How to visual Fig. 4 in your paper

ma-xu / pointMLP-pytorch

[ICLR 2022 poster] Official PyTorch implementation of "Rethinking Network Design and Local Geometry in Point Cloud: A Simple Residual MLP Framework"

Apache License 2.0

467 stars 60 forks source link

How to visual Fig. 4 in your paper #83

Closed Elbenhorn closed 8 months ago

Elbenhorn commented 1 year ago

Thank you for your amazing work! Could you share the code of drawing Fig.4 (Loss landscape along two rand directions.)

ma-xu commented 1 year ago

I will send it to you by the end of May (or early June). Sry that I'm really busy recently.

LiuHahahaha commented 10 months ago

I also have the same request. Can you share the code?

ma-xu commented 10 months ago

@Elbenhorn @LiuHahahaha Sorry about the late reply.

We firstly get the raw data by lossland.py, then we plot the figure by plot.py. Note that both scripts are not cleaned andn you may need to change some settiings accordingly (e.g., checkpoint paths, ranges, etc.). More details can be found here: Li, Hao, et al. "Visualizing the loss landscape of neural nets." NeurIPS 2018.

Also in issue: https://github.com/ma-xu/pointMLP-pytorch/issues/42

Elbenhorn commented 8 months ago

Very appreciate it!