Tubegen/bugs and feature requests - Revision history

Nathan Malkin at 18:35, 16 August 2010

2010-08-16T18:35:54Z

← Older revision		Revision as of 18:35, 16 August 2010
Line 35:		Line 35:
	''Anecdotal evidence:'' I ran Tubegen using the -OUTSF argument to output the seeds file it generated. Then I ran it (twice) using -USESF 1 and -SF to integrate streamlines based on the (previously generated) seed points (and then cull them). While the output from the second set of calls was the same, it was different from the output of the original run. Not sure why this is happening, but it probably shouldn't be.		''Anecdotal evidence:'' I ran Tubegen using the -OUTSF argument to output the seeds file it generated. Then I ran it (twice) using -USESF 1 and -SF to integrate streamlines based on the (previously generated) seed points (and then cull them). While the output from the second set of calls was the same, it was different from the output of the original run. Not sure why this is happening, but it probably shouldn't be.
	--- [[User:Nathan Malkin\|Nathan Malkin]] 12:39, 16 August 2010 (EDT)		--- [[User:Nathan Malkin\|Nathan Malkin]] 12:39, 16 August 2010 (EDT)

			=== Documentation flaws===
			Though listed under "Integration Parameters," the argument ''-dth'' is used for the T<sub>t</sub> value described in the Zhang paper (distance cut-off value). Similarly, the ''-Longshort'' argument is used for determining how to compare curves when finding the distance between them. If it is set to anything other than 0 (for example, the recommended value in the wiki [[Tubegen#Current_Values\|is 1]]), the invariant described in the paper (distance is from shorter to longer trajectory) is violated.
			--- [[User:Nathan Malkin\|Nathan Malkin]] 14:35, 16 August 2010 (EDT)

	== Feature Requests ==		== Feature Requests ==

Nathan Malkin: Tubegen bug report

2010-08-16T16:39:07Z

Tubegen bug report

@@ Line 23: / Line 23: @@
 === RK4 ===
 Using <tt>-ODE 1</tt> to specify fourth-order Runge-Kutta seems to break some intrinsic culling metric; all tubes get preemptively culled. --- [[User:Jadrian Miles|Jadrian Miles]] 17:08, 24 March 2009 (UTC)
 == Feature Requests ==

Jadrian Miles at 17:08, 24 March 2009

2009-03-24T17:08:04Z

← Older revision		Revision as of 17:08, 24 March 2009
Line 1:		Line 1:
	== Slowdown ==		== Bugs ==
			=== Slowdown ===
	Streamtube generation gets slower and slower as the program runs. What's going on? Is it running out of memory, reading inefficiently through a big file, doing something probabilistic that gets less likely the more tubes there are? We need a speedup!		Streamtube generation gets slower and slower as the program runs. What's going on? Is it running out of memory, reading inefficiently through a big file, doing something probabilistic that gets less likely the more tubes there are? We need a speedup!

Line 19:		Line 20:
	* <code>void jot_tube_mesh_generator::generate()</code>		* <code>void jot_tube_mesh_generator::generate()</code>
	These obviously do different things but there seems to be a lot of repeated code among them that should be factored out. Some of them are truly enormous functions, too, and should be broken apart for clarity.		These obviously do different things but there seems to be a lot of repeated code among them that should be factored out. Some of them are truly enormous functions, too, and should be broken apart for clarity.

			=== RK4 ===
			Using <tt>-ODE 1</tt> to specify fourth-order Runge-Kutta seems to break some intrinsic culling metric; all tubes get preemptively culled. --- [[User:Jadrian Miles\|Jadrian Miles]] 17:08, 24 March 2009 (UTC)

			== Feature Requests ==
			=== Adaptive Step Size ===
			dhl thinks it would be a good idea to change the advection algorithm to use an adaptive, rather than fixed, step size.

	[[Category:Diffusion MRI]][[Category:Bug Tracking Pages]]		[[Category:Diffusion MRI]][[Category:Bug Tracking Pages]]

Jadrian Miles: New page: == Slowdown == Streamtube generation gets slower and slower as the program runs. What's going on? Is it running out of memory, reading inefficiently through a big file, doing something p...

2009-03-24T14:22:22Z

New page: == Slowdown == Streamtube generation gets slower and slower as the program runs. What's going on? Is it running out of memory, reading inefficiently through a big file, doing something p...

New page

== Slowdown ==
Streamtube generation gets slower and slower as the program runs. What's going on? Is it running out of memory, reading inefficiently through a big file, doing something probabilistic that gets less likely the more tubes there are? We need a speedup!

:In <tt>streamtube.cpp</tt>, function <code>tube_collection_generator::generateSpacing()</code>, unless <code>getCullingDistance() == 0.0</code> (set this by passing <code>-cd 0.0</code> to <tt>[[tubegen]]</tt>), a pairwise inter-tube distance-based culling check happens after each streamtube is generated, for a total complexity of O(n<sup>2</sup>).

It would be a good idea to separate out seeding, streamline/tube generation, and culling, since each of these can be modified or optimized separately. There's a lot of code repetition in <tt>streamtube.cpp</tt> --- . Also want to figure out the function of <code>tube_collection_generator::getDistThreshold()</code> (<code>-dth</code> argument to <tt>tubegen</tt>) in <code>streamtube::avgdistance()</code>.

Just in <tt>streamtube.cpp</tt>, there's:
* <code>void brown_streamtube::generate(double seed_point[3])</code>
* <code>void MC_streamtube::generate(double seed_point[3])</code>
* <code>void cross_fiber::generate(double seed_point[3])</code>
* <code>void tube_collection_generator::generate2()</code>
* <code>void tube_collection_generator::generatecross()</code>
* <code>void tube_collection_generator::generatecrossradii()</code>
* <code>void tube_collection_generator::generatehalfcircle()</code>
* <code>streamtube* tube_collection_generator::generateOneTube(double p[3])</code>
* <code>void tube_collection_generator::generate(int MC)</code>
* <code>void tube_collection_generator::generateSpacing(double sspace[3], int MC)</code>
* <code>void jot_tube_mesh_generator::generate()</code>
These obviously do different things but there seems to be a lot of repeated code among them that should be factored out. Some of them are truly enormous functions, too, and should be broken apart for clarity.

[[Category:Diffusion MRI]][[Category:Bug Tracking Pages]]