Repository files navigation

The Torch-MLIR project aims to provide first class compiler support from thePyTorch ecosystem to the MLIR ecosystem.

This project is participating in the LLVM Incubator process: as such, it isnot part of any official LLVM release. While incubation status is notnecessarily a reflection of the completeness or stability of the code, itdoes indicate that the project is not yet endorsed as a component of LLVM.

PyTorchPyTorch is an open source machine learning framework that facilitates the seamless transition from research and prototyping to production-level deployment.

MLIRThe MLIR project offers a novel approach for building extensible and reusable compiler architectures, which address the issue of software fragmentation, reduce the cost of developing domain-specific compilers, improve compilation for heterogeneous hardware, and promote compatibility between existing compilers.

Torch-MLIRSeveral vendors have adopted MLIR as the middle layer in their systems, enabling them to map frameworks such as PyTorch, JAX, and TensorFlow into MLIR and subsequently lower them to their target hardware. We have observed half a dozen custom lowerings from PyTorch to MLIR, making it easier for hardware vendors to focus on their unique value, rather than needing to implement yet another PyTorch frontend for MLIR. The ultimate aim is to be similar to the current hardware vendors adding LLVM target support, rather than each one implementing Clang or a C++ frontend.

We have few paths to lower down to the Torch MLIR Dialect.

• ONNX as the entry points.
• Fx as the entry points

• #torch-mlir channel on the LLVMDiscord - this is the most active communication channel
• Github issueshere
• torch-mlir section of LLVM Discourse

At the time of writing, we releasepre-built snapshots of torch-mlir for Python 3.11 and Python 3.10.

If you have supported Python version, the following commands initialize a virtual environment.

python3.11 -m venv mlir_venvsource mlir_venv/bin/activate

Or, if you want to switch over multiple versions of Python using conda, you can create a conda environment with Python 3.11.

conda create -n torch-mlir python=3.11conda activate torch-mlirpython -m pip install --upgrade pip

Then, we can install torch-mlir with the corresponding torch and torchvision nightlies.

pip install --pre torch-mlir torchvision \  --extra-index-url https://download.pytorch.org/whl/nightly/cpu \  -f https://github.com/llvm/torch-mlir-release/releases/expanded_assets/dev-wheels

Torch-MLIR is primarily a project that is integrated into compilers to bridge them to PyTorch and ONNX. If contemplating a new integration, it may be helpful to refer to existing downstreams:

• IREE
• Blade

While most of the project is exercised via testing paths, there are some ways that an end user can directly use the APIs without further integration:

# Get the latest example if you haven't checked out the codewget https://raw.githubusercontent.com/llvm/torch-mlir/main/projects/pt1/examples/fximporter_resnet18.py# Run ResNet18 as a standalone script.python projects/pt1/examples/fximporter_resnet18.py# Outputload image from https://upload.wikimedia.org/wikipedia/commons/2/26/YellowLabradorLooking_new.jpg...PyTorch prediction[('Labrador retriever', 70.65674591064453), ('golden retriever', 4.988346099853516), ('Saluki, gazelle hound', 4.477451324462891)]torch-mlir prediction[('Labrador retriever', 70.6567153930664), ('golden retriever', 4.988325119018555), ('Saluki, gazelle hound', 4.477458477020264)]

The project follows the conventions of typical MLIR-based projects:

• include/torch-mlir,lib structure for C++ MLIR compiler dialects/passes.
• test for holding test code.
• tools fortorch-mlir-opt and such.
• python top level directory for Python code

If you would like to develop and build torch-mlir from source please look atDevelopment Notes