Today's Summer 2020 daily update:
| Name |
Update for Today |
Plans for Tomorrow |
| Alex M. |
(See Alex P. for details on working on fixing loop issues)
Went over running the loop with Ryan. Worked with Ryan on trying to get the X11 forwarding for XF to work, but we couldn't get the license. Steven suggested that we try running from a Pitzer desktop, which I think is a good idea if the problem is with the X11 forwarding; if it's a license issue, then we need to figure out where the license is located. I thought this would have been written somewhere in our code, but I couldn't find it, so I may need to email OSC.
Worked more on the symmetric bicone part of the paper. I think I could afford to add detail to the results portion. I included the plots from the APS presentation, but we still need to finish a rerun with the same number of generations and individuals to see if the data makes sense (particularly, I want to see if there was a mistake made when continuing the run after it had finished the 10th generation because I'm skeptical about the fact that we got such large angles in generation 12).
|
I'll meet with Evelyn and Ryan tomorrow to discuss running the loop in more detail. I'll help Ryan use the Pitzer desktop, which will hopefully work. If all goes well, we can get them started on running the loop to check against the data from the run I mentioned in the paper.
I'm going to try writing in the section for Genetic Algorithms in the paper, since Kai said he wanted me to take the first attempt at it. I think for now I'd like to get comments on what I've written on the symmetric bicone portion before I clean it up, especially on whether or not I've given enough detail (and clearly enough).
|
| Alex P. |
Continued to test the database functionality. In the process of testing and debugging we also tested random seed functionality of the genetic algorithm as we used a fixed seed and successfully got repeat results across multiple runs, which allowed better testing as we could be certain to encounter the same scenario when testing and figuring out how to overcome an error.Altered some of the xmacros and xf calls to work with database and created the new macro skeleton for output.xmacro to use just with the database, in the process of looking through the xmacros we found some segments of the code to investigate further. Currently, we just use values 1 to NPOP of the XF simulation and rewrite the contents of 1 to NPOP each generation, but more simulation values continue to be generated even if unpopulated. Also looked more into this and started implementing a way to take out the excess directories while also planning out a way to have each generation use new values rather than overwriting so that we can have a full record of all the simulations across generations rather than just the most recent generations. |
Going to continue to test database, have a run starting gen 1 now and going to continue that tonight, hopefully will find useful results and if not then things to debug. Eliot and Leo asked for help tomorrow with C++ code as they have had some problems with pointers and I agreed to help them tomorrow. |
| Leo |
I, along with Eliot, worked on implementing the new parameter. We have been working on 2 methods, one being adding a 4th gene and the second being adding another chromosome. We are focusing our attention on the chromosome method because we feel it will lead its way nicely into our future work later this summer with adding more complicated parameters. We are running into some issues with pointers and vectors within the GA that we worked on fixing. |
Tomorrow we hope to fix the issues with the pointers and vectors, and hope to test it with multiple generations. |
| Eliot |
Currently working on developing GA further to properly accept 2 chromosomes. The first chromosome has some radius, length, theta and the second chromosome has the same radius and theta, but its own unique length. I currently have the code properly populating an initial pool. However, it can not quite evolve. The GA is free of syntax errors, but I am getting some type of pointer error. I am working to resolve this. I believe it is the only obstacle left on the GA side of things, but it may not be.
|
I will continue working on fixing the errors in the GA, and hopefully finish that tomorrow. If so, I will begin learning about the XF implementation.
|
| Evelyn |
Had class, so I couldn't meet with the group
|
Get caught up with the stuff from today |
| Ryan |
|
|
Julie: Since I won't be doing GENETIS work every day necessarily, I'll add myself down here with a blurb when I am working with the others. I will also be continuing my daily updates on Dropbox for non-GENETIS specifics. |