Practical guide to hyperparameters search for deep learning models

Are you tired of babysitting your DL models? If so, you’re in the right place. In this post, we discuss motivations and strategies behind effectively searching for the best set of hyperparameters for any deep learning model. We’ll demonstrate how this can be done on FloydHub, as well as which direction the research is moving. When you’re done reading this post, you’ll have added some powerful new tools to your data science tool-belt – making the process of finding the best configuration for your deep learning task as automatic as possible. Unlike machine learning models, deep learning models are literally full of hyperparameters. Would you like some some evidence? Just take a look at the Transformer base v1 hyperparameters definition.I rest my case. Of course, not all of these variables contribute in the same way to the model’s learning process, but, given this additional complexity, it’s clear that finding the


Original URL: http://feedproxy.google.com/~r/feedsapi/BwPx/~3/q-b3S7v4FPQ/

Original article

Introduction to HAProxy Stick Tables

HTTP requests are stateless by design. However, this raises some questions regarding how to track user activities, including malicious ones, across requests so that you can collect metrics, block users, and make other decisions based on state. The only way to track user activities between one request and the next is to add a mechanism for storing events and categorizing them by client IP or other key.
Out of the box, HAProxy Enterprise Edition and HAProxy give you a fast, in-memory storage called stick tables. Originally, stick tables were created to solve the problem of server persistence. However, StackExchange, the network of Q&A communities that includes Stack Overflow, saw the potential to use them for rate limiting of abusive clients, aid in bot protection, and tracking data transferred on a per client basis. They sponsored further development of stick tables to expand the functionality. Today, stick tables are an incredibly powerful


Original URL: http://feedproxy.google.com/~r/feedsapi/BwPx/~3/mFEkQNL7xDw/

Original article

The Private Blockchain Fallacy

In all the blockchain hype, a simple, yet self-refuting idea
keeps popping up: a Private Blockchain.

In order to understand why a private blockchain is nonsense, we must
first define what a block chain is and what it is not. Since Nakamoto
coined (PDF) the term, lets see if
his description helps:

a peer-to-peer distributed timestamp server to generate computational
proof of the chronological order of transactions

It does highlight some important traits of “a block chain” (Nakamoto
used a space between both words, I use the current popular term
blockchain):

peer-to-peer: implying distribution; at least ruling out a central
authority
computational proof: implying it to be verifiable
timestamp-server/chronological ordering: it’s goal, but also implying
permanence
Marco Iansiti and Karim R Lakhani have a more accessible explanation:

an open, distributed ledger that can record transactions between two
parties efficiently and in a verifiable and permanent way.

open (often called permissionless)
distributed (often called decentralized)
verifiable
permanent (often called immutable)
In other words: anything that does not match those criteria, is by
definition,


Original URL: http://feedproxy.google.com/~r/feedsapi/BwPx/~3/R__JpTUr_VQ/

Original article

A Cat Meme Photographer from a Century Ago

Much as internet surfers — i.e. everyone in 2018 — enjoy sharing silly and picturesque JPEGs with one another that feature clever quips or inspirational sayings, Americans of a century ago passed around similar memes. They were called postcards, or souvenir cards, and mailing them to friends and relatives was immensely popular for sharing a gilded, snowy holiday scene or even a lolcat.
One hundred years before e-mail inboxes crowded with pictures of cats adorned with text like “I CAN HAS CHEEZBURGER?” and “CEILING CAT IS WATCHING YOU,” lolcats (and loldogs and lolrabbits) were already at the height of fancy. The rise of postcards at the turn of the century enabled Pennsylvanian Harry Whittier Frees to build a career out of photographing cute animals donning hats and britches.
Frees’s work was unique at the time because his shots seemed to display an unlikely cooperation from his furry talent in assuming human poses.


Original URL: http://feedproxy.google.com/~r/feedsapi/BwPx/~3/AzwV30PFGR0/cat-meme-photographer-century-ago.html

Original article

Walmart Is Putting 17,000 Oculus Go Headsets In Its Stores To Help Train Employees In VR

Walmart is reportedly planning to send Oculus Go headsets to each of its nearly 5,000 stores so that more of its employees can get instruction more often. TechCrunch reports: The big box giant will begin sending four headsets to each Walmart supercenter and two headsets to each Neighborhood Market in the country. That may not necessarily seem like a ton to train a store full of employees, but at Walmart’s scale that amounts to about 17,000 headsets being shipped by year’s end. The move is the evolution of an announcement that the company made last year that it was working with STRIVR Labs to bring virtual reality training to its 200 “Walmart Academy” training centers. Those training sessions were done on PC-tethered Oculus Rifts, the move to Oculus Go headsets really showcases how much more simple standalone headset hardware is to set up and operate.

Read more of this story at


Original URL: http://rss.slashdot.org/~r/Slashdot/slashdot/~3/KQnqUXG5Ru4/walmart-is-putting-17000-oculus-go-headsets-in-its-stores-to-help-train-employees-in-vr

Original article

Whatever Happened to the Semantic Web?

In 2001, Tim Berners-Lee, inventor of the World Wide Web, published an article
in Scientific American. Berners-Lee, along with two other researchers, Ora
Lassila and James Hendler, wanted to give the world a preview of the
revolutionary new changes they saw coming to the web. Since its introduction
only a decade before, the web had fast become the world’s best means for
sharing documents with other people. Now, the authors promised, the web would
evolve to encompass not just documents but every kind of data one could
imagine.

They called this new web the Semantic Web. The great promise of the Semantic
Web was that it would be readable not just by humans but also by machines.
Pages on the web would be meaningful to software programs—they would have
semantics—allowing programs to interact with the web the same way that people
do. Programs could exchange data across the Semantic Web without having to be
explicitly engineered to talk to each other. According to


Original URL: http://feedproxy.google.com/~r/feedsapi/BwPx/~3/Qyjo9wIDYBs/semantic-web.html

Original article

MariaDB acquires Clustrix

MariaDB, the company behind the eponymous MySQL drop-in replacement database, today announced that it has acquired Clustrix, which itself is a MySQL drop-in replacement database, but with a focus on scalability. MariaDB will integrate Clustrix’s technology into its own database, which will allow it to offer its users a more scalable database service in the long run.
That by itself would be an interesting development for the popular open source database company. But there’s another angle to this story, too. In addition to the acquisition, MariaDB also today announced that cloud computing company ServiceNow is investing in MariaDB, an investment that helped it get to today’s acquisition. doesn’t typically make investments, though it has made a few acquisitions. It is a very large MariaDB user, though, and it’s exactly the kind of customer that will benefit from the Clustrix acquisition.
MariaDB CEO Michael Howard tells me that ServiceNow current supports about 80,000 instances


Original URL: http://feedproxy.google.com/~r/feedsapi/BwPx/~3/drDI7njI9Ns/

Original article

Proudly powered by WordPress | Theme: Baskerville 2 by Anders Noren.

Up ↑

%d bloggers like this: