https://github.com/witmemtech/algorithm-deployed-in-wtm2101

The deployment of the different NN algorithm on the WTM2101, this tutorial covers the entire process of deploying the nn algorithm on the WTM2101 chip, including CIM Chip, mnist data , mapper, project and studio

algorithms computinginmemory machine-learning-algorithms mapper mnistdemo

Last synced: 12 Apr 2025

https://github.com/witmemtech/cim-technical-papers-collection

Computing in memory optimizes data handling by performing operations directly in memory, ideal for high-speed data processing needs. This compilation highlights its technologies and applications, offering a concise overview for professionals in the field.

computinginmemory nerual-network paper papercollection witmem

Last synced: 19 Feb 2026

https://github.com/witmemtech/Witin-NN-Tool-

The "witin_nn" framework, based on PyTorch, maps neural networks to chip computations and supports operators including Linear, Conv2d, and GruCell. It enables 8-12 bit quantization for inputs/outputs and weights, implementing QAT.

computinginmemory conv2d convtranspose2d grucell linear nat natural-language-processing nerual-network pytorch qat

Last synced: 30 Sep 2025

https://github.com/witmemtech/witin-nn-tool-

The "witin_nn" framework, based on PyTorch, maps neural networks to chip computations and supports operators including Linear, Conv2d, and GruCell. It enables 8-12 bit quantization for inputs/outputs and weights, implementing QAT.

computinginmemory conv2d convtranspose2d grucell linear nat natural-language-processing nerual-network pytorch qat

Last synced: 03 Feb 2026

https://github.com/witmemtech/the-2nd-actic-test

在存算一体阵列中，权重的排布直接影响到存算单元的利用率与阵列的计算效率，是存算芯片在编译环节的重点优化对象。排布算法的效率、阵列面积的利用率以及阵列的计算效率是神经网络在存算一体芯片上部署的重要指标。

Last synced: 26 Jan 2026