A DL Framework for Tensor Computation, inspired by Needle Framework in CMU 10-414/714: Deep Learning Systems.
git clone https://github.com/Paran0idy/Tenet.git
cd ./Tenet
make
- OpenAI Triton
- Pytorch
pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118
pip3 install triton
# using ndarray backend cuda, cpu or triton
export PYTHONPATH=./python && export NEEDLE_BACKEND=nd
- matmul
- MMA instruction using Tensor Cores, like Ampere Arch
mma.sync.align.m16n8k16
ldmatrix.sync
ldmatrix.trans.sync
- MMA instruction using Tensor Cores, like Ampere Arch
- reduce
- element-wise
- matmul
- reduce
- element-wise
- matmul
- reduce
- element-wise