feng wang

ceras. Ceras is yet another tiny deep learning engine, in pure c++ and header only.

matrix. A modern, C++20-native, single-file header-only dense 2D matrix library.

Stable-Diffusion-NCNN. Stable Diffusion inference in pure C++

MCNN. Demo of Multi-scale Deep Convolutional Neural Networks

float16_t. CPP20 implementation of a 16-bit floating-point type mimicking most of the IEEE 754 behavior. Single file and header-only.

nbody-simulation. a simulation of a dynamical system of particles

LLaMA-Factory-docker. Dockerfile

colorize. A Library to colorize program output to terminals

subpixel_conv2d. Sub-Pixel Convolutional Layer with Tensorflow/Keras

Noise2Atom. Unsupervised Denoising for STEM Images.

super_resolution. Image super resolution using deep convolutional neural networks.

derivative. numerical derivative lib

minimal-capsule. CapsNet implementation in a minimal manner

random_variate_generator. random variate generator lib, written in c++, generic style, compatible with STL.

tuple_binding. Simple relfection in C++17

nnl. a low-latency and high-performance inference engine for large models on low-memory GPU platform.

EfficientNetAndroid. Deploying EfficientNet to Android platform using flutter-tflite

mcnn-demo. Demonstration of multi-scale convolutional neural networks

meta-structure. Meta structure CRUD in C++20

recaption_images. Rename Images with Auto Generated Neural Image Captions

fdtd-tm. a electro-magnetic field fdtd solver

better_assert. Yet another assert macro, enabling additional debug information dump.

o_o. o_o: Using Ollama to create o1-like reasoning chains

electron_diffraction_simulation. Electron Diffraction Simulation

larbed-refinement. Dynamic inverse of LARBED patterns

lua_interpreter. a lua interpreter using boost::spirit

tetris-lspe. Reinforcement learning for tetris based on least squares policy evaluation method

clc_book_structure. Python

simple_json.hpp. A tiny little JSON parsing library in C++26 with Reflection

VITS-fast-fine-tuning-docker. This docker brings up a pipeline of VITS finetuning for fast speaker adaptation TTS, and many-to-many voice conversion

operator_traits. Checks if an operator exist or not.

poster-template. My post template

ga. genetic algorithm in c++

di. projects related to STEM

cumar. CUDA Map and Reduce in CPP

fmt. A new `fmt::format` wheeeeeeeeeeeeeeeeeeeel.

onnx-modifier. A tool to modify onnx models in a visualization fashion, based on Netron and flask.

CReLU. CReLU layer for TensorFlow/Keras

ImagePlayer. Deep image models at ease.

color_spray. STEM images colorization using Deep Convolutional Neural Networks.

nn-denoising. TEM image denoising with artifical neural network

sparse-matrix. sparse matrix with related numeric algorithms

Noisy_Predestination. Self-supervised denoising

redux. Redux pattern in C++

cunn. Deep neural network with multi-GPU support in a minimal fashion