samim samim23

Instructions for making a Neural-Style movie

The following instructions are for creating your own animations using the style transfer technique described by Gatys, Ecker, and Bethge, and implemented by Justin Johnson. To see an example of such an animation, see this video of Alice in Wonderland re-styled by 17 paintings.

Setting up the environment

The easiest way to set up the environment is to simply load Samim's a pre-built Terminal.com snap or use another cloud service like Amazon EC2. Unfortunately the g2.2xlarge GPU instances cost $0.99 per hour, and depending on parameters selected, it may take 10-15 minutes to produce a 512px-wide image, so it can cost $2-3 to generate 1 sec of video at 12fps.

If you do load the

Asymco: Hollywood by the numbers. IN-DEPTH
The Critical Path 20: Below the (belt)line. Explores Hollywood financials.
The Critical Path 21: Negative Costs. Dan and Horace take on the way income from movies is distributed and why no movies ever make money.
The Critical Path 71: Bingewatch.
The Critical Path 80: Functional Structure. What is a functional organization and why is that a thing of beauty? What do Pixar and Apple have in common? EXCELLENT
Horace hypothesizes that apps will become the new distribution mechanism for entertainment. INSIGHTFUL

In one Critical Path episode (can't remember which), Horace posits that Hollywood is like a distributed functional organization:

Workers are organized by specialty (acting,

Extracting financial disclosure reports and police blotter narratives using OpenAI's Structured Output

tl;dr this demo shows how to call OpenAI's gpt-4o-mini model, provide it with URL of a screenshot of a document, and extract data that follows a schema you define. The results are pretty solid even with little effort in defining the data — and no effort doing data prep. OpenAI's API could be a cost-efficient tool for large scale data gathering projects involving public documents.

OpenAI announced Structured Outputs for its API, a feature that allows users to specify the fields and schema of extracted data, and guarantees that the JSON output will follow that specification.

For example, given a Congressional financial disclosure report, with assets defined in a table like this:

	#!/usr/bin/env python
	# -- coding: utf-8 --
	# vim: ai ts=4 sts=4 et sw=4


	"""
	Tools to extract feed links, test if they are valid and parse them
	with feedparser, returning content or a proper error.
	"""

	# Alec Radford, Indico, Kyle Kastner
	# License: MIT
	"""
	Convolutional VAE in a single file.
	Bringing in code from IndicoDataSolutions and Alec Radford (NewMu)
	Additionally converted to use default conv2d interface instead of explicit cuDNN
	"""
	import theano
	import theano.tensor as T
	from theano.compat.python2x import OrderedDict

	# -- coding: utf-8 --
	"""
	Reproducing the results of Auto-Encoding Variational Bayes by Kingma and Welling
	With a little help from the code from van Amersfoort and Otto Fabius (https://github.com/y0ast)

	@author: Pedro Tabacof (tabacof at gmail dot com)
	"""

	import random
	import numpy as np

	Start a g2.2xlarge or better (GPU instance) with https://console.aws.amazon.com/ec2/v2/home?region=us-east-1#LaunchInstanceWizard:ami=ami-ffba7b94

	Login, username is ubuntu

	Update a bunch of stuff and make sure cudnn R2 is used:

	luarocks install image
	luarocks install loadcaffe
	luarocks install torch
	export LD_LIBRARY_PATH=/home/ubuntu/torch-distro/install/lib:/home/ubuntu/torch-distro/install/lib:/home/ubuntu/cudnn-6.5-linux-x64-v2-rc2

	# A reduction of the post by Yoav Goldberg to a script
	# https://gist.github.com/yoavg/d76121dfde2618422139
	# Author: Kyle Kastner
	# License: BSD 3-Clause

	# Fun alternate settings
	# Download kjv.txt from http://www.ccel.org/ccel/bible/kjv.txt
	# python markov_lm.py kjv.txt 5 1.
	# Snippet:
	# Queen ording found Raguel: I kill.