Transmute AI Lab

Official implementation of "Data-Efficient Training of CNNs and Transformers with Coresets: A Stability Perspective"

Transmute AI Lab Model Efficiency Toolkit

This is the official repository for the ICASSP 2023 paper "On Designing Light-Weight Object Trackers Through Network Pruning: Use CNNs or Transformers?"

Official implementation "ChipNet: Budget-Aware Pruning with Heaviside Continuous Approximations"

Official code for our CVPR'22 paper “Dynamic Kernel Selection for Improved Generalization and Memory Efficiency in Meta-learning”

Official Repository related to UltraMNIST classification benchmark

Official Implementation "Rescaling CNN through Learnable Repetition of Network Parameters"