DynaQuant: Compressing Deep Learning Training Checkpoints via Dynamic Quantization

no code implementations • 20 Jun 2023 • Amey Agrawal, Sameer Reddy, Satwik Bhattamishra, Venkata Prabhakara Sarath Nookala, Vidushi Vashishth, Kexin Rong, Alexey Tumanov

With the increase in the scale of Deep Learning (DL) training workloads in terms of compute resources and time consumption, the likelihood of encountering in-training failures rises substantially, leading to lost work and resource wastage.

Model Compression Quantization +1