Since I left off week 4 with learning the method of flare energy calculations using one event as an example, the main work this week was to go through the steps 3 through 5 to actually improve the values for the three parameters. For the meetings with Chunming, I first received some comments about the week 4 progress report presentation, mainly on the images that I used and how I could improve for the midterm presentation.
I learned what specific test values I could start with and how I can make adjustments to improve the accuracy and precision. I also learned that the method of comparison between synthetic emission curves and lightcurves that Chunming has in mind is the area method, where we calculate the area difference between the curves and normalize it. We were not sure if this was the best method we could come up with, so we decided to try another method, where I would look at the Pearson correlation coefficent between the flux vectors of the synthetic emission curves and the original lightcurves. I wrote the code for finding the top 10 combinations of test values by the correlation method. Even though this was a lot of copying and pasting from the code for the area method, it was still exciting since it was the first code I had actually written on my own beyond adding comments or just a few lines of code. After running S5 once with 125 combinations of test values to find the top 10 combinations with both methods, I found that there were 4/125 sets of test values that made top 10 in both methods, giving us confidence that those were certainly good. However we didn't get to the point of finalizing the comparison methods since the codes weren't done until near the end of the week. Since we had the midterm presentation I presented my work up to this point.
For week 6, we had a midterm presentation instead of the biweekly progress report meetings. The slides I used for the midterm presentation can be found here.