https://github.com/Moe-Zbeeb/Optimizing-Deep-Neural-Networks-via-Safety-Guided-Self-Compression
공식 코드Optimizing Deep Neural Networks using Safety-Guided Self Compression
Papers with Code 아카이브
Quantization
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
- 하위 태스크
- 2
- 평가표
- 10
- 보존된 평가 기록
- 32
Quantization is a promising technique to reduce the computation cost of neural network training, which can replace high-cost floating-point numbers (e.g., float32) with low-cost fixed-point numbers (e.g., int8/int16).
Source: Adaptive Precision Training: Quantify Back Propagation in Neural Networks with Fixed-point Numbers
코드
관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.
https://github.com/NJU-Jet/SR_Mobile_Quantization
공식 코드https://github.com/NVlabs/Bi3D
공식 코드https://github.com/NVlabs/BiTTC
공식 코드https://github.com/NVlabs/condensa
공식 코드https://github.com/NervanaSystems/distiller
공식 코드Fast Adjustable Threshold For Uniform Neural Network Quantization (Winning solution of LPIRC-II)
https://github.com/NervanaSystems/nlp-architect/blob/master/nlp_architect/models/transformers/quantized_bert.py
공식 코드https://github.com/NeuroCompLab-psu/SNN-Conversion
공식 코드Exploring the Connection Between Binary and Spiking Neural Networks
https://github.com/NiCE-X/TSAA
공식 코드https://github.com/NoakLiu/LLMEasyQuant
공식 코드LLMEasyQuant: Scalable Quantization for Parallel and Distributed LLM Inference
https://github.com/OmarErak/CsiNet_Optimized
공식 코드Accelerating and Compressing Deep Neural Networks for Massive MIMO CSI Feedback
https://github.com/On-Point-RND/GIFT_SW
공식 코드GIFT-SW: Gaussian noise Injected Fine-Tuning of Salient Weights for LLMs
https://github.com/On-Point-RND/OpenSafetyMini-Investigating-the-Impact-of-Quantization-Methods-on-the-Safety-and-Reliability-of-LLM
공식 코드https://github.com/OwenSec/DeepDetector
공식 코드Detecting Adversarial Image Examples in Deep Networks with Adaptive Noise Reduction
https://github.com/Pachark/SVQ-Forecasting
공식 코드https://github.com/PaddlePaddle/PaddleClas
공식 코드https://github.com/QDucasse/nn_benchmark
공식 코드https://github.com/Qualcomm-AI-research/pruning-vs-quantization
공식 코드https://github.com/RICE-EIC/FracTrain
공식 코드https://github.com/RICE-EIC/ShiftAddNet
공식 코드https://github.com/Rayhane-mamah/Efficient-VDVAE
공식 코드https://github.com/RujingXiong/RIS_Optimization
공식 코드https://github.com/RunpeiDong/DGMS
공식 코드Finding the Task-Optimal Low-Bit Sub-Distribution in Deep Neural Networks
https://github.com/SAILResearch/wip-24-jaskirat-black-box-edge-operators
공식 코드On the Impact of Black-box Deployment Strategies for Edge AI on Latency and Model Performance
https://github.com/SITE5039/AgrLearn
공식 코드Aggregated Learning: A Vector-Quantization Approach to Learning Neural Network Classifiers
https://github.com/SaipingZhang/PeQuENet
공식 코드https://github.com/SalesforceAIResearch/uni2ts
공식 코드https://github.com/SebastianJanampa/DETRPose
공식 코드DETRPose: Real-time end-to-end transformer model for multi-person pose estimation
https://github.com/SebastianPeitz/QuaSiModO
공식 코드On the Universal Transformation of Data-Driven Models to Control Systems
https://github.com/SeyedMuhammadHosseinMousavi/Victoria-Amazonica-Optimization-VAO-Algorithm
공식 코드Victoria Amazonica Optimization (VAO): An Algorithm Inspired by the Giant Water Lily Plant
https://github.com/SforAiDl/KD_Lib
공식 코드KD-Lib: A PyTorch library for Knowledge Distillation, Pruning and Quantization
https://github.com/ShakyaJayakody/EdgeProfiler
공식 코드EdgeProfiler: A Fast Profiling Framework for Lightweight LLMs on Edge Using Analytical Model
https://github.com/Sheldon04/CABM-pytorch
공식 코드CABM: Content-Aware Bit Mapping for Single Image Super-Resolution Network with Large Input
https://github.com/SkoltechAI/fewbit
공식 코드Few-Bit Backward: Quantized Gradients of Activation Functions for Memory Footprint Reduction
https://github.com/Stability-AI/stable-codec
공식 코드Scaling Transformers for Low-Bitrate High-Quality Speech Coding
https://github.com/StefanSchwarzTUW/MultiStage-Grassmannian-DNN
공식 코드Recursive CSI Quantization of Time-Correlated MIMO Channels by Deep Learning Classification
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.