<?xml version="1.0" encoding="UTF-8"?>
<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns="http://purl.org/rss/1.0/" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel rdf:about="https://community.wolfram.com">
    <title>Community RSS Feed</title>
    <link>https://community.wolfram.com</link>
    <description>RSS Feed for Wolfram Community showing questions tagged with Chemistry sorted by active.</description>
    <items>
      <rdf:Seq>
        <rdf:li rdf:resource="https://community.wolfram.com/groups/-/m/t/3595346" />
        <rdf:li rdf:resource="https://community.wolfram.com/groups/-/m/t/3550739" />
        <rdf:li rdf:resource="https://community.wolfram.com/groups/-/m/t/3537445" />
        <rdf:li rdf:resource="https://community.wolfram.com/groups/-/m/t/1398247" />
        <rdf:li rdf:resource="https://community.wolfram.com/groups/-/m/t/3451447" />
        <rdf:li rdf:resource="https://community.wolfram.com/groups/-/m/t/3440348" />
        <rdf:li rdf:resource="https://community.wolfram.com/groups/-/m/t/3351665" />
        <rdf:li rdf:resource="https://community.wolfram.com/groups/-/m/t/3345812" />
        <rdf:li rdf:resource="https://community.wolfram.com/groups/-/m/t/3343527" />
        <rdf:li rdf:resource="https://community.wolfram.com/groups/-/m/t/3330200" />
        <rdf:li rdf:resource="https://community.wolfram.com/groups/-/m/t/2780745" />
        <rdf:li rdf:resource="https://community.wolfram.com/groups/-/m/t/3162229" />
        <rdf:li rdf:resource="https://community.wolfram.com/groups/-/m/t/3102154" />
        <rdf:li rdf:resource="https://community.wolfram.com/groups/-/m/t/3092743" />
        <rdf:li rdf:resource="https://community.wolfram.com/groups/-/m/t/3081212" />
        <rdf:li rdf:resource="https://community.wolfram.com/groups/-/m/t/1874271" />
        <rdf:li rdf:resource="https://community.wolfram.com/groups/-/m/t/2266597" />
        <rdf:li rdf:resource="https://community.wolfram.com/groups/-/m/t/2667943" />
        <rdf:li rdf:resource="https://community.wolfram.com/groups/-/m/t/2661218" />
        <rdf:li rdf:resource="https://community.wolfram.com/groups/-/m/t/2613468" />
      </rdf:Seq>
    </items>
  </channel>
  <item rdf:about="https://community.wolfram.com/groups/-/m/t/3595346">
    <title>Residue curves for a reactive ternary system</title>
    <link>https://community.wolfram.com/groups/-/m/t/3595346</link>
    <description>&amp;amp;[Wolfram Notebook][1]&#xD;
&#xD;
&#xD;
  [1]: https://www.wolframcloud.com/obj/91f937b7-01ac-4791-827a-9468625c9ca5</description>
    <dc:creator>Housam Binous</dc:creator>
    <dc:date>2025-12-22T17:59:47Z</dc:date>
  </item>
  <item rdf:about="https://community.wolfram.com/groups/-/m/t/3550739">
    <title>Computing integral errors: Mathematica vs Matlab Simulink</title>
    <link>https://community.wolfram.com/groups/-/m/t/3550739</link>
    <description>&amp;amp;[Wolfram Notebook][1]&#xD;
&#xD;
&#xD;
  [1]: https://www.wolframcloud.com/obj/dfba162b-19a8-43a5-96b9-06f2bc0091e0</description>
    <dc:creator>Housam Binous</dc:creator>
    <dc:date>2025-09-24T15:28:25Z</dc:date>
  </item>
  <item rdf:about="https://community.wolfram.com/groups/-/m/t/3537445">
    <title>CH4 molecular geometry: model verification and minimal enclosing tetrahedron</title>
    <link>https://community.wolfram.com/groups/-/m/t/3537445</link>
    <description>The question is:&#xD;
&#xD;
![enter image description here][1]&#xD;
&#xD;
    vertices = {{0, 0, Sqrt[2/3] - 1/(2 Sqrt[6])}, {0, &#xD;
        Sqrt[3]/3, -1/(2 Sqrt[6])}, {-1/2, -Sqrt[3]/&#xD;
         6, -1/(2 Sqrt[6])}, {1/2, -Sqrt[3]/6, -1/(2 Sqrt[6])}   };&#xD;
    &#xD;
    &#xD;
    scale = (6 Sqrt[2])/Sqrt[Total[(vertices[[1]] - vertices[[2]])^2]];&#xD;
    scaledVertices = scale*vertices;&#xD;
    &#xD;
    &#xD;
    center = Mean[scaledVertices];&#xD;
    &#xD;
    &#xD;
    a = 6 Sqrt[2]; &#xD;
    h = a Sqrt[2/3]; &#xD;
    R = 3/4 h; &#xD;
    r_inner = h/4; &#xD;
    maxBlackRadius = R - 1; &#xD;
    &#xD;
    &#xD;
    hydrogenBalls = Table[Sphere[scaledVertices[[i]], 1], {i, 4}];&#xD;
    carbonBall = Sphere[center, maxBlackRadius]; &#xD;
    sticks = Table[Cylinder[{center, scaledVertices[[i]]}, 0.1], {i, 4}];&#xD;
    &#xD;
    modelPlot = &#xD;
     Graphics3D[{{White, Opacity[0.8], hydrogenBalls}, {Black, &#xD;
        Opacity[0.8], carbonBall}, {Gray, Opacity[0.6], sticks}, {Red, &#xD;
        PointSize[0.02], Point[center]}, {Blue, PointSize[0.02], &#xD;
        Point[scaledVertices]}}, Boxed -&amp;gt; True, Lighting -&amp;gt; &amp;#034;Neutral&amp;#034;, &#xD;
      ViewPoint -&amp;gt; {1, 1, 1}, ImageSize -&amp;gt; 500, PlotLabel -&amp;gt; &amp;#034;Model&amp;#034;]&#xD;
&#xD;
![enter image description here][2]&#xD;
&#xD;
My question is: According to the requirements of the problem, what are some good methods to draw the three-dimensional structure of methane, and to draw a circumscribed regular tetrahedron that can accommodate this structure while meeting the requirements, and to determine the correctness of the four options.&#xD;
&#xD;
&#xD;
  [1]: https://community.wolfram.com//c/portal/getImageAttachment?filename=2025-08-29_195609.png&amp;amp;userId=3529132&#xD;
  [2]: https://community.wolfram.com//c/portal/getImageAttachment?filename=.png&amp;amp;userId=3529132</description>
    <dc:creator>Wen Dao</dc:creator>
    <dc:date>2025-08-29T12:12:40Z</dc:date>
  </item>
  <item rdf:about="https://community.wolfram.com/groups/-/m/t/1398247">
    <title>Solve a PDE with boundary conditions (chemical adsorption in fixed beds)?</title>
    <link>https://community.wolfram.com/groups/-/m/t/1398247</link>
    <description>Dear Wolfram team:&#xD;
&#xD;
I have been trying for week to solve a system of 2 partial differential equations describing the adsorption of a chemical substance on a fixed bed (for example, a column of activated carbon). The 2 equations are the following, taken from McCabe (1993):&#xD;
&#xD;
![Description of eq 1][1]&#xD;
&#xD;
![Description of eq 2][2]&#xD;
&#xD;
  [1]: http://community.wolfram.com//c/portal/getImageAttachment?filename=EQ1.png&amp;amp;userId=1020580&#xD;
  [2]: http://community.wolfram.com//c/portal/getImageAttachment?filename=EQ2.png&amp;amp;userId=1020580&#xD;
&#xD;
Unfortunately I cannot get past the general solution (with arbitrary constants) because when I try to put boundary conditions the Mathematica program  fails. Maybe I am using the wrong command or syntax, or maybe there are too much or too few boundary conditions.&#xD;
&#xD;
I have left attached the program, where I tryed to simplify the problem combining both equations in a third.&#xD;
Thank you in advance for your help.&#xD;
&#xD;
Best regards,&#xD;
&#xD;
Alberto Silva</description>
    <dc:creator>Alberto Silva Ariano</dc:creator>
    <dc:date>2018-08-06T01:04:00Z</dc:date>
  </item>
  <item rdf:about="https://community.wolfram.com/groups/-/m/t/3451447">
    <title>3D rotate a MoleculePlot3D</title>
    <link>https://community.wolfram.com/groups/-/m/t/3451447</link>
    <description>I am really impressed by MoleculePlot3D, but I am wondering how to make the sort of classic &amp;#034;spinning molecule&amp;#034; GIF. The things I have tried ViewAngle and ViewCenter, but only succeed in a sort of fisheye look. I have used the following code (from documentation) as a template to no real success:&#xD;
&#xD;
    With[{obj = PolyhedronData[&amp;#034;Icosahedron&amp;#034;]}, &#xD;
     Animate[Show[obj, SphericalRegion -&amp;gt; True, &#xD;
       ViewVector -&amp;gt; {5 Cos[t], 5 Sin[t], 10 Sin[t/2]}], {t, 0, 4 Pi}, &#xD;
      AnimationRunning -&amp;gt; False]]&#xD;
&#xD;
Thank you for any help, I think this is a really cool area of the Wolfram Language! Forgive me if I am missing an obvious solution.</description>
    <dc:creator>Robert Mendelsohn</dc:creator>
    <dc:date>2025-04-27T18:52:43Z</dc:date>
  </item>
  <item rdf:about="https://community.wolfram.com/groups/-/m/t/3440348">
    <title>Has the latest update to IsotopeData caused a loss of precision?</title>
    <link>https://community.wolfram.com/groups/-/m/t/3440348</link>
    <description>In version 14.2, the outputs of IsotopeData that produce fractional answers (e.g., isotope abundances, decay branching ratios), have suffered a change in format (compared to version 13.2). The new format not only is less convenient, but also appears to cause a loss in precision.&#xD;
&#xD;
For example, in version 13.2, the isotope abundances of lithium-6 and lithium-7 print as 0.0759 and 0.9241, respectively. The abundances of all other lithium isotopes output 0. (with a decimal point). These values agree with other databases.&#xD;
&#xD;
In version 14.2, the isotope abundances of  lithium-6 and lithium-7 print as 0. % (with a red square outline) and 1. x 10^2 %. The abundances of the other lithium isotopes are printed as 0%, without the decimal point or a red outline.&#xD;
&#xD;
This suggests there has been some rounding that caused some values to lose all precision. It is also worth noting that the version 13.2 outputs, with the simple fractions, were more convenient for calculations. &#xD;
&#xD;
There is a way to recover something similar to the outputs of version 13.2 in version 14.2, and that is to add 0.0 to the result. &#xD;
&#xD;
With this trick, in version 14.2, the isotope abundances of  lithium-6 and lithium-7 print as 0.0485 and 0.9515, respectively (no more percentages). While these values add to 1, as they should, they are different from the values obtained in version 13.2.&#xD;
&#xD;
It is possible that these differences are the result of an update of our knowledge of the abundances of the lithium isotopes, but, to my knowledge, other databases have not changed their values. It seems to me that this is the result of an unsuccessful attempt to standardize the reported precision of these data.&#xD;
&#xD;
There are many similar fractional outputs in this database that are close to zero (many isotopes are extremely rare, so that their abundances will be close to zero, and many others have decay modes that occur extremely seldom, causing very small branching ratios). The concern is that all these outputs will lose accuracy and be reported as zero, and a simple attempt to recover these values seems to lead to incorrect outputs.&#xD;
&#xD;
Does anybody have an explanation for this? Is this just a bug? Is there a way to control this behavior with some precision or formatting option?&#xD;
&#xD;
Thanks in advance for your time,&#xD;
Otto Linsuaín&#xD;
&#xD;
P.S. The outputs mentioned above can be obtained with commands like:&#xD;
IsotopeData[&amp;#034;Lithium6&amp;#034;,&amp;#034;IsotopeAbundance&amp;#034;]&#xD;
&#xD;
For the branching ratios one should use an unstable isotope, for example:&#xD;
IsotopeData[&amp;#034;Neptunium239&amp;#034;,&amp;#034;BranchingRatios&amp;#034;]</description>
    <dc:creator>Otto Linsuain</dc:creator>
    <dc:date>2025-04-06T06:44:29Z</dc:date>
  </item>
  <item rdf:about="https://community.wolfram.com/groups/-/m/t/3351665">
    <title>Can System Modeler simulate complex Chemical processes with multiple units?</title>
    <link>https://community.wolfram.com/groups/-/m/t/3351665</link>
    <description>![enter image description here][1]&#xD;
&#xD;
It has been few a years since I used Wolfram Mathematica.  Here is a flow chart of a chemical process with fractional distillation and recycling.&#xD;
&#xD;
Does System Modeler able to simulate a process with multiple process units where there is heat, material balance with chemical reactions and manipulate the heat and material input?&#xD;
&#xD;
I will like to see a working sample.&#xD;
&#xD;
I only see samples of single Stirr tank reactions, but nothing evoilving a recyclig process with dynamic processes.&#xD;
&#xD;
&#xD;
  [1]: https://community.wolfram.com//c/portal/getImageAttachment?filename=5752imageprocess.png&amp;amp;userId=120134</description>
    <dc:creator>Jose Calderon</dc:creator>
    <dc:date>2025-01-06T06:15:44Z</dc:date>
  </item>
  <item rdf:about="https://community.wolfram.com/groups/-/m/t/3345812">
    <title>Electrochemical corrosion process simulation</title>
    <link>https://community.wolfram.com/groups/-/m/t/3345812</link>
    <description>I cannot find any package, tool, examples, or library in System Modeler related to electrochemistry or electrochemical processes, like electrochemical cell simulation, anodes, cathodes, electrochemical corrosion, or anything related to these topics.&#xD;
&#xD;
Can someone guide me on how to use the system modeler to simulate the corrosion process and cathodic protection?&#xD;
&#xD;
Regards, Rogelio</description>
    <dc:creator>Rogelio de las Casas</dc:creator>
    <dc:date>2024-12-27T16:03:01Z</dc:date>
  </item>
  <item rdf:about="https://community.wolfram.com/groups/-/m/t/3343527">
    <title>Missing outputs from ElementData</title>
    <link>https://community.wolfram.com/groups/-/m/t/3343527</link>
    <description>I am facing some issue with the Built-in Function. How to fix this? ![See the screenshot attached][1]&#xD;
Thank you.&#xD;
&#xD;
  [1]: https://community.wolfram.com//c/portal/getImageAttachment?filename=Screenshotfrom2024-12-2311-47-43.png&amp;amp;userId=3343398</description>
    <dc:creator>Mahaveer Gangwar</dc:creator>
    <dc:date>2024-12-23T06:28:33Z</dc:date>
  </item>
  <item rdf:about="https://community.wolfram.com/groups/-/m/t/3330200">
    <title>Errors in code for fission yield analysis across different patterns</title>
    <link>https://community.wolfram.com/groups/-/m/t/3330200</link>
    <description>I am working on a Mathematica program to analyze fission yield data of U-235. The program calculates results for 5 different patterns based on neutron incident energies. While the calculation works correctly for pattern = 3, it produces errors for other patterns such as pattern = 1.&#xD;
&#xD;
Key errors include:&#xD;
&#xD;
Union::normal: Nonatomic expression expected at position 1 in Union[vars12].  &#xD;
Part::take: Cannot take positions 6 through 42 in datax2.  &#xD;
FindMinimum::vloc: The variable Union[vars12] cannot be localized so that it can be assigned to numerical values.&#xD;
&#xD;
The result Reff becomes undefined (Indeterminate) or infinite.  &#xD;
What might be causing these issues, and how can I modify the code to ensure correct calculations for all patterns?&#xD;
&#xD;
&#xD;
Note:  &#xD;
&#xD;
When running the program on Mathematica version 11.3 or later, changes in the `FindMinimum` function result in errors even for the previously working pattern. Therefore, I kindly request that this issue be evaluated on Mathematica version 11.2.</description>
    <dc:creator>Hirokazu Maruyama</dc:creator>
    <dc:date>2024-11-28T14:32:27Z</dc:date>
  </item>
  <item rdf:about="https://community.wolfram.com/groups/-/m/t/2780745">
    <title>Result change when variable has index in WA (modeling chemical equilibrium)</title>
    <link>https://community.wolfram.com/groups/-/m/t/2780745</link>
    <description>Dear community,&#xD;
I entered some equations related to chemical equilibrium. I entered the input once with an index to one of the variables (&amp;#034;a_1&amp;#034;) and once without the indices (just &amp;#034;a&amp;#034;). The result produced by Wolfram Alpha is completely different. Here are the two inputs to the solver:&#xD;
&#xD;
10^-2.16=b*h/a, 10^-7.21=c*h/b, 10^-12.32=d*h/c, a+b+c+d=1, 10^-14=o*h&#xD;
&#xD;
and&#xD;
&#xD;
10^-2.16=b*h/a_1, 10^-7.21=c*h/b, 10^-12.32=d*h/c, a_1+b+c+d=1, 10^-14=o*h&#xD;
&#xD;
Furthermore the result changes everytime I add an index to another variable (&amp;#034;b_1&amp;#034; instead of &amp;#034;b&amp;#034;,...). Does anybody know the reason and can point me towards a valid solution?&#xD;
&#xD;
Thank you very much! Andreas&#xD;
&#xD;
P.S.: My project is chemistry-related. I want to model the course of the concentrations of all species during the titration of phosphoric acid with sodium hydroxide. The Input above models the first step, where no sodium hydroxide.</description>
    <dc:creator>Andreas Uhe</dc:creator>
    <dc:date>2023-01-13T10:15:45Z</dc:date>
  </item>
  <item rdf:about="https://community.wolfram.com/groups/-/m/t/3162229">
    <title>Why my IAE plot is off with respect to result by Dorf and Bishop?</title>
    <link>https://community.wolfram.com/groups/-/m/t/3162229</link>
    <description>&amp;amp;[Wolfram Notebook][1]&#xD;
&#xD;
&#xD;
  [1]: https://www.wolframcloud.com/obj/99bc3297-6cf0-4328-ac97-fa0050003574</description>
    <dc:creator>Housam Binous</dc:creator>
    <dc:date>2024-04-21T14:24:12Z</dc:date>
  </item>
  <item rdf:about="https://community.wolfram.com/groups/-/m/t/3102154">
    <title>Ponchon-Savarit Diagram using Aspen-Hysys Thermodynamic Database</title>
    <link>https://community.wolfram.com/groups/-/m/t/3102154</link>
    <description>&amp;amp;[Wolfram Notebook][1]&#xD;
&#xD;
&#xD;
  [1]: https://www.wolframcloud.com/obj/7b2326c8-f441-4cf6-b6b6-15d41e547a32</description>
    <dc:creator>Housam Binous</dc:creator>
    <dc:date>2024-01-13T05:26:37Z</dc:date>
  </item>
  <item rdf:about="https://community.wolfram.com/groups/-/m/t/3092743">
    <title>Input forms in SpectralLineData</title>
    <link>https://community.wolfram.com/groups/-/m/t/3092743</link>
    <description>Hello, I&amp;#039;m using the SpectralLineData function documented at:&#xD;
&#xD;
https://reference.wolfram.com/language/ref/SpectralLineData.html&#xD;
&#xD;
Does anyone know what the yellow boxes in the Examples are called? I&amp;#039;d like to learn how to go back and forth between these boxes and normal code, but I don&amp;#039;t know what the boxes are called so I can&amp;#039;t research how to use them.&#xD;
&#xD;
Thanks.</description>
    <dc:creator>Joe Prendergast</dc:creator>
    <dc:date>2023-12-30T00:04:45Z</dc:date>
  </item>
  <item rdf:about="https://community.wolfram.com/groups/-/m/t/3081212">
    <title>Chemical Element Binding Energy vs Ionization Energies</title>
    <link>https://community.wolfram.com/groups/-/m/t/3081212</link>
    <description>&amp;amp;[Wolfram Notebook][1]&#xD;
&#xD;
*Binding* energy of an element&amp;#039;s outermost electron is the amount of energy it takes to produce a free electron from that in the element&amp;#039;s ground state.  Is the binding energy therefore the last entry in the list of wavenumbers and is that the same, converted to eV, as given by the &amp;#034;IonizationEnergies&amp;#034; property of hydrogen?&#xD;
&#xD;
Why is there only one energy supplied in response to &amp;#034;IonizationEnergies&amp;#034; when, it would appear that `EntityClass[&amp;#034;AtomicLevel&amp;#034;,{Hydrogen&amp;#034;,1}]` should have the requisite data?&#xD;
&#xD;
Is there a more straightforward way to convert from &amp;#034;kJ/mol&amp;#034; to &amp;#034;eV&amp;#034;?&#xD;
  [1]: https://www.wolframcloud.com/obj/66ebfb38-c00d-4584-a9cd-957617631142</description>
    <dc:creator>James Bowery</dc:creator>
    <dc:date>2023-12-10T15:55:00Z</dc:date>
  </item>
  <item rdf:about="https://community.wolfram.com/groups/-/m/t/1874271">
    <title>Import SPE files into Mathematica Notebook?</title>
    <link>https://community.wolfram.com/groups/-/m/t/1874271</link>
    <description>I am trying to import an .SPE file from WinView (Princeton Instruments), I am wondering if this is possible and how to do it?</description>
    <dc:creator>Nicholas Heller</dc:creator>
    <dc:date>2020-02-06T23:22:50Z</dc:date>
  </item>
  <item rdf:about="https://community.wolfram.com/groups/-/m/t/2266597">
    <title>Computational chemist - full time job opportunity</title>
    <link>https://community.wolfram.com/groups/-/m/t/2266597</link>
    <description>Hello!  I am hiring for a full time Computational Chemistry for one of my companies, SmarterSorting.  Very much a WolframLang, WolframAlpha opportunity.  Company is based in Austin, TX, but this role can work anywhere.&#xD;
&#xD;
If you love Wolfram, chemistry, computational thinking and fast growing startups that have lots of fun data, apply or message me.&#xD;
&#xD;
https://boards.greenhouse.io/smartersorting/jobs/4005347004&amp;amp;dep=4004121004</description>
    <dc:creator>Russell Foltz-Smith</dc:creator>
    <dc:date>2021-05-12T14:09:51Z</dc:date>
  </item>
  <item rdf:about="https://community.wolfram.com/groups/-/m/t/2667943">
    <title>How to set up a simple plot of pH against volume of titrant?</title>
    <link>https://community.wolfram.com/groups/-/m/t/2667943</link>
    <description>My very first notebook.  I have the following equation&#xD;
Vb = Va * ((Ca/(1 + H/Ka)) - H + Kw/H) / ((Cb/(1 + Kw/(H*Kb))) + H - &#xD;
     Kw/H)&#xD;
which I have managed to set up in Mathematica.  I have tested it by entering different values of pH one at a time.&#xD;
For given values of Ca, Cb, Ka, Kb, Kw, and Va, I want to plot pH (y-axis) against Vb (x-axis) by specifying a range and increment of pH (e.g., from pH 3.0 to 12.5 in steps of 0.2)&#xD;
I would also like to tabulate the values of pH and Vb.&#xD;
&#xD;
My longer-term goal is to use manipulate to illustrate the effects of changes to Ca, Cb, Ka, Kb, and Va&#xD;
&#xD;
Any help with giving me a kickstart will be much appreciated&amp;amp;[Wolfram Notebook][1]&#xD;
&#xD;
&#xD;
  [1]: https://www.wolframcloud.com/obj/3408892f-86b1-4033-bb34-39c090f32b45</description>
    <dc:creator>Paul Newton</dc:creator>
    <dc:date>2022-10-23T21:22:41Z</dc:date>
  </item>
  <item rdf:about="https://community.wolfram.com/groups/-/m/t/2661218">
    <title>Viewing molecular orbitals .cube files. Problem with version 13.1</title>
    <link>https://community.wolfram.com/groups/-/m/t/2661218</link>
    <description>I have been using &#xD;
&#xD;
    Import[&amp;#034;xxxx.cube&amp;#034;,&amp;#034;Graphics3D&amp;#034;]&#xD;
&#xD;
to view molecular orbitals. E.g. the homo orbital of CO2: &#xD;
![homo.CO2][1]&#xD;
&#xD;
This was generated with version 12 of Mathematica. &#xD;
&#xD;
Using version 13.1 the exact same .cube file now renders as&#xD;
&#xD;
![homo.CO2.13.1][2]&#xD;
&#xD;
Something changed. Anyone knows what happened?&#xD;
&#xD;
Best regards&#xD;
Per Hedegård&#xD;
&#xD;
&#xD;
  [1]: https://community.wolfram.com//c/portal/getImageAttachment?filename=CO2.Homo.png&amp;amp;userId=554734&#xD;
  [2]: https://community.wolfram.com//c/portal/getImageAttachment?filename=CO2.Homo.13.1.png&amp;amp;userId=554734</description>
    <dc:creator>Per  Hedegard</dc:creator>
    <dc:date>2022-10-19T20:51:11Z</dc:date>
  </item>
  <item rdf:about="https://community.wolfram.com/groups/-/m/t/2613468">
    <title>Graph Convolution Network, molecule activity</title>
    <link>https://community.wolfram.com/groups/-/m/t/2613468</link>
    <description>I am trying to translate an example code for graph convolution network from Deep Learning for Physical Scientists: Accelerating Research with Machine Learning. Edward O. Pyzer-Knapp et. al. They have an example (Chapter 6, Section 6.3, Page 104-106) where they build a graph convolution network using Keras.&#xD;
&#xD;
The model takes a 132*132 adjacency matrix built out of molecule connectivity, with a flag of 1 or 0 for active or inactive molecule.&#xD;
&#xD;
    from tensorflow.keras import datasets, layers, models &#xD;
     from sklearn.model_selection import train_test_split &#xD;
     model = models.Sequential() &#xD;
     model.add(layers.Conv2D(64, (3, 3), activation=&amp;#039;relu&amp;#039;, input_shape=(None, None, 1), padding=&amp;#039;SAME&amp;#039;)) &#xD;
     model.add(layers.MaxPooling2D((2, 2))) &#xD;
     model.add(layers.Conv2D(32, (3, 3), activation=&amp;#039;relu&amp;#039;)) &#xD;
     model.add(layers.GlobalMaxPooling2D()) &#xD;
     model.add(layers.Dense(2, activation=&amp;#039;softmax&amp;#039;)) &#xD;
     model.compile(optimizer=&amp;#039;adam&amp;#039;, &#xD;
                   loss=&amp;#039;sparse_categorical_crossentropy&amp;#039;, &#xD;
                   metrics=[&amp;#039;accuracy&amp;#039;]) &#xD;
&#xD;
I would like to build this model in WL. My attempt thus far is&#xD;
&#xD;
    net = NetChain[{ConvolutionLayer[64, {3, 3}], &#xD;
       ElementwiseLayer[&amp;#034;ReLU&amp;#034;], PoolingLayer[{2, 2}, &amp;#034;Function&amp;#034; -&amp;gt; Max], &#xD;
       ConvolutionLayer[32, {2, 2}], &#xD;
       PoolingLayer[{2, 2}, &amp;#034;Function&amp;#034; -&amp;gt; Max], LinearLayer[2], &#xD;
       SoftmaxLayer[]}]&#xD;
&#xD;
I can&amp;#039;t find the GlobalMaxPooling option, and I am uncertain if my convolution layers are set up correctly. I appreciate if anyone could show how this could be done.&#xD;
&#xD;
Thank you,</description>
    <dc:creator>Arun Sharma</dc:creator>
    <dc:date>2022-09-08T13:03:07Z</dc:date>
  </item>
</rdf:RDF>

