| Re: [eigen] Reshape function |
[ Thread Index |
Date Index
| More lists.tuxfamily.org/eigen Archives
]
- To: eigen@xxxxxxxxxxxxxxxxxxx
- Subject: Re: [eigen] Reshape function
- From: Gael Guennebaud <gael.guennebaud@xxxxxxxxx>
- Date: Sat, 21 Feb 2009 09:42:35 +0100
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :date:message-id:subject:from:to:content-type :content-transfer-encoding; bh=UPgDDcQ2YXWJEehuLHvby28Tmv09jbbGgSrVIxdlugY=; b=t0998FFajQJLfFu0kQjM6E5LLrJ4JrQo6LhF2imxGHGv6AddpSb2yF3+MYoaGz9mCW VH0+j5X76zS7q5XGgp1z7HqNbJoVwjw6DNQu7DRrM+YTJdvqQVFrIHbCXOMIt1Qy+HWb HDonsrNWbtS6AQ288cwTaTsNil514U2qTN1qY=
- Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type:content-transfer-encoding; b=fv4xyR7gBLrP/m2cphIdWYYpMTYb3l+MKi1240hibPe3c3Z10pVdJUSInk3y3oXSmH MdBPqgYeBkxmJ0oUA9f25pNLtdnKVNUKAnb7JhDDVLKYAflSBrNbT8Xlt/bnBpIrL0fe AWubOF9v7J80qoBfE7d/nL2efM2ycTOX3rD6Q=
yes this has been discussed a couple of time on the forum, and this
seems reasonable to me.
On Sat, Feb 21, 2009 at 7:27 AM, Keir Mierle <mierle@xxxxxxxxx> wrote:
> It would be nice if Eigen had a reshape function like numpy and matlab:
>
> In [15]: x = random.rand(12)
>
> In [16]: x
> Out[16]:
> array([ 0.98782169, 0.24576577, 0.43955455, 0.54305616, 0.00266062,
> 0.48190312, 0.55144744, 0.36710248, 0.04064906, 0.34286723,
> 0.13652416, 0.01933417])
>
> In [17]: reshape(x, (3, 4))
> Out[17]:
> array([[ 0.98782169, 0.24576577, 0.43955455, 0.54305616],
> [ 0.00266062, 0.48190312, 0.55144744, 0.36710248],
> [ 0.04064906, 0.34286723, 0.13652416, 0.01933417]])
>
> In [18]: reshape(x, (6, 2))
> Out[18]:
> array([[ 0.98782169, 0.24576577],
> [ 0.43955455, 0.54305616],
> [ 0.00266062, 0.48190312],
> [ 0.55144744, 0.36710248],
> [ 0.04064906, 0.34286723],
> [ 0.13652416, 0.01933417]])
>
> I mail because some of my old libmv1 code used this, and I am porting it to
> libmv2. I use this a fair amount in numpy. Here's what I propose:
>
> Matrix<...> mymatrix(12, 1);
> mymatrix.reshape(3, 4); // In place.
>
> This should do the same thing as resize, but does not allow changing the
> number of elements, and does not change the contents. I'm not sure what the
> semantics should be around column vs row major; I suspect it's best to not
> do anything fancy (i.e. reshape() with differing row/col major will give
> different results).
>
> Keir
>
>