Skip to content

Latest commit

 

History

History
168 lines (136 loc) · 4.2 KB

README.md

File metadata and controls

168 lines (136 loc) · 4.2 KB
Raw

mysoftware

This is the software repository for our paper solving problem x in a novel way.

Repository layout (what do I find where?)

|- container   # OCI recipes
|- mymodule    # source code for this software module
|- images
|- notes       # Project organization, paper LaTex, related work, proposal, meeting notes
|- scripts     # processing scripts, standalone CLI

Motivation

Short description of

  • problem in field (e.g. cancer)
  • what is current solution (handdrawn segmentation)
  • what this software does to solve it (sota segmentation)
  • proposed impact (accelerated/accurate diagnosis) Similar to abstract of the paper

Status

CircleCI License: AGPL v3

Table of contents

  1. Installation
  2. Test
  3. Usage
  4. Cite
  5. Reproducing results
    1. Datasets
    2. Preprocessing
  6. FAQ
  7. REMOVE ME ON TEMPLATE

Installation

Local

git clone <thisrepo>
# build instructions
Pip
pip install mysoftware
Conda
conda install mysoftware

or using the environment

conda env create --name myproject --file=myproject.yml

Container

Singularity
Building from recipes
sudo singularity build container/singularityrecipe.def
Released images
singularity pull repo:image.version

Test

To run the tests ensuring the installation is valid, please run

python -m mymodule.testmycode

Which should result in

Usage

Example snippets

Example inference on in silico data

import seaborn as sns
import mysoftware.preprocessing as stk
import numpy as np
np.random.seed(42)
silico_data = stk.generate_insilico(args)
q = stk.infer(silico_data)
sns.boxplot(x='method', y='accuracy', data=q)

Notebooks

See notebooks/example.ipynb for an example interactive workflow.

API

See docs for the complete API documentation

Cite

@article{you2023,
  title={yourtitle},
  author={authors et al},
  year={2023},
  publisher={venue}
}

Reproducing our published results

Data

In silico

See [mysoftware/insilico] on how to generate our in silico datasets. For your convenience, this can be downloaded as well:

mkdir insilico && cd insilico
wget -O insilico.hdf5 https://mydatarepo/insilico.hdf5

Real world

mkdir mydata
wget -O dataset.hdf5 https://mydatarepo/dataset.hdf5

Ensure the checksums match

md5sum dataset.hdf5

should produce

4a4f224c7b7c871855fd307ae323be93 dataset.hdf5

Pre/post processing code

See scripts/preprocessing.py for the preprocessing scripts that configure the dataset and generate the plots

FAQ

Help I can't figure out how function5 works

Please create a new issue detailing concisely, yet complete what issue you encountered, in a reproducible way.

FOR STUDENTS -- REMOVE ME

This a template for you to reuse in releasing software accompanying papers. At a minimum, ensure your software runs in a reproducible way. Place your (Python) code in [mymodule/mycode.py] (or different submodules). Processing scripts ideally are separate from the modules

Instructions on how to create packages/containers:

PIP

Conda

Singularity

Travis

CircleCI

CodeCov