Skip to content

Simulator for Executing Matrix-Vector Multiplications on Analog Computing-in-Memory Crossbars

License

Notifications You must be signed in to change notification settings

rpelke/analog-cim-sim

Repository files navigation

Analog-CIM-Sim

CMake Style Python Coverage (lines)

Yet another simulator for executing matrix-vector multiplications on analog computing-in-memory crossbars.

Build Instructions

To build the project, you need cmake and a dev version of python. If you don't have python3-dev, pybind won't compile.

Execute the following steps (replace the placeholders):

python3 -m venv .venv
source .venv/bin/activate
pip3 install -r requirements.txt

mkdir -p build/release/build && cd build/release/build

# <proj-dir> should be an absolute path to the analog-cim-sim repository
# <python-version> can be found in your .venv, e.g., python3.8
cmake \
    -DCMAKE_BUILD_TYPE=Release \
    -DPY_INSTALL_PATH=<proj-dir>/.venv/lib/<python-version>/site-packages \
    -DCMAKE_PREFIX_PATH=<proj-dir>/.venv/lib/<python-version>/site-packages/pybind11/share/cmake/pybind11 \
    -DCMAKE_INSTALL_PREFIX=../ \
    -DLIB_TESTS=ON \
    -DBUILD_LIB_CB_EMU=ON \
    -DBUILD_LIB_ACS_INT=ON \
    # -DDEBUG_MODE=ON \    # Add -DDEBUG_MODE=ON to enable debug output of matrix operations (cpy_mtrx and mvm)
    ../../../cpp

make -j `nproc`
make install

Execute the tests:

python3 -m unittest discover -s int-bindings/test -p '*_test.py'

To detect segmentation faults in the C++ part, you can also run:

gdb --batch --ex="run" --ex="bt" --ex="quit" --args python3 -m unittest discover -s int-bindings/test -p '*_test.py'

About

Simulator for Executing Matrix-Vector Multiplications on Analog Computing-in-Memory Crossbars

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •