<html>
  <head>
    <meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <p>I think I saw it in the Deep Learning book: 
      <a class="moz-txt-link-freetext" href="http://www.deeplearningbook.org/">http://www.deeplearningbook.org/</a><br>
    </p>
    Bill<br>
    <br>
    <div class="moz-cite-prefix">On 3/28/17 9:48 AM, Henrique C. S.
      Junior wrote:<br>
    </div>
    <blockquote
cite="mid:CAEeMmB=+wWLedcOJWu8yCPLbieyuuoDwROA3RYcb1C9B3xmL8Q@mail.gmail.com"
      type="cite">
      <div dir="ltr">
        <div class="gmail_default"
          style="font-family:monospace,monospace">@Tommaso, this is
          something like Internal Coordinates[1], right?</div>
        <div class="gmail_default"
          style="font-family:monospace,monospace">@Bill, thanks for the
          hint, I'll definitely take a look at this.</div>
        <div class="gmail_default"
          style="font-family:monospace,monospace"><br>
        </div>
        <div class="gmail_default"
          style="font-family:monospace,monospace">[1] - <a
            moz-do-not-send="true"
            href="https://en.wikipedia.org/wiki/Z-matrix_%28chemistry%29">https://en.wikipedia.org/wiki/Z-matrix_(chemistry)</a></div>
      </div>
      <div class="gmail_extra"><br>
        <div class="gmail_quote">On Tue, Mar 28, 2017 at 2:12 AM, Bill
          Ross <span dir="ltr"><<a moz-do-not-send="true"
              href="mailto:ross@cgl.ucsf.edu" target="_blank">ross@cgl.ucsf.edu</a>></span>
          wrote:<br>
          <blockquote class="gmail_quote" style="margin:0 0 0
            .8ex;border-left:1px #ccc solid;padding-left:1ex">
            <div bgcolor="#FFFFFF" text="#000000">
              <p>Image processing deals with xy coordinates by (as I
                understand) training with multiple permutations of the
                raw data, in the form of translations and rotations in
                the 2d space. If training with 3d data, there would be
                that much more translating and rotating to do, in order
                to divorce the learning from the incidentals.</p>
              <span class="HOEnZb"><font color="#888888">
                  <p>Bill<br>
                  </p>
                </font></span>
              <div>
                <div class="h5"> <br>
                  <div class="m_-7569063688226978064moz-cite-prefix">On
                    3/27/17 4:35 PM, Tommaso Costanzo wrote:<br>
                  </div>
                  <blockquote type="cite">
                    <div dir="ltr">
                      <div>
                        <div>
                          <div>
                            <div>
                              <div>
                                <div>
                                  <div>
                                    <div>
                                      <div>Dear Henrique,<br>
                                      </div>
                                      I am sorry for the poor email I
                                      wrote before. What I was saying is
                                      simply the fact that if you are
                                      trying to use the coordinates as
                                      "features" from an .xyz file then
                                      by machine learning you will learn
                                      at wich coordinate certain atoms
                                      will occur so you can only make
                                      prediction on the coordinate.
                                      However, if I correctly
                                      understood, the "features"
                                      representing the coupling J are
                                      distance, angle, and electron
                                      number. Definitely this properties
                                      can be derived from the XYZ file
                                      format from simple geometric
                                      calculations and the number of
                                      electrons will depend from the
                                      type of atom. So, what I was
                                      trying to say is that instead of
                                      using the XYZ file as input for
                                      scikit-learn, I was suggesting to
                                      do the calculation of angle,
                                      distances, electrons' number in
                                      advance (with other software(s) or
                                      directly in python)  and use the
                                      new calculated matrix as input for
                                      scikit-learn. In this case the
                                      machine will learn how J(AB)
                                      varies as a function of angle,
                                      distance, number of electrons. <br>
                                    </div>
                                    For example <br>
                                  </div>
                                  <br>
                                  distance     angle   n el.<br>
                                  1                  90      1<br>
                                  1                  90      1<br>
                                  2                  90      1<br>
                                  ....                ...        ...<br>
                                  <br>
                                </div>
                                If you are using a supervised learning
                                you will have to add a 4th column ( in
                                reality a separate column vector) with
                                your J(AB) on which you can train your
                                model and then predict the unknown
                                samples<br>
                                <br>
                              </div>
                              For example <br>
                              distance     angle   n el.    J(AB)<br>
                              1                  90      1        1<br>
                              1                  90      1        1<br>
                              2                  90      1         0.5<br>
                              ....                ...        ...      
                              ...<br>
                              <br>
                            </div>
                            <div>Now if you train the model on the
                              second matrix, and then you try to predict
                              the first one you should expect a results
                              like:<br>
                              <br>
                              1<br>
                              1<br>
                              0.5<br>
                              <br>
                            </div>
                            Of course in this case the "features" are
                            perfectly equal, hence the example is
                            completely unrealistic. However, I hope that
                            it will help to understand what I was
                            explaining in the previous email.<br>
                            If you want you can directly contact me at
                            this email, and I hope that you got
                            additional hints from Robert, that he seems
                            to be even more knowledgeable than me.<br>
                          </div>
                          <br>
                        </div>
                        Sincerely <br>
                      </div>
                      Tommaso<br>
                      <div>
                        <div><br>
                          <div><br>
                          </div>
                        </div>
                      </div>
                    </div>
                    <div class="gmail_extra"><br>
                      <div class="gmail_quote">2017-03-27 18:44
                        GMT-04:00 Henrique C. S. Junior <span dir="ltr"><<a
                            moz-do-not-send="true"
                            href="mailto:henriquecsj@gmail.com"
                            target="_blank">henriquecsj@gmail.com</a>></span>:<br>
                        <blockquote class="gmail_quote" style="margin:0
                          0 0 .8ex;border-left:1px #ccc
                          solid;padding-left:1ex">
                          <div dir="ltr">
                            <div class="gmail_default"
                              style="font-family:monospace,monospace">Dear
                              Tommaso, thank you for your kind reply.</div>
                            <div class="gmail_default"
                              style="font-family:monospace,monospace">I
                              know I have a lot to study before actually
                              starting any code and that's why any
                              suggestion is so valuable.</div>
                            <div class="gmail_default"
                              style="font-family:monospace,monospace">So,
                              you're suggesting that a simplification of
                              the system using only the paramagnetic
                              centers can be a good approach? (I'm not
                              sure if I understood it correctly).</div>
                            <div class="gmail_default"
                              style="font-family:monospace,monospace">My
                              main idea was, at first, try to represent
                              the systems as realistically as possible
                              (using coordinates). I know that the
                              software will not know what a bond is or
                              what an intermolecular interaction is but,
                              let's say, after including 1000s of
                              examples in the training, I was expecting
                              that (as an example) finding a C 0.000 and
                              an H at 1.000 should start to "make sense"
                              because it leads to an experimental trend.
                              And I totally agree that my way to
                              represent the system is not the better.</div>
                            <div class="gmail_default"
                              style="font-family:monospace,monospace"><br>
                            </div>
                            <div class="gmail_default"
                              style="font-family:monospace,monospace">Thank
                              you so much for all the help.</div>
                          </div>
                          <div class="m_-7569063688226978064HOEnZb">
                            <div class="m_-7569063688226978064h5">
                              <div class="gmail_extra"><br>
                                <div class="gmail_quote">On Mon, Mar 27,
                                  2017 at 4:15 PM, Tommaso Costanzo <span
                                    dir="ltr"><<a
                                      moz-do-not-send="true"
                                      href="mailto:tommaso.costanzo01@gmail.com"
                                      target="_blank">tommaso.costanzo01@gmail.com</a>></span>
                                  wrote:<br>
                                  <blockquote class="gmail_quote"
                                    style="margin:0 0 0
                                    .8ex;border-left:1px #ccc
                                    solid;padding-left:1ex">
                                    <div dir="ltr">
                                      <p
                                        style="margin:0px;text-indent:0px"><span
style="font-family:"ubuntu";font-size:12pt">Dear Henrique,</span></p>
                                      <p
style="margin:0px;text-indent:0px;font-family:"ubuntu";font-size:12pt"><br>
                                      </p>
                                      <p
                                        style="margin:0px;text-indent:0px"><span
style="font-family:"ubuntu";font-size:12pt">I agree with
                                          Robert on the use of a
                                          supervised algorithm and I
                                          would also suggest you to try
                                          a semisupervised one if you
                                          have trouble in labeling your
                                          data. </span></p>
                                      <p
style="margin:0px;text-indent:0px;font-family:"ubuntu";font-size:12pt"><br>
                                      </p>
                                      <p
                                        style="margin:0px;text-indent:0px"><span
style="font-family:"ubuntu";font-size:12pt">Moreover, as a
                                          chemist I think that the input
                                          you are thinking to use is not
                                          the in the best form for
                                          machine learning because you
                                          are trying to predict coupling
                                          J(AB) but in the future space
                                          you have only coordinates
                                          (XYZ). What I suggest is to
                                          generate the pair of atoms
                                          externally and then use a
                                          matrix of the form (Mx3),
                                          where M are the pairs of atoms
                                          you want to predict your J and
                                          3 are the features of the two
                                          atoms (distance, angle,
                                          unpaired electrons). For a
                                          supervised approach you will
                                          need a training set where the
                                          J is know so your training
                                          data will be of the form Mx4
                                          and the fourth feature will be
                                          the J you know.</span></p>
                                      <p
                                        style="margin:0px;text-indent:0px"><span
style="font-family:"ubuntu";font-size:12pt">Hope that this is
                                          clear, if not I will be happy
                                          to help more</span></p>
                                      <p
style="margin:0px;text-indent:0px;font-family:"ubuntu";font-size:12pt"><br>
                                      </p>
                                      <p
                                        style="margin:0px;text-indent:0px"><span
style="font-family:"ubuntu";font-size:12pt">Sincerely</span></p>
                                      <p
                                        style="margin:0px;text-indent:0px"><span
style="font-family:"ubuntu";font-size:12pt">Tommaso</span></p>
                                    </div>
                                    <div class="gmail_extra">
                                      <div>
                                        <div
                                          class="m_-7569063688226978064m_-419284271361902240h5"><br>
                                          <div class="gmail_quote">2017-03-27
                                            13:46 GMT-04:00 Henrique C.
                                            S. Junior <span dir="ltr"><<a
                                                moz-do-not-send="true"
                                                href="mailto:henriquecsj@gmail.com"
                                                target="_blank">henriquecsj@gmail.com</a>></span>:<br>
                                            <blockquote
                                              class="gmail_quote"
                                              style="margin:0 0 0
                                              .8ex;border-left:1px #ccc
                                              solid;padding-left:1ex">
                                              <div dir="ltr">
                                                <div
                                                  class="gmail_default"
style="font-family:monospace,monospace">Dear Robert, thank you. Yes, I'd
                                                  like to talk about
                                                  some specifics on the
                                                  project.</div>
                                                <div
                                                  class="gmail_default"
style="font-family:monospace,monospace">Thank you again.</div>
                                              </div>
                                              <div
class="m_-7569063688226978064m_-419284271361902240m_-8383123951498439579HOEnZb">
                                                <div
class="m_-7569063688226978064m_-419284271361902240m_-8383123951498439579h5">
                                                  <div
                                                    class="gmail_extra"><br>
                                                    <div
                                                      class="gmail_quote">On
                                                      Mon, Mar 27, 2017
                                                      at 2:25 PM, Robert
                                                      Slater <span
                                                        dir="ltr"><<a
moz-do-not-send="true" href="mailto:rdslater@gmail.com" target="_blank">rdslater@gmail.com</a>></span>
                                                      wrote:<br>
                                                      <blockquote
                                                        class="gmail_quote"
                                                        style="margin:0
                                                        0 0
                                                        .8ex;border-left:1px
                                                        #ccc
                                                        solid;padding-left:1ex">
                                                        <div dir="ltr">You
                                                          definitely can
                                                          use some of
                                                          the tools in
                                                          sci-kit learn
                                                          for supervised
                                                          machine
                                                          learning.  The
                                                          real trick
                                                          will be how
                                                          well your
                                                          training
                                                          system is
                                                          representative
                                                          of your future
                                                          predictions. 
                                                          All of the
                                                          various
                                                          regression
                                                          algorithms
                                                          would be of
                                                          some value and
                                                          you make even
                                                          consider an
                                                          ensemble to
                                                          help
                                                          generalize. 
                                                          There will be
                                                          some important
                                                          questions to
                                                          answer--what
                                                          kind of loss
                                                          function do
                                                          you want to
                                                          look at?  I
                                                          assumed
                                                          regression
                                                          (continuous
                                                          response) but
                                                          it could also
classify--paramagnetic, diamagnetic, ferromagnetic, etc...
                                                          <div><br>
                                                          </div>
                                                          <div>Another
                                                          task to think
                                                          about might be
                                                          dimension
                                                          reduction.</div>
                                                          <div>There is
                                                          no guarantee
                                                          you will get
                                                          fantastic
                                                          results--every
                                                          problem is
                                                          unique and
                                                          much will
                                                          depend on
                                                          exactly what
                                                          you want out
                                                          of the
                                                          solution--it
                                                          may be that we
                                                          get '10%'
                                                          accuracy at
                                                          best--for some
                                                          systems that
                                                          is quite good,
                                                          others it is
                                                          horrible.<br>
                                                          </div>
                                                          <div><br>
                                                          </div>
                                                          <div>If you'd
                                                          like to talk
                                                          specifics,
                                                          feel free to
                                                          contact me at
                                                          this email.  I
                                                          have a
                                                          background in
                                                          magnetism (PhD
                                                          in magnetic
                                                          multilayers--i
                                                          was physics,
                                                          but as you are
                                                          probably aware
                                                          chemisty and
                                                          physics blend
                                                          in this area)
                                                          and have a
                                                          fairly good
                                                          knowledge of
                                                          sci-kit learn
                                                          and machine
                                                          learning.  </div>
                                                          <div><br>
                                                          </div>
                                                          <div><br>
                                                          </div>
                                                        </div>
                                                        <div
                                                          class="gmail_extra"><br>
                                                          <div
                                                          class="gmail_quote">
                                                          <div>
                                                          <div
class="m_-7569063688226978064m_-419284271361902240m_-8383123951498439579m_6033336047822367828h5">On
                                                          Mon, Mar 27,
                                                          2017 at 10:50
                                                          AM, Henrique
                                                          C. S. Junior <span
                                                          dir="ltr"><<a
moz-do-not-send="true" href="mailto:henriquecsj@gmail.com"
                                                          target="_blank">henriquecsj@gmail.com</a>></span>
                                                          wrote:<br>
                                                          </div>
                                                          </div>
                                                          <blockquote
                                                          class="gmail_quote"
style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
                                                          <div>
                                                          <div
class="m_-7569063688226978064m_-419284271361902240m_-8383123951498439579m_6033336047822367828h5">
                                                          <div dir="ltr">
                                                          <div
                                                          class="gmail_default"
style="font-family:monospace,monospace">
                                                          <p
                                                          style="margin:0cm
                                                          0cm
12pt;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><span
style="color:rgb(36,39,41)" lang="EN-US">I'm a chemist with some
                                                          rudimentary
                                                          programming
                                                          skills
                                                          (getting
                                                          started with
                                                          python) and in
                                                          the middle of
                                                          the year I'll
                                                          be starting a
                                                          Ph.D. project
                                                          that uses
                                                          computers to
                                                          describe
                                                          magnetism in
                                                          molecular
                                                          systems.<span></span></span></p>
                                                          <p
                                                          style="margin:0cm
                                                          0cm
12pt;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><span
style="color:rgb(36,39,41)" lang="EN-US">Most of the time I get my
                                                          results after
                                                          several
                                                          simulations
                                                          and
                                                          experiments,
                                                          so, I know
                                                          that one of
                                                          the hardest
                                                          tasks in
                                                          molecular
                                                          magnetism is
                                                          to predict the
                                                          nature of
                                                          magnetic
                                                          interactions.
                                                          That's why
                                                          I'll try to
                                                          tackle this
                                                          problem with
                                                          Machine
                                                          Learning
                                                          (because such
                                                          interactions
                                                          are dependent,
                                                          basically, of
                                                          distances,
                                                          angles and
                                                          number of
                                                          unpaired
                                                          electrons).
                                                          The idea is to
                                                          feed the
                                                          computer with
                                                          a large
                                                          training set
                                                          (with number
                                                          of unpaired
                                                          electrons, XYZ
                                                          coordinates of
                                                          each molecule
                                                          and
                                                          experimental
                                                          magnetic
                                                          couplings) and
                                                          see if it can
                                                          predict the
                                                          magnetic
                                                          couplings
                                                          (J(AB)) of new
                                                          systems:<span></span></span></p>
                                                          </div>
                                                          <div>
                                                          <div
                                                          class="gmail_default"
style="font-family:monospace,monospace">(see example in the attached
                                                          image)</div>
                                                          <div
                                                          class="gmail_default"
style="font-family:monospace,monospace"><br>
                                                          </div>
                                                          <div
                                                          class="gmail_default"
style="font-family:monospace,monospace">Can Scikit-Learn handle the
                                                          task, knowing
                                                          that the
                                                          matrix used to
                                                          represent
                                                          atomic
                                                          coordinates
                                                          will probably
                                                          have a
                                                          different
                                                          number of
                                                          atoms (because
                                                          some molecules
                                                          have more
                                                          atoms than
                                                          others)? Or is
                                                          this a job
                                                          better suited
                                                          for another
                                                          software/approach?
                                                          ​</div>
                                                          <span
class="m_-7569063688226978064m_-419284271361902240m_-8383123951498439579m_6033336047822367828m_-1717598575983325084HOEnZb"><font
color="#888888"><br>
                                                          </font></span></div>
                                                          <span
class="m_-7569063688226978064m_-419284271361902240m_-8383123951498439579m_6033336047822367828m_-1717598575983325084HOEnZb"><font
color="#888888">
                                                          <div><br>
                                                          </div>
                                                          -- <br>
                                                          <div
class="m_-7569063688226978064m_-419284271361902240m_-8383123951498439579m_6033336047822367828m_-1717598575983325084m_-4201444065020757644gmail_signature">
                                                          <div dir="ltr">
                                                          <div>
                                                          <div dir="ltr">
                                                          <div>
                                                          <div dir="ltr">
                                                          <div>
                                                          <div dir="ltr">
                                                          <div>
                                                          <div dir="ltr">
                                                          <div>
                                                          <div dir="ltr">
                                                          <div>
                                                          <div dir="ltr">
                                                          <div>
                                                          <div dir="ltr"><span
style="color:rgb(139,139,139)"><font face="monospace, monospace"><b><font
color="#808080">Henrique C. S. Junior</font></b><br>
                                                          Industrial
                                                          Chemist -
                                                          UFRRJ</font></span></div>
                                                          <div dir="ltr"><span
style="color:rgb(139,139,139)"><font face="monospace, monospace">M. Sc.
                                                          Inorganic
                                                          Chemistry -
                                                          UFRRJ<br>
                                                          Data
                                                          Processing
                                                          Center - PMP</font><br>
                                                          </span></div>
                                                          </div>
                                                          <div><span
                                                          style="color:rgb(139,139,139)"><font
face="monospace, monospace">Visite o <a moz-do-not-send="true"
                                                          href="http://mundoquimico.com.br"
target="_blank">Mundo Químico</a></font></span></div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </font></span></div>
                                                          <br>
                                                          </div>
                                                          </div>
                                                          <span>______________________________<wbr>_________________<br>
                                                          scikit-learn
                                                          mailing list<br>
                                                          <a
                                                          moz-do-not-send="true"
href="mailto:scikit-learn@python.org" target="_blank">scikit-learn@python.org</a><br>
                                                          <a
                                                          moz-do-not-send="true"
href="https://mail.python.org/mailman/listinfo/scikit-learn"
                                                          rel="noreferrer"
target="_blank">https://mail.python.org/mailma<wbr>n/listinfo/scikit-learn</a><br>
                                                          <br>
                                                          </span></blockquote>
                                                          </div>
                                                          <br>
                                                        </div>
                                                        <br>
______________________________<wbr>_________________<br>
                                                        scikit-learn
                                                        mailing list<br>
                                                        <a
                                                          moz-do-not-send="true"
href="mailto:scikit-learn@python.org" target="_blank">scikit-learn@python.org</a><br>
                                                        <a
                                                          moz-do-not-send="true"
href="https://mail.python.org/mailman/listinfo/scikit-learn"
                                                          rel="noreferrer"
target="_blank">https://mail.python.org/mailma<wbr>n/listinfo/scikit-learn</a><br>
                                                        <br>
                                                      </blockquote>
                                                    </div>
                                                    <br>
                                                    <br clear="all">
                                                    <div><br>
                                                    </div>
                                                    -- <br>
                                                    <div
class="m_-7569063688226978064m_-419284271361902240m_-8383123951498439579m_6033336047822367828gmail_signature"
data-smartmail="gmail_signature">
                                                      <div dir="ltr">
                                                        <div>
                                                          <div dir="ltr">
                                                          <div>
                                                          <div dir="ltr">
                                                          <div>
                                                          <div dir="ltr">
                                                          <div>
                                                          <div dir="ltr">
                                                          <div>
                                                          <div dir="ltr">
                                                          <div>
                                                          <div dir="ltr">
                                                          <div>
                                                          <div dir="ltr"><span
style="color:rgb(139,139,139)"><font face="monospace, monospace"><b><font
color="#808080">Henrique C. S. Junior</font></b><br>
                                                          Industrial
                                                          Chemist -
                                                          UFRRJ</font></span></div>
                                                          <div dir="ltr"><span
style="color:rgb(139,139,139)"><font face="monospace, monospace">M. Sc.
                                                          Inorganic
                                                          Chemistry -
                                                          UFRRJ<br>
                                                          Data
                                                          Processing
                                                          Center - PMP</font><br>
                                                          </span></div>
                                                          </div>
                                                          <div><span
                                                          style="color:rgb(139,139,139)"><font
face="monospace, monospace">Visite o <a moz-do-not-send="true"
                                                          href="http://mundoquimico.com.br"
target="_blank">Mundo Químico</a></font></span></div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                          </div>
                                                        </div>
                                                      </div>
                                                    </div>
                                                  </div>
                                                </div>
                                              </div>
                                              <br>
______________________________<wbr>_________________<br>
                                              scikit-learn mailing list<br>
                                              <a moz-do-not-send="true"
href="mailto:scikit-learn@python.org" target="_blank">scikit-learn@python.org</a><br>
                                              <a moz-do-not-send="true"
href="https://mail.python.org/mailman/listinfo/scikit-learn"
                                                rel="noreferrer"
                                                target="_blank">https://mail.python.org/mailma<wbr>n/listinfo/scikit-learn</a><br>
                                              <br>
                                            </blockquote>
                                          </div>
                                          <br>
                                          <br clear="all">
                                          <br>
                                          -- <br>
                                        </div>
                                      </div>
                                      <div
class="m_-7569063688226978064m_-419284271361902240m_-8383123951498439579gmail_signature"
                                        data-smartmail="gmail_signature">
                                        <div dir="ltr"><span></span><span>Please
                                            do NOT send Microsoft Office
                                            Attachments:</span><br>
                                          <div> <a
                                              moz-do-not-send="true"
                                              href="http://www.gnu.org/philosophy/no-word-attachments.html"
                                              target="_blank">http://www.gnu.org/philosophy/<wbr>no-word-attachments.html</a></div>
                                        </div>
                                      </div>
                                    </div>
                                    <br>
                                    ______________________________<wbr>_________________<br>
                                    scikit-learn mailing list<br>
                                    <a moz-do-not-send="true"
                                      href="mailto:scikit-learn@python.org"
                                      target="_blank">scikit-learn@python.org</a><br>
                                    <a moz-do-not-send="true"
                                      href="https://mail.python.org/mailman/listinfo/scikit-learn"
                                      rel="noreferrer" target="_blank">https://mail.python.org/mailma<wbr>n/listinfo/scikit-learn</a><br>
                                    <br>
                                  </blockquote>
                                </div>
                                <br>
                                <br clear="all">
                                <div><br>
                                </div>
                                -- <br>
                                <div
                                  class="m_-7569063688226978064m_-419284271361902240gmail_signature"
                                  data-smartmail="gmail_signature">
                                  <div dir="ltr">
                                    <div>
                                      <div dir="ltr">
                                        <div>
                                          <div dir="ltr">
                                            <div>
                                              <div dir="ltr">
                                                <div>
                                                  <div dir="ltr">
                                                    <div>
                                                      <div dir="ltr">
                                                        <div>
                                                          <div dir="ltr">
                                                          <div>
                                                          <div dir="ltr"><span
style="color:rgb(139,139,139)"><font face="monospace, monospace"><b><font
color="#808080">Henrique C. S. Junior</font></b><br>
                                                          Industrial
                                                          Chemist -
                                                          UFRRJ</font></span></div>
                                                          <div dir="ltr"><span
style="color:rgb(139,139,139)"><font face="monospace, monospace">M. Sc.
                                                          Inorganic
                                                          Chemistry -
                                                          UFRRJ<br>
                                                          Data
                                                          Processing
                                                          Center - PMP</font><br>
                                                          </span></div>
                                                          </div>
                                                          <div><span
                                                          style="color:rgb(139,139,139)"><font
face="monospace, monospace">Visite o <a moz-do-not-send="true"
                                                          href="http://mundoquimico.com.br"
target="_blank">Mundo Químico</a></font></span></div>
                                                          </div>
                                                        </div>
                                                      </div>
                                                    </div>
                                                  </div>
                                                </div>
                                              </div>
                                            </div>
                                          </div>
                                        </div>
                                      </div>
                                    </div>
                                  </div>
                                </div>
                              </div>
                            </div>
                          </div>
                          <br>
                          ______________________________<wbr>_________________<br>
                          scikit-learn mailing list<br>
                          <a moz-do-not-send="true"
                            href="mailto:scikit-learn@python.org"
                            target="_blank">scikit-learn@python.org</a><br>
                          <a moz-do-not-send="true"
                            href="https://mail.python.org/mailman/listinfo/scikit-learn"
                            rel="noreferrer" target="_blank">https://mail.python.org/mailma<wbr>n/listinfo/scikit-learn</a><br>
                          <br>
                        </blockquote>
                      </div>
                      <br>
                      <br clear="all">
                      <br>
                      -- <br>
                      <div class="m_-7569063688226978064gmail_signature"
                        data-smartmail="gmail_signature">
                        <div dir="ltr"><span></span><span>Please do NOT
                            send Microsoft Office Attachments:</span><br>
                          <div> <a moz-do-not-send="true"
                              href="http://www.gnu.org/philosophy/no-word-attachments.html"
                              target="_blank">http://www.gnu.org/philosophy/<wbr>no-word-attachments.html</a></div>
                        </div>
                      </div>
                    </div>
                    <br>
                    <fieldset
                      class="m_-7569063688226978064mimeAttachmentHeader"></fieldset>
                    <br>
                    <pre>______________________________<wbr>_________________
scikit-learn mailing list
<a moz-do-not-send="true" class="m_-7569063688226978064moz-txt-link-abbreviated" href="mailto:scikit-learn@python.org" target="_blank">scikit-learn@python.org</a>
<a moz-do-not-send="true" class="m_-7569063688226978064moz-txt-link-freetext" href="https://mail.python.org/mailman/listinfo/scikit-learn" target="_blank">https://mail.python.org/<wbr>mailman/listinfo/scikit-learn</a>
</pre>
    </blockquote>
    

  </div></div></div>


______________________________<wbr>_________________

scikit-learn mailing list

<a moz-do-not-send="true" href="mailto:scikit-learn@python.org">scikit-learn@python.org</a>

<a moz-do-not-send="true" href="https://mail.python.org/mailman/listinfo/scikit-learn" rel="noreferrer" target="_blank">https://mail.python.org/<wbr>mailman/listinfo/scikit-learn</a>


</blockquote></div>

<div>
</div>-- 
<div class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><span style="color:rgb(139,139,139)"><font face="monospace, monospace"><b><font color="#808080">Henrique C. S. Junior</font></b>
Industrial Chemist - UFRRJ</font></span></div><div dir="ltr"><span style="color:rgb(139,139,139)"><font face="monospace, monospace">M. Sc. Inorganic Chemistry - UFRRJ
Data Processing Center - PMP</font>
</span></div></div><div><span style="color:rgb(139,139,139)"><font face="monospace, monospace">Visite o <a moz-do-not-send="true" href="http://mundoquimico.com.br" target="_blank">Mundo Químico</a></font></span></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div>
</div>


<fieldset class="mimeAttachmentHeader"></fieldset>
<pre wrap="">_______________________________________________
scikit-learn mailing list
<a class="moz-txt-link-abbreviated" href="mailto:scikit-learn@python.org">scikit-learn@python.org</a>
<a class="moz-txt-link-freetext" href="https://mail.python.org/mailman/listinfo/scikit-learn">https://mail.python.org/mailman/listinfo/scikit-learn</a>
</pre>

</blockquote>
</body></html>