infinitone

I've used a bunch of these for object detection labeling.. the best one i found so far is labelmg (free) and rectlabel for Pascal VOC format.

It seems like a good idea in theory but i'm not so sure it'd work in many environments in practice. If i understand it correctly, you're storing all the state such as files but there is state that is tied to that specific machine (ie. machine fqdn, machine-specifc filepaths) and you wouldn't want to apply that state on another machine. I guess you could do some data wrangling and .stateignore that stuff but it would require quite the effort on a large application that spans many components and many teams.

On a very small app, i can see the utility of dotmesh.

There are many things produced that bring no value but pollute the world... take cigarettes for example.

Not if you need specialized vms like a p2.

I tried setting up a tagbox, image gets pulled fine, it starts fine and says 'box ready' but when i go to localhost:8080 - it doesn't load.

Running latest docker 17.12.0-ce on windows 10.

Yeah, i was confused too. Most engineers are actually backend: building APIs, distributed systems, large data compute... i would think these are the things people are doing when using 'high-demand technologies' such as AWS.

I'm still confused at what voice app is. And the link takes me a to a login page- there's you first problem in terms of discover-ability...

Another Esri HNer i see!

Yeah, i was confused about that... that whole 'closed system' section was much weaker than i thought. I thought somehow they were going to build a standardized app that can interface with all these closed apps- but then again, how would they do that to begin with?

I do... and i know a lot of others in tech that aren't as hyper to line up for a new phone. Also, so far, in my opinion- after the iphone 6- there hasn't been any major differences.

So this isn't OSS? Seems atypical in the ML community.

For those looking for alternative OSS solutions: BRAT, labellmg are decent.

The article starts with: > With GraphQL, the client speaks a request language which: 1) eliminates the need for the server to hardcode the shape or size of the data

And then ends with: > There are some mitigations we can do here. We can do cost analysis on the query in advance and enforce some kind of limits on the amount of data one can consume.

... so basically you did nothing in the way of perf but add complexity.

So... did anyone sub?

I never heard/read that quote by Einstein; interesting. Here is the full quote for the curious: http://www.goodreads.com/quotes/17014-the-release-of-atomic-...

You do realize it can become indefinite right? It is 90 days only if the countries share 'data' which the trump regime poorly defined. And its highly unlikely countries like Iran would accept that.

This feels like it can just be implemented by an Alexa skill? Would anyone be interested?

Heh.. i guess no one told you about LA or San Diego then.

This is pretty cool, love the visual graph- do you have a demo/sample code of the graph alone?

This is really interesting. How much data does it require to build a generative model with say 60% accuracy for story type data?

Its all about the over-ear mate.. over-ear..

I've survived a 4 story building fall when i was 3. Came out with no scratches or broken bones.

We use dashing at my workplace extensively. From my research into the space a year ago, dashing is the one with the largest library of plugins [0], and its super easy to make your own- here's mine [1].

[0] https://github.com/Shopify/dashing/wiki/Additional-Widgets

[1] https://github.com/aysark/dashing-timeline

Yes, in general what they refer to are things like the IRS Tax records (250 TB), Yahoo Ad data (900 TB). You just can't use a single machine to work with such data.

This is awesome! I've been meaning to do something like this for awhile.. great that someone else finally did it.

Any chance you'd open source it?

While this is great and serves as a good starter for those looking to get a basic understanding of computer vision.

I'm not so sure if thats what real self-driving cars use. It seems too basic to apply a Canny-Edge detector which is taught in 3rd/4th year comp vision classes. I'm skeptical but also not very knowledgeable about the tech that goes into self driving cars- so if anyone with real experience can inform would be great.

I've written a survey paper about various edge detection algorithms and while Canny is one of the better ones, it still leaves a lot to be desired.

Yup, agreed. I think overtime we'll also try to compensate for this by build a ML model from all the emails being sent to provide more 'fuzziness' to it.

You're not harsh at all- i'd appreciate feedback, what parts do you think we should improve?

I'll admit we kind of rushed the sign up page as we've been busy building the product.

Check out http://foxtype.com - does some of that but more grammar-like heuristics such as conciseness, complexity.

On a side note, I'm part of a team working on http://emailfox.co which will provide 'Smart Sentences' for you when composing an email, based on a recipient. Allowing you to write personal, relevant emails faster.

We're building a tool that does something similar but for email. Currently we're targeting cold sales emails- the idea is, you enter a recipient's email and we aggregate data about them and surface relevant, personable sentences that you can use in the email. You'll also be able to change the tone of these sentences (funny, professional, casual, etc.)

Learn more: http://emailfox.co

Is there something like this but for windows?