hughperkins

Suggest passing the cost and the openai api calls through to us. eg we provide an openai api key. Just make it very easy for us to stay in control.

If your target audience is techie, you could maybe provide us a short python proxy script, that we run on our laptop, and that calls openai for you. Your website then connects to this somehow.

This would reduce load on your server, improve lag for us, and avoid cost issues for you.

And that's bad because?

Just set it to equal the current number of h1b slots

Note that looks like you are assuming 4% investments after retirement. Per the passive investment chart, this implies a fairly high level of risk? I'm unclear if you're also assuming we will live forever?

also: broadband. It originally contrasted with narrowband, and they describe modes for propagation of light through fiberglass cables... I thought it was really weird when I heard people on mainstream radio discussing such things. Eventually I realized they were using broadband as a synonym for 'fast internet' :P

^^^ This. Why is this not in fact the case?

A strange game. The only winning move is not to play.

Correlation != causation.

Seems perfectly plausible to me that the more confident candidates are more likely to be poached by other companies, for example.

Looks like 'bowling' is quite a challenge, for now...

> believe that question was cleared up once they responded to me that they were going to see their boyfriend/girlfriend.

This sentence is kind of interesting. So, you were 'just making small talk', until they told you they had a boy/girlfriend, and then, you just ... stopped 'making smalltalk'? :-P

Since when did TDD fail? Which is not to say it needs to be applied systematically to everything. But there are often bits of code which are better off being correct, and TDD works well for those.

If one looks at it from a 'what a today's compiler can do', then sure, one needs to statically declare everything.

If one looks at things from a 'any technical task can be handled by machine learning, sooner or later' point of view, there seems to be no obvious reason why parallelization, which is a purely technical task, not like say, writing music, could not sooner or later be handled by machine learning algorithms?

I've used oracle. Its awesome. Expensive. But awesome. If I have to pay, postgres is OK. But oracle is really really good. If they were laptop operating systems, postgres would be Debian, oracle would be Mac os x.

Is a 100% filled in linked in profile something to strive for? It seems like the poster has chosen an arbitrary goal, then seeks to manipulate the goal so he can finish it. Why not just choose a different goal?

This is excellent news! Guidelines to follow implies that if the manufacturers can meet these guidelines then they could plausibly have a somewhat legal basis for putting self driving cars on the roads.

If you are living on expenses, so can/must eat out every night, you quickly notice that empty restaurants have shocking food; and conversely the ones that are full, and have lines, have the most awesome food/atmosphere.

Whenever i see a bunch of code written in a language different fron whatever the rest of tge team are using, that language is often: go :)

I vote personal project. Its meritocratic. It involves approximately similar skills to being successful in a technical role. It contributes something new and novel to society. Its probably a better predictor than fizzbuzz, or knowing some technology buzzwords...

According to the comment just above yours, those "signatures" are unauthenticated?

Wrong way around: only allow http:// and https:// (and generally filtering out anything thats not letters, numbers, slash or dot is probably a good idea. Remove any sequences of more than one slash or dot.

Its no longer representative in fact. Tobacco usage within China has dropped radically over the last few years.

Ok. Fixed this by two things:

- using branch r0.10, as suggested by https://news.ycombinator.com/item?id=12464835 - making sure to install the new r0.10 wheel, which has a different name than the r0 wheel built by master :-D

Still doesnt work for me though: even on a new box, I get:

    ubuntu@somewhere:~/tensorflow$ python3 -c 'import tensorflow'             
    Traceback (most recent call last):
      File "<string>", line 1, in <module>
      File "/home/ubuntu/tensorflow/tensorflow/__init__.py", line 23, in <module>
        from tensorflow.python import *
      File "/home/ubuntu/tensorflow/tensorflow/python/__init__.py", line 49, in <mod
    ule>
        from tensorflow.python import pywrap_tensorflow
    ImportError: cannot import name 'pywrap_tensorflow'

Ah, I see what I was missing before: there's a patch 1 of cuda 8.0, which adds gcc 5.4 support, and is installed in the cuda section using:

    sudo sh cuda_8.0.27.1_linux.run --silent --accept-eula

Thats not really true. I dont use facebook (i use wechat, never had anything censored, and the privacy controls are awesome), but i dont think facebook wakes up in the morning and thinks "why doesnt Hugh use facebook? We should make it more like wechat". If you want change, the most effective way to get it is to ask for it.

I dont think hes expecting other people to contribute code to his project whilst hes working on it.

This seems to be a frankenstein of cuda 7.5 instructions and cuda 8.0. Similarly ubuntu 14.04 and 16.04. As far as i know, these instructions will fail from gcc 5.4 errors, amongst other issues.

Havent tried on arm. Dont think Ive noticed anyone trying on arm. Not quite sure what would work/not work.

To be fair, my OpenCL distro follows the same approach http://github.com/hughperkins/distro-cl Mainstream torch is kind of like 'sid' debian: it's cutting edge, but not always as stable as one would like, if one wants to tweak it a bit, eg add OpenCL.

As far as the title of this news thread, note that the 'self contained torch folder' title of the intel distro comes from the original distro that Soumith made https://github.com/torch/distro It's not something that Intel wrote/added/assert.

There's a thread at https://github.com/torch/torch7/pull/287 of people who have managed to install on Windows, and a wiki page at https://github.com/torch/torch7/wiki/Windows , which is looking pretty comprehensive :-)