- Beijing
Stars
A tool to modify ONNX models in a visualization fashion, based on Netron and Flask.
Compute CNN receptive field size in pytorch in one line
IREE plugin repository for the AMD AIE accelerator
An unnecessarily tiny implementation of GPT-2 in NumPy.
Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.
Symbolic Expression and Statement Module for new DSLs
LLaMA: Open and Efficient Foundation Language Models
HeteroCL: A Multi-Paradigm Programming Infrastructure for Software-Defined Heterogeneous Computing (FPGA'19 Best Paper)
HeteroCL-MLIR dialect for accelerator design
An API standard for single-agent reinforcement learning environments, with popular reference environments and related utilities (formerly Gym)
A scalable High-Level Synthesis framework on MLIR
An MLIR-based compiler framework bridges DSLs (domain-specific languages) to DSAs (domain-specific architectures).
NumPy aware dynamic Python compiler using LLVM
A high-performance, zero-overhead, extensible Python compiler with built-in NumPy support
Metaprogramming with Python, published by Packt
Template Metaprogramming with C++, published by Packt
Instant visualization of Python AST and Code Objects
BladeDISC is an end-to-end DynamIc Shape Compiler project for machine learning workloads.
Sample scripts and examples for my CPython Internals book
Dive into CPython internals, trying to illustrate every detail of CPython implementation