pavel/fedavg_mpi
1
0 Fork 0
Code Issues Pull Requests Releases Wiki Activity
42 Commits 1 Branch 0 Tags
master
Commit Graph
42 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
pavel 3e850713ca final fixes on the report 2020-08-05 22:12:57 +02:00
pavel 27a1beff70 purged the visualizer 2019-12-15 22:17:46 -08:00
pavel 24ca380cbf even nicer plots can you beleive it 2019-12-15 21:05:11 -08:00
pavel 670f69e0df ya sdelal' 2019-12-15 19:51:30 -08:00
pavel 05480606b0 small refactoring and kicked the visualizer again 2019-12-15 09:51:22 -08:00
pavel 06d0b2d565 added some figures to the docs 2019-12-15 09:29:01 -08:00
pavel 086b021f26 end of report reached, yay 2019-12-14 22:30:41 -08:00
pavel cbe62bae02 wrote some reports, broke some code 
as usual, u know
 2019-12-14 19:39:33 -08:00
pavel aac9aed6e7 suppressed the FutureWarnings 2019-12-14 10:47:28 -08:00
pavel 79e8b9874b for some reason I dockerized it and it works 2019-12-13 17:35:06 -08:00
pavel 4a9e328884 visualizer kinda working, more work needs done 2019-12-13 11:05:38 -08:00
pavel 1338df9db2 very nice scalability curve up to 12 procs 
parameters: wiki, BPE 250, BS 32, TARGET 8.3, the rest is commited

also it might be that more pipelines would work if there were
MANY more pipelines, but that remains to be seen until i buy
myself a Ryzen 3700x and destroy the world with it i guess i
should go to sleep now
 2019-12-12 22:26:50 -08:00
pavel 2bbb1d243c stop refactoring and get some stuff huggin' done 2019-12-12 21:10:41 -08:00
pavel 966bbc904c mainly faster startup thanks to 'lazy' loading 2019-12-12 20:02:03 -08:00
pavel d954a9832b somehow i managed to improve the build system 
to become even more awesome
 2019-12-12 10:17:17 -08:00
pavel 0372f0ee9c using adam seems like a sane idea 2019-12-11 22:44:18 -08:00
pavel 598e59ca2e parallelizing doesn't bring anything 2019-12-11 22:23:11 -08:00
pavel 2e5042f0e3 termination seems to be implemented okayish 2019-12-11 15:29:59 -08:00
pavel 136e5938c3 side-quest successful, go eat mom's spaghetti 
the side-quest was to flatten the pipeline so that the only
branching is allowed after batcher
 2019-12-11 11:19:56 -08:00
pavel 32742059c7 commit working termination code for two nodes 
before embarking on yet another side-quest
 2019-12-11 11:12:32 -08:00
pavel 7043b65532 this is the baseline for evaluation 2019-12-11 10:31:16 -08:00
pavel 5abe7bb413 a lot of small stuff 2019-12-07 19:56:23 -08:00
pavel 997d3a7618 the pipeline kinda works, you can even demo it 2019-12-01 22:04:57 -08:00
pavel b2a4681fb2 implemented some kind of dispatching 2019-12-01 17:49:09 -08:00
pavel bc6d34e253 now it kinda learns again and code is kinda clean 2019-12-01 15:44:34 -08:00
pavel 5d14171631 for some two other files weren't added to commit 2019-12-01 15:05:48 -08:00
pavel a2ef842ef8 take a more windowed approach 2019-12-01 15:04:59 -08:00
pavel 4bf66bf85e network trains and tests but slow? -- investigate 2019-12-01 11:26:54 -08:00
pavel b115077c3b now slave correctly gets data 2019-12-01 10:21:58 -08:00
pavel 569bbf7397 tokenize FASTER 2019-12-01 09:44:40 -08:00
pavel 7409eca38b added batcher and even made it twice as fast 
as it was a minute ago
 2019-11-30 22:40:23 -08:00
pavel 101248965c tokenizer and batcher are sane and work 
i wonder how much pain it's gonna cause me to change
something in the code
 2019-11-30 22:16:42 -08:00
pavel 04c35ed9b6 after literally hours of pain, keras learns mnist 2019-11-29 22:10:25 -08:00
pavel 87208d599d switched to meson, let's see if it's gonna bite me 2019-11-28 16:04:00 -08:00
pavel c4a21b6be8 reached some mildly amusing result 
of slaves having "independent" data sources and master still
training through them
 2019-11-27 17:37:40 -08:00
pavel 5af90383f0 commiting halfways because i'm feeling stuck 2019-11-26 22:11:54 -08:00
pavel 9727499f4d set up synchronous comm for many slaves 
and also abusing the c preprocessor
 2019-11-26 21:20:40 -08:00
pavel 725130e51f implemented a very smooth frankenstein function 2019-11-25 22:37:34 -08:00
pavel 76f8d7dcb6 a great work has been done here 
now there is no mentioning python in c code (except for the
boilerplate at the beginning, but the rest is nice clean c).
all the bridging is being done in cython (where it belongs).
and no memory leaks so there's that!
 2019-11-25 22:01:39 -08:00
pavel 253c833b5b commit small improvements before going craZy 2019-11-25 20:54:27 -08:00
pavel 04e0b9829c well done sending a network over mpi, pat pat 
now do a lot of nets in parallel and then we'll talk
 2019-11-25 20:23:33 -08:00
pavel a80a3de4fa it started working so commit it 2019-11-24 22:14:26 -08:00