Jay Baker jaybaker

Neural Network Training Recipe

I loved this "recipe" from Andrej Karpathy. Read the whole thing at the source. I summarized the key points.

A brief review of the last week or so can be a useful habit.

Add some regimen to make this weekly.

That review may include items such as these.

	"""
	This gist is intended to be loaded into a python notebook via wget.
	It is used to retrieve a secret from the google cloud.

	See https://cloud.google.com/secret-manager/docs/reference/libraries#client-libraries-install-python

	Usage:
	1. !wget
	2. !pip install google-cloud-secret-manager
	3. import load_secret

	# concat files but just keep header from first file
	awk '
	FNR==1 && NR!=1 { while (/^header/) getline; }
	1 {print}
	' file*.txt >all.txt

	import os

	GB1 = 102410241024 # 1GB
	size = 50 # desired size in GB
	with open('large_file', 'wb') as fout:
	for i in range(size + 1):
	fout.write(os.urandom(GB1))

	# from http://www.rackspace.com/knowledge_center/article/generate-a-csr-with-openssl

	# create key
	openssl genrsa -out ~/domain.com.ssl/domain.com.key 2048

	# create csr
	openssl req -new -sha256 -key ~/domain.com.ssl/domain.com.key -out ~/domain.com.ssl/domain.com.csr

	# Taken from http://flask.pocoo.org/snippets/57/
	from flask import Flask
	app = Flask(__name__)

	@app.route('/', defaults={'path': ''})
	@app.route('/<path:path>')
	def catch_all(path):
	return 'You want path: %s' % path

	if __name__ == '__main__':