated or used by stream)
public native @ByRef @Cast("c10::cuda::CUDACachingAllocator::StatArray*") BoolPointer active(); public native DeviceStats active(BoolPointer setter);
public native @ByRef @Cast("c10::cuda::CUDACachingAllocator::StatArray*") Stat active(); public native DeviceStats active(Stat setter);
// COUNT: number of inactive, split memory blocks (unallocated but can't be
// released via cudaFree)
public native @ByRef @Cast("c10::cuda::CUDACachingAllocator::StatArray*") BoolPointer inactive_split(); public native DeviceStats inactive_split(BoolPointer setter);
public native @ByRef @Cast("c10::cuda::CUDACachingAllocator::StatArray*") Stat inactive_split(); public native DeviceStats inactive_split(Stat setter);

// SUM: bytes allocated by this memory alocator
public native @ByRef @Cast("c10::cuda::CUDACachingAllocator::StatArray*") BoolPointer allocated_bytes(); public native DeviceStats allocated_bytes(BoolPointer setter);
public native @ByRef @Cast("c10::cuda::CUDACachingAllocator::StatArray*") Stat allocated_bytes(); public native DeviceStats allocated_bytes(Stat setter);
// SUM: bytes reserved by this memory allocator (both free and used)
public native @ByRef @Cast("c10::cuda::CUDACachingAllocator::StatArray*") BoolPointer reserved_bytes(); public native DeviceStats reserved_bytes(BoolPointer setter);
public native @ByRef @Cast("c10::cuda::CUDACachingAllocator::StatArray*") Stat reserved_bytes(); public native DeviceStats reserved_bytes(Stat setter);
// SUM: bytes within active memory blocks
public native @ByRef @Cast("c10::cuda::CUDACachingAllocator::StatArray*") BoolPointer active_bytes(); public native DeviceStats active_bytes(BoolPointer setter);
public native @ByRef @Cast("c10::cuda::CUDACachingAllocator::StatArray*") Stat active_bytes(); public native DeviceStats active_bytes(Stat setter);
// SUM: bytes within inactive, split memory blocks
public native @ByRef @Cast("c10::cuda::CUDACachingAllocator::StatArray*") BoolPointer inactive_split_bytes(); public native DeviceStats inactive_split_bytes(BoolPointer setter);
public native @ByRef @Cast("c10::cuda::CUDACachingAllocator::StatArray*") Stat inactive_split_bytes(); public native DeviceStats inactive_split_bytes(Stat setter);
// SUM: bytes requested by client code
public native @ByRef @Cast("c10::cuda::CUDACachingAllocator::StatArray*") BoolPointer requested_bytes(); public native DeviceStats requested_bytes(BoolPointer setter);
public native @ByRef @Cast("c10::cuda::CUDACachingAllocator::StatArray*") Stat requested_bytes(); public native DeviceStats requested_bytes(Stat setter);

// COUNT: total number of failed calls to CUDA malloc necessitating cache
// flushes.
Expand Down
4 changes: 2 additions & 2 deletions pytorch/src/gen/java/org/bytedeco/pytorch/global/torch.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -76579,11 +76579,11 @@ scalar_t sf(scalar_t x, scalar_t y)
public static final int TORCH_VERSION_MINOR = 1;

/** Indicates the patch version of LibTorch. */
public static final int TORCH_VERSION_PATCH = 1;
public static final int TORCH_VERSION_PATCH = 2;

/** Indicates the version of LibTorch. */
public static final String TORCH_VERSION =
"2.1.1";
"2.1.2";


// Parsed from torch/csrc/autograd/InferenceMode.h
Expand Down
3 changes: 3 additions & 0 deletions pytorch/src/main/java/org/bytedeco/pytorch/presets/torch_cuda.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -104,6 +104,9 @@ public void map(InfoMap infoMap) {
.put(new Info("std::vector<c10::cuda::CUDAKernelLaunchInfo>").pointerTypes("CUDAKernelLaunchInfoVector").define())
.put(new Info("const std::vector<c10::cuda::CUDACachingAllocator::TraceEntry>", "std::vector<c10::cuda::CUDACachingAllocator::TraceEntry>").pointerTypes("TraceEntryVector").define())

//// std::array
.put(new Info("std::array<c10::cuda::CUDACachingAllocator::Stat,3>", "c10::cuda::CUDACachingAllocator::StatArray").cast().pointerTypes("Stat"))

//// Function pointers
// Function pointer returning shared_ptr don't compile on windows
// "D:\a\javacpp-presets\javacpp-presets\pytorch\target\native\org\bytedeco\pytorch\windows-x86_64\jnitorch.cpp(98904): error C2526: 'JavaCPP_org_bytedeco_pytorch_functions_GatheredContextSupplier_allocate_callback': C linkage function cannot return C++ class 'std::shared_ptr<c10::GatheredContext>'"
Expand Down