hurwitzlab/viral-learning

Name: viral-learning

Owner: Hurwitz Lab

Description: Deep learning demonstration code.

Created: 2017-11-29 16:56:13.0

Updated: 2017-12-13 17:48:48.0

Pushed: 2018-01-06 02:07:07.0

Homepage: null

Size: 11

Language: Jupyter Notebook

GitHub Committers

User	Most Recent Commit	# Commits

Other Committers

User	Email	Most Recent Commit	# Commits

README

viral-learning

Deep learning demonstration code.

Repository Setup

This repository contains Jupyter notebooks, which are not friendly to git. One way to help git handle these files is to set up jq to automatically strip out non-code elements from notebooks on commit.

The following recipe comes from http://timstaley.co.uk/posts/making-git-and-jupyter-notebooks-play-nice/.

First install jq.

Next add the following to ~/.gitconfig. (Note: it may be necessary to specify the full path to jq for tools such as SourceTree.)

e]
ibutesfile = ~/.gitattributes_global

ter "nbstrip_full"]
n = "jq --indent 1 \
    '(.cells[] | select(has(\"outputs\")) | .outputs) = []  \
    | (.cells[] | select(has(\"execution_count\")) | .execution_count) = null  \
    | .metadata = {\"language_info\": {\"name\": \"python\", \"pygments_lexer\": \"ipython3\"}} \
    | .cells[].metadata = {} \
    '"
ge = cat
ired = true

Finally add the following to ~/.gitattributes_global:

ynb filter=nbstrip_full

Demonstration data

The first_try notebook needs two data files:

• 'virus.fasta' https://www.ncbi.nlm.nih.gov/nuccore/NC_031261.1?report=fasta
• 'bacterium.fna.gz' ftp://ftp.ncbi.nlm.nih.gov/genomes/all/GCF/000/195/955/GCF_000195955.2_ASM19595v2/GCF_000195955.2_ASM19595v2_genomic.fna.gz

Install

The first_try notebook requires Jupyter, Keras, and TensorFlow. These can be installed in a virtual environment as follows:

thon3.6 -m venv ~/venv/vl
venv/vl/bin/activate
 $ pip install jupyter keras tensorflow

Run

Start the Jupyter notebook server and run the first_try notebook.

 $ jupyter notebook