Create puias-computational.repo file:
$ vim /etc/yum.repos.d/puias-computational.repo
Paste the following code into the above file:
import urllib2 | |
from lxml import html | |
from collections import defaultdict | |
url = "http://www.youtube.com/watch?v=IyLZ6RXJ8Eg" | |
doc = html.parse(urllib2.urlopen(url)) | |
data = defaultdict(dict) | |
props = doc.xpath('//meta[re:test(@name|@property, "^twitter|og:.*$", "i")]', | |
namespaces={"re": "http://exslt.org/regular-expressions"}) |
A comparison of Theano with other deep learning frameworks, highlighting a series of low-level design choices in no particular order.
Overview
Symbolic: Theano, CGT; Automatic: Torch, MXNet
Symbolic and automatic differentiation are often confused or used interchangeably, although their implementations are significantly different.