most citedWhere Is My Training Bottleneck? Hidden Trade-Offs in Deep Learning Preprocessing Pipelines

31 citations · 31 across the 1 of their papers we have counts for

collaborators

1 paper

cs.LG2022★ 31 cited

Where Is My Training Bottleneck? Hidden Trade-Offs in Deep Learning Preprocessing Pipelines

Alexander Isenko, Ruben Mayer, Jeffrey Jedele +1

Preprocessing pipelines in deep learning aim to provide sufficient data throughput to keep the training processes busy. Maximizing resource utilization is becoming more challenging…