-
From: clojure@googlegroups.com [mailto:clojure@googlegroups.com] On Behalf Of
Huahai Yang
Sent: Thursday, June 30, 2011 12:33 AM
To: Clojure
Subject: Re: Translating Java code with nested for loops
Missed that one. Yeah, that's much nicer. -huahai
On Jun 29, 6:39 pm, Alan Malloy
Missed that one. Yeah, that's much nicer. -huahai
On Jun 29, 6:39 pm, Alan Malloy wrote:
> (* daily (reduce * 1 coll))
> ;; is exactly equal to
> (reduce * daily coll)
>
> My point is you can start with daily instead of with 1, and thus not
> have to special-case it in at the end.
>
> On Jun 29,
(* daily (reduce * 1 coll))
;; is exactly equal to
(reduce * daily coll)
My point is you can start with daily instead of with 1, and thus not
have to special-case it in at the end.
On Jun 29, 8:22 am, Huahai Yang wrote:
> The multiplication outside the reduce is required because it is part
> of
.
-Original Message-
From: clojure@googlegroups.com [mailto:clojure@googlegroups.com] On Behalf Of
Alan Malloy
Sent: Wednesday, June 29, 2011 4:50 AM
To: Clojure
Subject: Re: Translating Java code with nested for loops
On Jun 29, 12:55 am, Ken Wesson wrote:
> On Wed, Jun 29, 2011 at 3:53
The multiplication outside the reduce is required because it is part
of the original logic: "contrib[i] = sum * dailyValues[i];". Using the
drop function is a very good call though.
-huahai
On Jun 29, 12:53 am, Alan Malloy wrote:
> If you're already doing a bunch of multiplication in the reduce,
On Jun 29, 12:55 am, Ken Wesson wrote:
> On Wed, Jun 29, 2011 at 3:53 AM, Alan Malloy wrote:
> > This layout makes it fairly clear that the first element of total-
> > values is never being used, but it would be nice if we could say so
> > explicitly. So finally, we can wrap the whole thing in a
On Wed, Jun 29, 2011 at 3:53 AM, Alan Malloy wrote:
> This layout makes it fairly clear that the first element of total-
> values is never being used, but it would be nice if we could say so
> explicitly. So finally, we can wrap the whole thing in a let:
>
> (defn calc [total-values daily-values]
If you're already doing a bunch of multiplication in the reduce, you
don't need to do it outside as well, do you?
And because reduce doesn't care about laziness, you can use drop
instead of nthnext (which I think is clearer in almost all cases).
(defn calc [total-values daily-values]
(map-index
How about this one?
(defn calc [total-values daily-values]
(map-indexed
(fn [i daily]
(* daily
(reduce #(* %1 (inc (/ %2 100.0))) 1.0 (nthnext total-
values (inc i)
daily-values))
Happy coding.
-huahai
On Jun 28, 10:11 pm, Justin Kramer wrote:
> Here's one wa
On Jun 29, 2011, at 1:45 AM, David Sletten wrote:
>
> (defn compute-contrib [daily-values total-values]
> (loop [contrib []
> daily-values daily-values
> total-values total-values]
> (if (empty? daily-values)
> contrib
> (recur (conj contrib (* (first daily-val
On Jun 28, 2011, at 11:37 PM, David Sletten wrote:
>
> On Jun 28, 2011, at 8:20 PM, Bhinderwala, Shoeb wrote:
>> The inputs are two arrays of type double of the same length – dailyValues
>> and totalValues. The output is the array contrib of the same length.
>>
>>
>> int n = dailyValu
Here's one way.
(defn tails [coll]
(take-while seq (iterate rest coll)))
(defn calc [total-values daily-values]
(map * daily-values (for [tail (tails total-values)]
(reduce #(* %1 (inc (/ %2 100.0)))
1.0
(
On Jun 28, 2011, at 8:20 PM, Bhinderwala, Shoeb wrote:
> The inputs are two arrays of type double of the same length – dailyValues and
> totalValues. The output is the array contrib of the same length.
>
>
> int n = dailyValues.length;
>
> for (int i = 0; i < n; i++)
>
>
13 matches
Mail list logo
