https://software.intel.com/pt-br/forums/topic/434694/feed
pt-brQuote:Chi-Hung W. wrote:also
https://software.intel.com/pt-br/comment/1749394#comment-1749394
<a id="comment-1749394"></a>
<div class="field field-name-comment-body field-type-text-long field-label-hidden"><div class="field-items"><div class="field-item even" property="content:encoded"><p><strong>Quote:</strong></p><blockquote><em>Chi-Hung W.</em> wrote:also the left/right eigenvectors are always not in accordance</blockquote>Again, a simple statement such as that is not enough to go on. Note also that if <strong>v</strong> is an eigenvector, so is <em>c</em><strong>v</strong>, where <em>c</em> is a complex constant different from zero. Different packages may use different conventions for choosing the constant.
</div></div></div>Tue, 27 Aug 2013 16:26:11 +0000Imecej4comment 1749394 at https://software.intel.comsorry for the wrong results..
https://software.intel.com/pt-br/comment/1749386#comment-1749386
<a id="comment-1749386"></a>
<div class="field field-name-comment-body field-type-text-long field-label-hidden"><div class="field-items"><div class="field-item even" property="content:encoded"><p>sorry for the wrong results...the real part of the eigenvalue was not what I had written here. I am new to mathematica. </p>
<p>I discovered that if I set <br /><em> N[Eigenvalues[mat],14]</em><br />in mathematica, it would yield the same eigenvalues as what I have got using fortran+zgeev<br />though, if I set<br /> <em>N[ Eigenvalues[mat] ]</em><br />in mathematica then it is somehow different. therefore the precision of the eigenvalues seems not converged when I use fortran..?</p>
<p>also the left/right eigenvectors are always not in accordance...</p>
</div></div></div>Tue, 27 Aug 2013 16:06:45 +0000chwengcomment 1749386 at https://software.intel.comIf you ran Mathematica on the
https://software.intel.com/pt-br/comment/1749324#comment-1749324
<a id="comment-1749324"></a>
<div class="field field-name-comment-body field-type-text-long field-label-hidden"><div class="field-items"><div class="field-item even" property="content:encoded"><p>If you ran Mathematica on the 3 X 3 matrix that you gave above, you would not have obtained the eigenvalues as 2, -1 and -1. If the matrix that you gave Mathematica was the same matrix but with the diagonal elements replaced by zero, your comparison is not valid. Furthermore, your example does not agree with your original statement that "The obtained eigenvalues from the subroutine zgeev are precise".</p>
<p>Please make valid comparisons and report what you did accurately</p>
</div></div></div>Tue, 27 Aug 2013 13:16:22 +0000Imecej4comment 1749324 at https://software.intel.comfor more explicitly, I am
https://software.intel.com/pt-br/comment/1749296#comment-1749296
<a id="comment-1749296"></a>
<div class="field field-name-comment-body field-type-text-long field-label-hidden"><div class="field-items"><div class="field-item even" property="content:encoded"><p>for more explicitly, I am doing this as a check, it is a 3X3 matrix. and there is a complex number only at matrix(1,1)</p>
<p>matrix = {{ -6.75490294261523187*(10^-4) + (i *(-10^-3)), 1, 1 }, <br /> { 1, 0, 1 }, <br /> { 1, 1, 6.75490294261523187*(10^-4) }}</p>
<p>The matrix is non-symmetric. therefore I call zgeev for solving this. and it yields the following eigenvalues(here I output the real part of it only):<br /> 2.0000000273480896 <br /> -1.0003672906735881 <br /> -0.99963273667450081 </p>
<p>However, in mathematica, it is (real part of the eigenvalue)<br /> 2<br /> -1<br /> -1<br />I am wondering that maybe due to quasi-degeneracy, the eigenvalues are not precise enough (also the eigenvectors). Should I find ways to increase the precision or, try other subroutines for dealing degeneracy eigenvalues?</p>
</div></div></div>Tue, 27 Aug 2013 10:35:43 +0000chwengcomment 1749296 at https://software.intel.comPlease present details,
https://software.intel.com/pt-br/comment/1749100#comment-1749100
<a id="comment-1749100"></a>
<div class="field field-name-comment-body field-type-text-long field-label-hidden"><div class="field-items"><div class="field-item even" property="content:encoded"><p>Please present details, preferably using a short example, to support your claim. </p>
</div></div></div>Mon, 26 Aug 2013 15:50:40 +0000Imecej4comment 1749100 at https://software.intel.com