- 博客(14)
- 资源 (5)
- 收藏
- 关注
原创 Autodim论文
大部分的RecSys都会将稀疏特征映射成连续的表征向量,即在onehot- encoding后加入embedding层。在推荐系统中,特征的embedding是非常重要的部分,记录了各个特征所包含的信息,对之后的特征交叉和训练有至关重要的作用。一个直观的直觉是,embedding向量的维度与其所能代表的信息成比例关系。因此,各个特征域的特征使用的embedding向量的维度应该有所区别,文章提出了一种可以端到端训练的框架,自动为不同域赋予embedding维度,取得了SOTA的效果。 框架的主...
2022-05-25 10:52:39 320
原创 colab引用已写好的代码文件
from google.colab import drivedrive.mount('/content/drive')#挂载google dreveimport syssys.path.append('/content/drive/My Drive/Colab Notebooks')
2022-04-07 15:27:43 312
原创 HTGC连接和使用(for Mac)
终端输入连接htgcssh #eid@htgc1.cs.cityu.edu.hk根据学校要求:Besides users' home directories, all nodes in the HTGC mount a 12TB shared NFS storage on path '/public'. Users can make their own folder there. Each user account has a default quota of 200GB disk space
2022-04-06 16:02:10 1060
原创 Review of paper“Parallelizing exploration-exploit tradeoffs in gaussian process bandit optimization“
As executing experiments simultaneously or in batches is often desirable in many experimental ecenarios, this paper proposed gaussian Process Batch Upper confidence Bound, an upper confidence bound-based algorithm, which models the reward function as a sam
2022-03-15 16:28:10 576
原创 Review of Papar“Dual Averaging Method for Regulized Stochastic Learning and Online Optimization“
RDA(regularized dual average) is a algorithm to efficiently solve regularized stochastic learning and regularized online optimization. This paper shows RDA's effectivity for sparse online learning with L1-regularization.For the traditional online algorit
2022-02-13 16:27:20 292
原创 Review Report of paper “Online Convex Programming and Generalized Infinitesimal Gradient Ascent“
Summary of the paperThe paper contains two main parts. Part A is online convex programming. The author describe a Greedy Projection algorithm and analyze the performance of it, proving that the average regr
2022-01-24 15:37:45 194
原创 Pytorch深度学习(1)
建立network的一个模板:from torch import nn, optimimport torch.nn.functional as Fclass network(nn.Module): def __init__(self): super().__init__(self): self.fc1 = nn.Linear(784, 256) self.fc2 = nn.Linear(256, 64) self.fc3 = nn.L
2022-01-16 14:32:36 1124
原创 使用Django制作的简陋数据库
提示:文章写完后,目录可以自动生成,如何生成可参考右边的帮助文档文章目录前言 一、如何在社区版Pycharm建立Django项目 二、模型的建立 1.ER图的绘制和数据库结构 2.在Django中建立数据库 3.管理网站的网站界面 4.视图和URL传输 总结前言数据库课程的大作业,需要编写一个完整数据库项目。因为没有前端经验,这里从零开始学习Django框架,时间有限,最终的结果十分简陋,仅作记录用一、如何在社区Pycharm建立Django项目
2022-01-15 11:31:43 744
原创 pytorch环境配置(for mac)
由于anaconda暂时不能完全适配mac M1芯片的原因,使用miniconda进行环境配置。参考GitHub项目完成安装:https://github.com/udacity/deep-learning-v2-pytorchminiconda主页下载安装包miniconda官网下载自行寻找合适的版本查看是否安装成功终端中输入:conda -v这里我出现了 zsh: command not found: conda 的结果,查询后进行了如下操作vim ~/.zshrc
2022-01-14 14:34:49 987
Neural Architecture Searching survey paper reading
2022-01-27
空空如也
TA创建的收藏夹 TA关注的收藏夹
TA关注的人