happen2me

# Install broadcom wireless drivers
sudo pacman -S linux-headers base-devel broadcom-wl-dkms

# Install yay
sudo pacman -S --needed git base-devel
git clone https://aur.archlinux.org/yay-bin.git
cd yay-bin
makepkg -si

# Install rclone chrome, zotero, slack, etc. (xdg-desktop-portal is for automatic dark theme switch)

happen2me

Deploy an End-to-End Chinese Knowledge-based Dialogue System in a MLOps fashion on AWS

Knowledge graph database structures data as nodes and vertex. It provides flexibility and allows diversity and heterogeneity in real-world data. However, managing and querying such databases requires professionality in understanding graph query language and the graph database itself. Luckily, natural language processing (NLP) technics can help both in creating graph databases and understanding users' queries.

In this post, we will build a system heavily relied on NLP that can extract information from unstructured texts and interpret users' natural language queries as graph query languages. Moreover, we will develop this system in an MLOps manner such that it can automatically update itself to cope with changes in data and schema.

Walkthrough of this post:

• Part I Algorithm: Introduce the model used for knowledge extraction and natural question understanding.
• Part II Deployment: An overview of how the train

happen2me

1. 概念

   Project：一个标注项目，里面有类型相同的标注任务。

   Task：一个标注任务。有自己的标签集合，包含一定数量的待标注帧。可以属于一个项目，并且会继承该项目的标签集合。

   Job: 一个job是一个任务里的一个标注作业。Job的存在是为了对标注帧很多的Task进一步细分，如果Task有设置segment size，Task里会包含多个Jobs，每个Job含指定数量的帧。

2. 筛选名字包含wattman的图像，单独创建一个job

happen2me

# Return a list of vertices
def get_vertices(context):
    selected_object = context.active_object
    if selected_object is None:
        raise Exception('No object selected: please select the object to deform')
    return selected_object.data.vertices

# Returns a list of triangles of vertex indices (you need to perform simple triangulation) 
def get_faces(context):	
    selected_object = context.active_object