Review for NeurIPS paper: Learning to Detect Objects with a 1 Megapixel Event Camera

Weaknesses: - I believe that the details given in this work are detailed enough to reproduce the experimental results in the paper. The used neural network layers/functions are well established and the training schedule looks clean. However, I would like to encourage the authors to publish their code. However all components are well chosen from previous published work. Examples include the representation of events based on [48, 22, 49, 23], keeping temporal state from [39] or the detector head from [37].