-
事情的起因是我打算把自动检测依赖项的功能整合进 CI 里,这样就不会因为加了宏包后忘记改脚本而编译失败了。然而 `njuthesis` 那边拿 `clist` 批量载入了几个包,不符合标准流程,自然都没有被识别到。
我目前对这种情况想不出怎么做自动化处理(当然也很有可能是自己写法太烂……),就直接把这几个包当作常量手动加进脚本里面了。
```tex
\cs_new_protected…
-
### 先决条件
- [X] 我已尝试更新[模板版本](https://github.com/nju-lug/NJUThesis/blob/master/CHANGELOG.md)
- [X] 我已检索[模板手册](http://mirrors.ctan.org/macros/unicodetex/latex/njuthesis/njuthesis.pdf)
- [X] 我已检索[项目 wiki…
-
This is just to keep track of potentially interesting candidates and their advantages and disadvantages.
- https://www.h2o.ai/
- https://github.com/JohnLangford/vowpal_wabbit (!!!)
- FACTORIE http…
-
## The Purpose of This Issue
We are consistently eager to learn more about Tai-e users, understand what led you to choose our framework, attentively address your requirements, and help promote your w…
-
概率论学的一般,对于偏差与方差的推导有以下疑问,希望大佬解答我的疑惑.
南瓜书2.5.1节,在推导公式 2.41 的3步 → 4步时, 写到“由于噪声和 f 无关,所以 f(x; D) 和 y_D 是两个相互独立的随机变量。”
疑问:D是随机抽取的数据集,y_D是数据集D中数据x的标签, f(x;D)是用数据集D训练的模型然后对y_D所对应的x的预测结果,也就是D的不同会同时导致训练的…
-
这两天在重新拜读关于封面部分的代码。`fduthesis` 使用了 `xtemplate` 进行此部分的实现,但在实际阅读代码的时候,我并没有感到抽象化起到了简化的作用,反而是大量的嵌套导致直观上艰深难懂(也有很大可能是因为我菜)。仔细阅读这一宏包的使用手册后我发现了一个华点,简言之就是:*`Instance` 不应当是对整个页面内容的描述,而是对单个页面部件的描述。*
原文是这样写的:
…
-
# 总结
- 官方文档:https://www.qemu.org/docs/master/
- https://github.com/NJU-ProjectN/nemu
- https://github.com/qemu/qemu
- qemu+riscv:https://zhuanlan.zhihu.com/p/258394849
- 可以模拟很多开发板:https://blog.cs…
cisen updated
2 years ago
-
#### Description
IsolationForest allows choosing weighted sampling of attributes when building isolation trees. Adding this part to the algorithm can help to avoid noise attributes or redundant featu…
-
I want to know what depth image and raw-depth image refer to. Is depth image a post-processing depth map? I did this operation (depths-min)/(max-min)*255 when I generated raw-depth from the depth data…
-
2016.1月1版第54页公式3.7的分子上x与y似乎反了。
![微信图片_20240713155008](https://github.com/user-attachments/assets/e0b75a70-d94b-4e28-a474-b111fc1398f4)