Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Project Submission #7

Open
wants to merge 37 commits into
base: master
Choose a base branch
from

Conversation

AkshayaSrivastava
Copy link

I have a draft version in here entitled Final_Project_Draft. Please ignore this unless you would like to check something that i was working on in the process of completing the project. If needed, I can remove it as well, or file it in a separate folder for your convenience.

The actual final project is entitled: "Telegrapher's Equations Final Project by Akshaya Srivastava"

Please use that for grading purposes.

AkshayaSrivastava and others added 20 commits December 14, 2015 01:51
…t sure if it's an issue with markdown syntax or the web address. Will investigate on personal machine, but isn't strictly necessary for derivation.
…sion. Trying out new method to source information.
…rent wave equation. Added a dig deeper segment, to keep scope of project in check.
…on and then the theory will be done and coding can begin!
… issue, and figured out how to create reference links and comment out the references in the Markdown Block. May list references again at the end, if time permits; if not, they are scattered throughout the text where used. Code pending.
…un together. No longer trying that. also, wrote code for initial signal generation function, and tested to make sure it's running as expected. May remove rhb/lhb params if needed
…HB/LHB params. Don't think I can ignore them...Might have something to do with the initial condition i'm looking at. Worst Case Scenario, use Equation 32 only for the n = 1, and then recode Equation 24 for all others. Also want to experiment with timesteps to figure out exactly what's happening
…mation throughout. Will keep the previous versions of the code in here and add text to explain what's going on and why it isn't working. also, need to debug the wave somewhat, just a smidge...
…meters, define functions, invoke functions, plot. All code cells will follow this progression to help keep scope of variables and functions in check. Also assists in Debugging if everything that the code needs is called in the same cell.
…ext discussing first order and the traps you can fall into if you are not careful. Called out references explicitely in the end of the report. Now I just need to fix the code and run the code once in JupyterHub to test viridis, and this should be done!
…ns and timestep size. Have to do a CFL analysis now. Leaving the current and resistivity as a Dig Deeper Exercise for now.
…r exist. May need to leave that as a Dig Deeper exercise...
…boundary conditions to the solver, which lead to strang boundary conditions past the initial values. Switching to Jupyter Hub to make sure viridis works and better visualize. Should be able to finish up soon from there.
…e a final version that doesn't take so long to run, and uses cells better.
@ncclementi
Copy link
Member

  • Add license like the ones we have in our notebooks.
  • You first paragraph is almost a copy form wikipedia, you need to re-write it using your own words, not just changing couple of them from the original source, that is technically plagiarism.
  • Before equation 1 you say: "From these Equations, the following model arises:" when you say "from these equations" to which ones are you referring? there are no equations before.

*In the Voltage section, after "That is:" there should be a vertical space, between the text and the equation.

*In equations 21, 22, 23., 24 and so on. I would recomnd you to use \left[ and \right] in your
LaTex so the brackets adjust to the size of what is it between them.

  • When you want to import many modules from a library, you can just write:

        from matplotlib import pyplot, cm 
    

    you don't need to write them separately.

  • If you have warnings, you should try to fixed them, not add a line to ignore them.

                warnings.filterwarnings('ignore')
    

    the problem is when you put a float as an index. Like here v0[0:nx/2+1]=alpha,
    nx/2+1 this is a float.

  • When you are plotting, if you add a semicolon in the last line of the cell,
    this:
    <matplotlib.legend.Legend at 0x7f3fdd068b70>

    won't appear and looksmore neat.

  • You are using the same name (init) for 4 different functions, that might cause problems and /or confusion.

  • When you say you are "getting first order answers" what are you trying to say? I don't understand when you say that "it looks like a first order problem"

  • Two of the "init" functions, are exactly the same. Once you define a function you don't need to write it again. You might have problems because you are using the same name 4 different times.

  • In the In[19] the voltage appear to 1e37 which units are you using? Why this number? and it also appear the Initial in blue in the legend, but it isn't in the plot.

  • In the In[21] the legend says Initial, Neumann, Solved, but you only have "Solved".

  • In In[22] you have this line:

    ERROR! Session/line number was not unique in database. History logging moved to new    session 66 
    

    What is it?
    I'm trying to run this one, sometimes I get errors and when it runs it takes forever. I guess it is because of the issue of calling 4 different functions "init". We should be able to run your code with no errors.

  • How we know that this results are correct, do you have any source or problem for comparison/test?

typos:

functions or the ----> functions of the
And What ------------> And what

@AkshayaSrivastava
Copy link
Author

Thanks for all the feedback!

I'll get started on making those changes immediately. Do I have to do another pull request after each commit, or once I'm done making the changes?

@labarba
Copy link
Member

labarba commented Dec 22, 2015

Any new commits get automatically added to this pull request.

@AkshayaSrivastava
Copy link
Author

ok. thanks!

…tive commons stuf correctly as well, following the samples on the course materials.
…bout half of the document (discussion of first order) adds more confusion than clarity. The purpose of that section was to show how using a first order approximation would lead to first order solutions, even with a second order discretization scheme. I think this could be better conveyed through equations, and would clarify where the solution is going to. Also, added a basic outline of the rest of the lesson, which will help when writing, and when readers are reading. Created brand new version of the assignment in order to 'start fresh' and write better, cleaner code, rather than trying to modifexisting structure, and potentially missing errors, etc.
… in place, we can use sympy to plot them and that will be our exact solutions with which to compare the numerical solutions.
…heoretical calculations in latex and created equation 5 through sympy.
…nitial conditions now. also realized that sympy can only do first order pdes. moved code for sympy solution into last cell in case it's needed. may try using it to solve the ODEs that result from Seperation of Variables, if possible.
…or both, so we can alway sturn this into a function to plot against our numerical solution.
…n code design. Next step is to set up the loop.
…debugging code in there for now, which will be commented out and can be used again to see variables change as the loop runs.
…rk on conclusion and source to CFL condition derivation. Need to clean up code for debugging as well.
…de. Will run once more to make sure everything is good and proofread.
…reading. Figuring out how to avoid runtime errors now, like the overflows and sesssion/line number not unique in database.
… session/line number not unique errors occur is when you run the same cell twice without interuppting the kernel first. I did that with the exact plot and got the error and two plots, even though there was only one command - that leads me to believe that there was a risidual command in the kernal that was still being executed as I ran it again. The 3-d plots may take some time to load, but the actual calculations are done in under a minute. On Jupyter Hub and my machine, the entire notebook rendered in under 10 minutes. In any case, I think I'm done. Merry Xmas to All , and to All a good night!
@AkshayaSrivastava
Copy link
Author

I rewrote the assignment, focusing less on showing the 1st order neumann condition fails, and more on the second order scheme, I also fixed all of the errors that were appearing earlier, (the last commit message has more info on that), and I've moved the older versions into a separate folder. Please use V2 for grading.

Merry Xmas to all and to all a good night!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants