Re: [Clojure Book WIP] Deep Learning for Programmers: an Interactive Tutorial with CUDA, OpenCL, MKL-DNN, Java and Clojure, New Release 0.4.0

2019-06-18 Thread Duncan McGreggor
Hey! This looks *nice* ... d On Tue, Jun 18, 2019 at 4:49 AM Dragan Djuric wrote: > basically… > >1. the only DL book for programmers >2. interactive & dynamic >3. step-by-step implementation >4. incredible speed >5. yet, No C++ hell (!) >6. Nvidia GPU (CUDA and cuDNN) >

Re: [ANN] metasoarous/oz 1.5.1 - clojure dataviz updates (jupyter notebooks, markdown extensions, and static export, oh my!)

2019-01-24 Thread Duncan McGreggor
Yeah, I've been using CL+Maxima+Jupyter for my maths, etc.; will be really nice to start doing some of that in Clojure ... thanks! d On Wed, Jan 23, 2019 at 11:37 AM Christopher Small wrote: > Thank you both for the feedback! I'm really happy to be able to empower > our budding Clojure Data Sci

Re: Who Uses import-vars?

2017-11-07 Thread Duncan McGreggor
On 7 November 2017 at 13:11, Timothy Baldridge wrote: > I structure my code very explicitly. Normally the most common constructs > are put in a single file named after the library itself (not in core.clj, > do that half your files will be named core). > > https://github.com/halgari/odin/blob/mast

Re: Who Uses import-vars?

2017-11-07 Thread Duncan McGreggor
I use it in some large fraction of my Clojure projects (25-50%?). The primary driver is to balance an "internal" organization (e.g., that allows me to keep file line counts low) that would be cumbersome as an API vs. an ideal "developer experience" (e.g., less hierarchy and easier access to functio

Re: How to try/catch Let bindings?

2017-10-02 Thread Duncan McGreggor
Didier, I've done something similar a few times just using core.async -- no extra deps required ;-) d On 1 October 2017 at 23:21, Didier wrote: > I can't emphasize enough the utility of the interceptor chain pattern, as >> employed heavily in pedestal. >> > > Interesting... Its almost like a w

Clojure/conj hotel reservations

2017-09-18 Thread Duncan McGreggor
Hey all, If anyone else is registering late for conj, the website is showing rooms as sold out, but there are still a few available -- you have to call the hotel. Let them now the discount is under "Cognitect, Inc." I as of a few minutes ago there about 8 rooms still available ... See you there!

Re: [ANN] Eastwood, the Clojure lint tool, version 0.2.4 released

2017-05-31 Thread Duncan McGreggor
I use both all the time (part of the gating in the build process); they are nicely complementary. d On 23 May 2017 at 11:16, Travis Daudelin wrote: > Hi, thanks for posting this looks great! > > Is there any overlap in functionality between Eastwood and Kibit >

Re: Monorepos with separate artifacts?

2017-05-01 Thread Duncan McGreggor
The NASA project I'm working on right now does something similar, though with lein: * https://github.com/nasa/Common-Metadata-Repository The project was started a few years back, before the Components library (among others) were ready for use, so you'll see some bespoke bits in there. Similarly,

Re: Startup time of standalone .jar executable

2017-03-21 Thread Duncan McGreggor
Has the application owner defined an interface for long-running components? If not, maybe convince them to. Then they could just (.start yourCode) and then make whatever API calls you have defined, with zero start-up costs per-call from their side. (This is what I do when integrating non-Clojure .

Re: defpromise?

2012-11-20 Thread Duncan McGreggor
On Tue, Nov 20, 2012 at 9:12 AM, Jim foo.bar wrote: > I was just wondering...we have a 'defsomething' macro for almost > everything... > > How come there is no 'defpromise'? It is very easy to write one and since > we always initialize promises like this: (def x (promise)), I don't see any > reas

Re: defpromise?

2012-11-20 Thread Duncan McGreggor
On Tuesday, November 20, 2012 9:13:03 AM UTC-8, Jim foo.bar wrote: > > I was just wondering...we have a 'defsomething' macro for almost > everything... > > How come there is no 'defpromise'? It is very easy to write one and > since we always initialize promises like this: (def x (promise)), I