specimen-tools

specimen-tools contains code related to the iOS application Specimen. This can be used to: build a SQLite database of Specimen data, query it, and includes additional utilities for simple things like plotting common color-related data.

In order to avoid additional spamming of the original Specimen authors, for access to the underlying Specimen data, please reach out to Jose Cambronero (jcamsan@mit.edu). Access to the data is decided on a case-by-case basis and there is no guarantee of approval. Applications related to research or open-source experimentation are encouraged.

Requirements

specimen-tools assumes you have access to:

• Python 2.*
• SQLite

Most *nix systems come with SQLite installed, and should satisfy this requirement with no additional work. If this is not the case, please see SQlite for the appropriate way to install on your system. On Linux, this should be as simple as sudo apt-get install sqlite3. On Mac OSX, with homebrew, the same command is sudo brew install sqlite3.

The necessary Python libraries are installed automatically.

Make sure that your python and pip are linked to the correct version. You can check this by calling

pip --version
python --version

in both cases, the Python major version should correspond to 2.7.

Installation and Basic Usage

Assuming you have satisfied the requirements above, in order to install specimen-tools, you can use pip

sudo pip install specimen-tools

Or you can copy this repository and build locally

git clone https://github.com/josepablocam/specimen-tools.git
cd specimen-tools/
pip install -e .

This installs the specimen module. Note that this does not build the database. In order to do so, you need to explicitly execute specimen-tools/scripts/build_db.py (which is only included in the repository). Additionally, you will need to have access to the raw Specimen data to build the database. The database built is a SQLite database, which means it is serverless and consists of a single file.

Acknowledgements

We thank Erica Gorochow, Salvatore Randazzo, and Charlie Whitney for providing access to the Specimen dataset and for building the Specimen application.