jaewoosong / pocketnn
The official, proof-of-concept C++ implementation of PocketNN.
☆31Updated 5 months ago
Related projects ⓘ
Alternatives and complementary repositories for pocketnn
- Implementation of "NITI: Training Integer Neural Networks Using Integer-only Arithmetic" on arxiv☆77Updated 2 years ago
- Fast matrix multiplication for few-bit integer matrices on CPUs.☆27Updated 5 years ago
- Official implementation of "Searching for Winograd-aware Quantized Networks" (MLSys'20)☆27Updated last year
- CMix-NN: Mixed Low-Precision CNN Library for Memory-Constrained Edge Devices☆39Updated 4 years ago
- Implementation of convolution layer in different flavors☆67Updated 7 years ago
- Post-training sparsity-aware quantization☆33Updated last year
- ColTraIn HBFP Training Emulator☆16Updated last year
- This repository containts the pytorch scripts to train mixed-precision networks for microcontroller deployment, based on the memory contr…☆49Updated 6 months ago
- PyTorch extension for emulating FP8 data formats on standard FP32 Xeon/GPU hardware.☆100Updated 11 months ago
- ☆30Updated 3 years ago
- A Winograd Minimal Filter Implementation in CUDA☆23Updated 3 years ago
- Train neural networks with joint quantization and pruning on both weights and activations using any pytorch modules☆40Updated 2 years ago
- ☆123Updated last year
- INT-Q Extension of the CMSIS-NN library for ARM Cortex-M target☆18Updated 4 years ago
- This is the implementation for paper: AdaTune: Adaptive Tensor Program CompilationMade Efficient (NeurIPS 2020).☆13Updated 3 years ago
- CUDA templates for tile-sparse matrix multiplication based on CUTLASS.☆49Updated 6 years ago
- A tool to deploy Deep Neural Networks on PULP-based SoC's☆79Updated 8 months ago
- Improving Post Training Neural Quantization: Layer-wise Calibration and Integer Programming☆95Updated 3 years ago
- GEMM and Winograd based convolutions using CUTLASS☆25Updated 4 years ago
- QONNX: Arbitrary-Precision Quantized Neural Networks in ONNX☆127Updated 3 weeks ago
- Fork of upstream onnxruntime focused on supporting risc-v accelerators☆81Updated last year
- You Only Search Once: On Lightweight Differentiable Architecture Search for Resource-Constrained Embedded Platforms☆10Updated last year
- Chameleon: Adaptive Code Optimization for Expedited Deep Neural Network Compilation☆26Updated 5 years ago
- Implementation for the paper "Latent Weights Do Not Exist: Rethinking Binarized Neural Network Optimization"☆73Updated 4 years ago
- Reference implementations of popular Binarized Neural Networks☆104Updated 3 weeks ago
- Converting a deep neural network to integer-only inference in native C via uniform quantization and the fixed-point representation.☆21Updated 2 years ago
- Code for ICML 2021 submission☆35Updated 3 years ago
- ☆68Updated 2 years ago
- SAMO: Streaming Architecture Mapping Optimisation☆32Updated last year