Skip to main content

Ownage of ESPI image inference

Project description

espiownage

Ownage (domination) of ESPI image inference. (Pronounced like "espionage" but with a little "own" in the middle.)

Welcome to the new phase of SPNet developement -- IN PROGRESS.
In this incarnation, we'll be making it an image-segmentation code instead of an object detector, and we'll use fast.ai.

Install

Preliminaries

Ubuntu (& probably other Linuxes):

sudo apt-get install python3-tk

Mac (with Homebrew)

brew install python-tk

Then on all systems, let's set up a virtual environment called espi. I like to put my environments in ~/envs:

mkdir ~/envs; python3 -m venv ~/envs/espi; source ~/envs/espi/bin/activate

And then you want/need to update pip in case it gave you an ancient version:

python3 -m pip install pip --upgrade

Then to install this package all you do is...

pip install espiownage

Note: the requirements on this package follow a "kitchen sink" approach so that everything a student might need gets installed, e.g. jupyter and more. (And wheel because it speeds up the installations...I think.)

How to use

If you're reading this, you probably have access to the "real" data, which sits (on my machine) in ~/Dropbox/Data/espiownage-data. So cd to that directory, i.e.,

$ cd ~/Dropbox/Data/espiownage-data

...(or whereever you've got it) for what follows.

AND THEN, so we don't "clobber" each other's work, make your own copy (~17MB) of the main annotations directory, as in append your last name (hawley, morrison, morgan, etc):

cp -r annotations annotations_yourlastname

and then we'll each edit our own copy just to avoid...confusion.

Note:If you don't have access to the real data, you can still grab the fake SPNet data and then, for each of those datasets: Move (or symlink) all the images to a directory called images/, and all the .csv files to a directory called annotations/, and proceed.

ellipse editor

ellipse_editor --files=annotations_yourlastname/*.csv

See ellipse_editor -h for command-line options. You can, for example, edit only one strike's worth of data by running

ellipse_editor --files=annotations_yourlastname/06241902*.csv

or a range of annotations, as in ellipse editor --files=annotations_yourlastname/06241902_proc_001*.csv

generate segmentation masks

...from annotations:

gen_masks --files=annotations/*.csv

Contributing / Development

You'll want to install more things:

pip install nbdev twine 

Fork this repo. When you want to update your repo, one macro does it all (see Makefile):

make git_update

Asides

Handy tips for students

I can never remember how to start up virtual environments / or I don't want to remember. So in my ~/.bashrc file (you may have a ~/.zshrc) I put in a line where I define an alias/function I call gimme, that reads like so:

gimme() { source ~/envs/"$1"/bin/activate;  }

(note that in order for this alias to be recognized, you need to either logout and log back in or else run $ source ~/.bashrc)

Then when I want to load environment like espi I just type...

gimme espi

--Scott H. Hawley, September 2021

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

espiownage-0.0.27.tar.gz (23.3 kB view details)

Uploaded Source

Built Distribution

espiownage-0.0.27-py3-none-any.whl (20.7 kB view details)

Uploaded Python 3

File details

Details for the file espiownage-0.0.27.tar.gz.

File metadata

  • Download URL: espiownage-0.0.27.tar.gz
  • Upload date:
  • Size: 23.3 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.4.2 importlib_metadata/4.8.1 pkginfo/1.7.1 requests/2.26.0 requests-toolbelt/0.9.1 tqdm/4.62.2 CPython/3.9.5

File hashes

Hashes for espiownage-0.0.27.tar.gz
Algorithm Hash digest
SHA256 2f83c853e0312bb1a30f1817efe8f33450e68c1ec5fdecb586b00991d2889060
MD5 f554f456be314a9475b4c5e2e86a7a9f
BLAKE2b-256 8a958829bfb2b6dcc9ad1142a5fe5589e8d74979ffa6d8cf0be633ed36fb1c2a

See more details on using hashes here.

File details

Details for the file espiownage-0.0.27-py3-none-any.whl.

File metadata

  • Download URL: espiownage-0.0.27-py3-none-any.whl
  • Upload date:
  • Size: 20.7 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.4.2 importlib_metadata/4.8.1 pkginfo/1.7.1 requests/2.26.0 requests-toolbelt/0.9.1 tqdm/4.62.2 CPython/3.9.5

File hashes

Hashes for espiownage-0.0.27-py3-none-any.whl
Algorithm Hash digest
SHA256 0f85661e668c68425e11fbd63e712a6a9c36d9080197ba100b9d868a9090ca81
MD5 e53841c705fab162de416df451385f83
BLAKE2b-256 30260ad61788ae9b630a3ec1d594929f66f1899b07d583474742a66f4db00710

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page