151 reputation
7
bio website
location Colorado Springs, CO
age
visits member for 1 year, 11 months
seen Aug 25 at 21:50

Jul
20
comment Plot Legends and Frame Ticks in Array Plots
Interesting, I wonder why it affects the frame ticks?
May
30
comment Function that is a good fit to the plot made with SmoothHistogram
What is an actual function though? For instance, it looks like I could fit a polynomial to this curve. I want to know the actual best fitting function, not just recreate the plot via some other means.
May
24
comment Replacing High Densities of 1.s in Very Large Binary Arrays
It does appear to accomplish what I originally intended. What if the "chunk" were to move one increment at a time versus an entire "chunk" size at a time? That way, any given "chunk" in the data would never exceed "lim".
May
24
comment Replacing High Densities of 1.s in Very Large Binary Arrays
I like the idea of the window moving one increment at a time versus moving the entire length of the window each time. By having it moving one increment at a time it guarantees that there will never be more than "threshold" 1s in any given "bandwidth".
May
23
comment Replacing High Densities of 1.s in Very Large Binary Arrays
If the window is 3 and my max density say 2 or more and I am moving the window 3 indices at a time then the result would be {0,0,0,0,0,0,0,0,0,1,0,0}.
May
23
comment Replacing High Densities of 1.s in Very Large Binary Arrays
I like the moving window idea. However, I think you have what I am trying to do backwards. If the density is too high (meaning that if there are too many 1s close together in a given region), I want to make everything zero. I want less ones, not more 1s.
May
23
comment Replacing High Densities of 1.s in Very Large Binary Arrays
No, not quite. I think it is my fault for not being clear enough. Let me revise my posting and give a small scale example.
May
22
comment Replacing Patterns In a List of Varying Length
@rasher I think I may just start a new question...
May
22
comment Replacing Patterns In a List of Varying Length
Thanks guys. This has been very helpful. It is very fast and doing exactly what I had originally intended. Unfortunately, it has led me to another problem that is very similar to this one. What I am finding is that just looking for a minimum number of 1s in a row isn't good enough. I am trying to carve out areas in my data where there is a high density of 1s without completely zeroing out everything. This means that there could be some 0s sprinkled in randomly that break up the pattern and I still want that area zeroed out. I would think this requires an entirely different approach?
May
21
comment Replacing Patterns In a List of Varying Length
This method appears to work with the test data above. However, when I use data on the order of {792001,111}, for some reason it doesn't work anymore. For some reason it yielded {792001,104} array and it didn't replace any patterns of 1 to 0.
May
21
comment Replacing Patterns In a List of Varying Length
I also forgot something in my test data above. These numbers are floating point numbers, as it is much faster to manipulate large arrays using floating point numbers. So, my data looks something like this.
May
21
comment Replacing Patterns In a List of Varying Length
I like the idea of converting to a string. However, it isn't apparent to me how I can adjust the number of ones in the pattern before making them zeros. For instance, let's say I need to see 100 ones in a row before making them zeros.
May
21
comment Replacing Patterns In a List of Varying Length
Ultimately, for the large arrays I am working with, what I want to do is replace any sequence of ones that are larger than 100 or maybe even 1000. I'm not sure yet. What I posted above was an example of what I am trying to do at a smaller scale so I can propose the question.
May
21
comment Replacing Patterns In a List of Varying Length
Thanks. I knew there was probably a way to do it without indexing through it. Hopefully this will be fast. I am working with arrays on the order of {100,100000}
Nov
20
comment Fluorescence Correlation Spectroscopy Simulation
Yes, I suspect that it is a memory issue. I am working with 4GB of memory and a 2GHz Intel Core 2 Duo processor.
Nov
19
comment Fluorescence Correlation Spectroscopy Simulation
Yes, when i set n to 50 and Deltat to 10^6 'particleintensity' is the slowest bit. However, this is not the case when n is set to 5 and Deltat is 10^6. I need a way to only calculate particle intensity for particles that are close to the origin.
Nov
18
comment Fluorescence Correlation Spectroscopy Simulation
This has been working great. However, I have found that I need Deltat to be at least 1000000 and n to be at least 50. Using these parameters, the calculation of particle intensity still takes a considerable amount of time. Realistically, I only need to calculate the particle intensity when particles are close to the origin, say within a distance of Omegaz or Omegar. If I could only calculate particle intensity when particles are close to the origin, I think that would seriously reduce my calculation time. Any thoughts?
Nov
5
comment Fluorescence Correlation Spectroscopy Simulation
That's even better! Thanks.
Nov
5
comment Fluorescence Correlation Spectroscopy Simulation
@ruebenko No, I haven't attempted to compile it. My programming skills are still rather basic. I'm not quite sure how to use Compile yet.
Nov
5
comment Fluorescence Correlation Spectroscopy Simulation
@drN In a sense, yes this is molecular dynamics. Once I have an total intensity profile I can autocorrelate it and extract information such as concentration and diffusion coefficients among many other things.