Rewrite MATLAB functions with matlab.io.xml.dom#347
Closed
Jerry1144 wants to merge 5 commits into
Closed
Conversation
Circumvent Java heap size problems with recent MATLAB dealing with large XML. MATLAB's `io.xml.xpath.*` appears to execute eagerly and over-consumes memory Walking the dom tree ourselves is slower than native `readtable()` when every sub-node are consolidated into one big table, but the latter cannot easily handle "one edge table per track" which is required by trackmateEdges.m
Migrate from Java-based `xmlread()` to MATLAB's native `matlab.io.xml.dom.*`. Minor style change: use 1-based node() instead of 0-based item(), and access properties directly instead of getProperty().
Revert to querying the nodes with getAttribute(). Strangely, copying the attributes out with getAttributes(), and querying that copy, is noticeably slower than just repeatedly node.getAttribute().
The correct way to attach the caught ME is error(ME.identifier, ...)
Fixed the missing break-condition check when inspecting the first Spot node.
Author
|
Apparently |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Rewrite with matlab.io.xml.dom.* facilities, avoiding Java heap size problems. Available since R2021a. One path to #343. Force-push of #346
A detour from #345 -
trackmateEdgesrequires one table per track node, but it isn't possible to express a parent node's attribute in XPath 1.0, and callingreadtable()once per track is slow (50+h from my 16k track file).Then, to keep the total memory use in check, all function in the series must adhere to a single path. Apparently memory reuse does not happen across
readstruct,readtable, andmatlab.io.xml.*, and GC only occurs during inactivity. Thus the total rewrite.It's possible to mix readtable's
trackmateSpots.minto this. Used alone it is 5x faster than walking the dom tree ourselves, but when called bytrackmateGraph.mit doubles temp memory usage, and slows things down when memory pressure kicks in.