# Meeting Note 11/13
###### tags: `meeting note`

[TOC]

## TODO list
- github
- public or private
- 目錄架構
- python專案
- initialize module
- worker flow detail?
-
- simulation
- 不能建邊啊，2300萬個點再建要爆炸了
- 不能建，也不用建
- networkX
- 需要$O(1)$
- node to group list
- `group_list(u)`
- iterate elements in a given group list
- `group_members(g)`
- $p_{max,u}$ of a contact group $g$
- $u$ is a age group
- $p_{max,u}=\max\{p_{u,v}\mid v\in (S\cup F)\cap g\}$
- $N$ of a contact group
- $|(S\cup F)\cap g|$
- 機率的
- 所以可能要repeat幾次取平均?
- 不同community的contact probability不一樣

## 內容
- raw
- worker flow format
- $A$ to $B$ 工作人數的list (matrix)
- $A,B$是鄉鎮市區
- data processing
- use worker flow data(from [US](https://www.census.gov/data/tables/2015/demo/metro-micro/commuting-flows-2015.html)) to build **probability of commuting from A to B daily** where A and B are census tract.
- convert all necessary data to JSON or CSV format

- contact group
- period
- day only
- night only
- both
- age dependent
- 0-5
- elementary school
- ...
- 19-64
- 64-
- if $u$ and $v$ are 屬於多個共同community
- 就會判定多次
- 加乘效果，合理
- day night不同
- dynamic community(X
- 兩個static community
- 但幾乎一樣，只差在某些worker加入或離開

- data format
- 規定 input data
- Geographics
- census_tract.csv
- Demographics
- age_population.csv
- Worker Flow
- ratio_commute_worker.csv
- worker_flow.csv
- Contact Matrix
- contact_prob.csv
- COVID-19 configuration
- R0.csv [(Next-generation Matrix)](https://eehh-stanford.github.io/gceid/sens.html)
- age_dependent_mortality.csv
- latent_period.csv
- Infectious period.csv
- natural immunity
- initial_adopter.csv ($\{\{地區_0, 人數_0\}, \{地區_1, 人數_1\}, ... \}$)
- input data to init_module.data
- graph.txt
- init_module to simulation_module

## graph.txt 格式細節

- first line
- number of node $n$
- number of group class $m$
- number of age group $a$
- second line
- each group class $0\leq i<m$, the number of that group class $m_i$
- followed by $m$ blocks
- each block contains $a$ line, $a$ doubles each line
- $c_{i,j,m_i}$ indecates contact probability of contact group $m_i$ from age group $i$ to age group $j$
- the following $n$ line, for line $i$
- indicate information of node $i$
- first the age group it belongs
- then the number of day group
- and followed by group class $m_i$ and group id $0\leq idx_{m_i}<m_i$
- then the number of night group
- and followed by group class $m_i$ and group id $0\leq idx_{m_i}<m_i$

2 3 2
1 2 3
0.6 0.4
0.2 0.3
0.5 0.2
0.2 0.4
0.9 0.4
0.4 0.2
0 1 0 0 2 0 0 2 1
1 2 0 0 1 0 0

## 禮拜三前須完成事項
- 郭庭佑
- ML
- data format from init_module to simulation_module
- SI module
- naive?
- Sieve?
- 廖廷恩
- Data processing
- 沈家駿
- 廖晊程
- define data format to load in init_module
- process data in init_module and output graph.txt