explorer

README

Explorer

Generic data explorer, designed for viewing hundreds of thousands of images, pre-computed metrics, and multiple OCR transcriptions all on a single dashboard. Adapted from the Old Bailey and OCR explorer, which is associated with https://github.com/ughe/old-bailey.

Data

The data/ directory contains the following structure:

├── config.csv
├── imgs
│   └── .gitkeep
├── txts
│   └── .gitkeep
└── results.csv

The data/ directory is essentially an empty scaffolding in this repo, since it can be interchanged to swap the contents of the viewer. For example, see https://obo.cs.princeton.edu/mini/data for a deployment of this repo with a specialized data dir. The config.csv and results.csv must be set up correctly, along with the actual images in imgs and the different text transcriptions in subfolders within txts.

config.csv contains the site title, image extension, list of directories in data/txts/, and optionally: navbar links & metric min/max ranges. Here is an example config.csv:

title,Sample Title
imgs-fmt,jpg
txts-dirs,FirstDirName;SecondDirName;EtcDirName
[]links,Button1;https://google.com
[]links,Button2;data/
[]range,ExampleMetricSubstring;Min;Max
[]range,Time;0;12000

Note that the semicolon-separated list of txts-dirs are expected to be dir names inside data/txts/ and each dir should contain every single image pointer with .txt on the end.

results.csv contains a row of image pointers (image names without file extension) and then a row for each pre-computed metric. For example:

ptr,Image1,Image2, ...
Time (Millis),300,400, ...

Get Started

Initialize the data/ directory and then serve from the explorer's top-level dir:

python3 -m http.server

open http://127.0.0.1:8000

Documentation

Index

• Variables

Variables

var Grid []byte

var Index []byte

var Main []byte

var Style []byte