VisualVoteData

Visualize vote data from the Central Election Commission of Taiwan

Usage

1. Convert SHP files into Python pickles, if they are not already in shapes/.

   > cd collect_shapes
   > <python> collect_shapes.py 臺北市 南港區
   > <python> collect_shapes.py 臺北市 內湖區

2. Distill data from 選舉資料庫, if they are not already in data/.

   > cd distill_data
   > <python> distill_legislators.py 臺北市 南港區
   > <python> distill_legislators.py 臺北市 內湖區

3. Select what values to visualize with select_data.py or with something like Google Sheets. Intermediate CSV files will be generated in visualize/rgb/.

   > cd visualize
   > <python> select_data.py ../data/臺北市_南港區_立委第4選區_投開票所.csv --out=南港
   > <python> select_data.py ../data/臺北市_內湖區_立委第4選區_投開票所.csv --out=內湖

4. Export to images (TODO: interactive webpages, and KML files for Google Earth).

   > cd visualize
   > <python> export.py ../output/港湖 南港 內湖

Data sources

• 選舉資料庫

  • It seems that elbese.csv is for elections with electoral districts, while elbase.csv is for elections without electoral districts (ECODE == 0 for all rows).

    # 臺北市: province code: 63
    # 臺北市: county code: 0
    # 松山區: town code: 10
    # 全區合計: village code: '0000'
    
    # elbese.csv:
    >>> df[(df.PCODE==63) & (df.CCODE==0) & (df.TCODE==10) & (df.VCODE=='0000')]
          PCODE  CCODE  ECODE  TCODE VCODE NAME
    457      63      0      3     10  0000  松山區
    458      63      0      7     10  0000  松山區
    
    # elbase.csv:
    >>> df[(df.PCODE==63) & (df.CCODE==0) & (df.TCODE==10) & (df.VCODE=='0000')]
          PCODE  CCODE  ECODE  TCODE VCODE NAME
    1        63      0      0     10  0000  松山區

    These rows are only present in elbese.csv:

    >>> df[(df.ECODE!=0) & (df.TCODE==0) & (df.VCODE=='0000')]
          PCODE  CCODE  ECODE  TCODE VCODE      NAME
    472      63      0      1      0  0000  臺北市第01選區
    473      63      0      2      0  0000  臺北市第02選區
    474      63      0      3      0  0000  臺北市第03選區
    475      63      0      4      0  0000  臺北市第04選區
    476      63      0      5      0  0000  臺北市第05選區
    ...     ...    ...    ...    ...   ...       ...
    8263     10     17      1      0  0000  基隆市第01選區
    8264     10     18      1      0  0000  新竹市第01選區
    8265     10     20      1      0  0000  嘉義市第01選區
    8352      9      7      1      0  0000  連江縣第01選區
    8353      9     20      1      0  0000  金門縣第01選區
    
    [73 rows x 6 columns]

• 鄉鎮市區界線(TWD97經緯度)

• 村里界歷史圖資(TWD97經緯度)

• 臺北市鄰界圖

  • Converted the TM2 (2-degree zone transverse Mercator) coordinates to longitudes and latitudes with ShpTrans 2.2 by 范成棟 (2006.03.15)

Font

• Noto Sans Traditional Chinese

Compatibility

• Only tested with Python 3.11.9 on Windows 10 22H2, but should be compatible with most Python 3 distributions.