Lu Chongkai

@pdmosses Sure. My repo site is [makecent/makecent.github.io](https://github.com/makecent/makecent.github.io) and the related page is [chongkai.site](https://chongkai.site/docs/collections/python_tips/).

@pdmosses How did you get the Python snippet highlighted by GitHub?, i.e., the below:  I never have successfully done that.

@pdmosses The dark mode is not the point, but the lighting style. 1. I marked my codes with `python`:  2. I am using the same theme...

@pdmosses But you have already succeeded. My "wanted results" is token from what you have shown before. Do you mean that it only works locally?

> @makecent it's only on GitHub **repos** (e.g., in issue comments such as this one) that I can produce the style that you want. I don't know how to get...

> @makecent from the page resources used when browsing a GitHub repos, it seems they use their own (not released) [Prettylights](https://github.com/primer/github-syntax-theme-generator) framework. The CSS for their light theme is [here](https://github.com/primer/github-syntax-light/blob/master/lib/github-light.css)....

Me too. According to the original paper, EfficientNet-B4 should have similar FLOPs with ResNet-50. While the practical used memory of the former is much larger than the latter. **And it's...

Thanks for the response. The regression network I mean here is a model that has the same architecture with your model but only the last layer is dropped, and use...

> Maybe he means it is extracted after training with ActivityNet-1.3 pretrained model? It will make sense then. The feature extractor (pretrained model) is **fixed** without further training.

@HAOCHENYE Thanks for the reply. My case is that I want to validate every 0.5 epoch because my dataset size is huge (I think such demand is not that strange)....