Search Results for author:
Found 2 papers, 2 papers with code
Data Generation for Hardware-Friendly Post-Training Quantization
We analyze existing data generation methods based on batch normalization (BN) matching and identify several gaps between synthetic and real data: 1) Current generation algorithms do not optimize the entire synthetic dataset simultaneously; 2) Data augmentations applied during training are often overlooked; and 3) A distribution shift occurs in the final model layers due to the absence of BN in those layers.
HPTQ: Hardware-Friendly Post Training Quantization
Neural network quantization enables the deployment of models on edge devices.
Ranked #1 on
Quantization
on MS COCO
Cannot find the paper you are looking for? You can
Submit a new open access paper.
