Deborah Gross

Show:

What's happening?

  • Comment: Refined centroids needs testing/fixing

    If this is being addressed: it would be great to be able to specify not just K = 12 - 15, but also K = 5 - 55, counting by 5 or 10 or whatever.

    2009-10-19 01:24:22 UTC in EDAM ENCHILADA

  • Comment: Analyze Particle Adjustments

    Oh, sorry -- I meant to finish my statement in 1c, but I had to check in the software first, and then I forgot to update. It's hard to zoom from low to high on the x-axis, if you want to catch zero, based on the way the zoom-bar works.

    2009-07-15 13:54:37 UTC in EDAM ENCHILADA

  • Aggregate timelines bins when there is no data

    It would be very useful if timebins that are empty, i.e. which contain no data, are put in with a blank value for the data (not a zero value!), so that when we export the data we have a time point for every unit of time requested. Zero values are reserved for times when the actual data value is zero. Plots look bad otherwise, as this example describes: Say you ran the instrument for 4 days...

    2009-07-15 13:49:16 UTC in EDAM ENCHILADA

  • Analyze Particle Adjustments

    Here are some things on the wish list: 1) the axes in the graphs in "Analyze Particle" are pretty weird. It would be nice to have the following: a) Tick marks that are always at integer values, and always at reasonable (i.e. 1's, 5's or 10's ) values. b) Zooms that can be "locked," meaning that if one zooms to a particular region of the spectrum, one can maintain that zoom while using the...

    2009-07-15 13:43:59 UTC in EDAM ENCHILADA

  • Aggregate Timeline Bins Can't Be > 24hr

    Manually entered time bins cannot be longer than 24 hours. For example, when manually creating time-bins that are 24:13:00 long, the data is aggregated using time bins that are 00:13:00 long.

    2009-07-10 13:12:31 UTC in EDAM ENCHILADA

  • Deleting Data Collections Doesn't Work Right

    Text here from my student Sam Thompson, who did some sleuthing. Note also the attached error log which has the text of the errors she was getting. Problem: When you delete a data collection, Enchilada doesn’t recognize it in the total particle count, which leads to errors when trying to cluster. Tests: 1. Fresh new data set (117 particles) with two collections in it. Clustered it.

    2009-06-23 19:14:50 UTC in EDAM ENCHILADA

  • Export Cluster Centers

    Here is some info about a simple format to export cluster centers that would be very useful. We already have some tools that are relatively widely used with our instrument that can grab this file format and make a nice graph, and it's easy to work with if that's not desired. The one thing that I would like to ask that you build into this is an opportunity for the user to define the highest...

    2009-02-16 18:33:21 UTC in EDAM ENCHILADA

  • Comment: Scale for cluster centers is hacked, with significant issues

    I run the risk of making this unreadable, but I will throw my comments in below Dave's, where needed, but in ALL CAPS. "1. Relative area and height are always 1.0 for all peaks in the cluster center. This is the case no matter which clustering algorithm you use. I'm not entirely sure what the correct calculation would be to get the relative area and height or whether there is one."...

    2009-02-13 16:18:14 UTC in EDAM ENCHILADA

  • Comment: Scale for cluster centers is hacked, with significant issues

    I run the risk of making this unreadable, but I will throw my comments in below Dave's, where needed, but in ALL CAPS. "1. Relative area and height are always 1.0 for all peaks in the cluster center. This is the case no matter which clustering algorithm you use. I'm not entirely sure what the correct calculation would be to get the relative area and height or whether there is one."...

    2009-02-13 16:14:18 UTC in EDAM ENCHILADA

  • Comment: Scale for cluster centers is hacked, with significant issues

    I run the risk of making this unreadable, but I will throw my comments in below Dave's, where needed, but in ALL CAPS. "1. Relative area and height are always 1.0 for all peaks in the cluster center. This is the case no matter which clustering algorithm you use. I'm not entirely sure what the correct calculation would be to get the relative area and height or whether there is one."...

    2009-02-13 15:22:08 UTC in EDAM ENCHILADA

About Me

  • 2009-01-21 (10 months ago)
  • 2370061
  • dgross25 (My Site)
  • Deborah Gross

Send me a message