Skip to content

Latest commit

 

History

History
29 lines (22 loc) · 1.17 KB

readme.md

File metadata and controls

29 lines (22 loc) · 1.17 KB

Basic Annotator Tool

This is a simple annotator tool

Installation & Use

  1. conda create --name anna python=3.7
  2. pip install -r requirements.txt
  3. Go down a level cd anna
  4. Review config.yaml it will describe what you want to analyze.
    labels_config is a list of labels you want to assign.
    index_cols - the column used as index and target the column you absolutely need to annotate

Usage

  1. Place the input file into data/datasets make sure it matches some config available or add a new config. Note, the tool expects csv files and those will be read with pandas.read_csv()
  2. Run as python app.py. Annotations are saved to data/annotations
  3. Do not forget to restart the server if you add new files or new configs,

NB: Annotations use the column specified as index_cols if that column is not unique, annotation will be ambiguous

Alex the example

This is Alex, he is an example

  1. Notice how there can be a number of labels per row (defined in the config file)
  2. You can label and un-label as you like and feel
  3. If two lines have the same index, labels will be ambiguous