ceari / edacc_web

Experiment Design and Analysis for Computer Clusters: Web frontend
http://edacc.github.io
MIT License
3 stars 3 forks source link

instance name duplicates #4

Closed MariusSchneider closed 12 years ago

MariusSchneider commented 12 years ago

Hi,

there are some duplicates of instance names on the overview of the runtimes and on the corresponding CSV export. I cannot distinguish between these instances.

For example, three times: SAT_dat.k90.debugged.cnf on

http://edacc2.informatik.uni-ulm.de/SATChallenge2012/experiment/18/results/?cost=resultTime&display_measure=par10&solver_configs=679&i=679&i=4351&i=4352&i=2677&i=2675&i=4329&i=4328&i=4332&i=4327&i=532&i=534&i=533&i=2685&i=4330&i=4333&i=2681&i=2691&i=4331&i=2666&i=2670&i=4326&i=528&i=529&i=2667&i=2672&i=4325&i=4464&i=4471&i=4426&i=4474&i=4449&i=4458&i=4444&i=4425&i=4465&i=4441&i=4445&i=4455&i=4477&i=4473&i=4457&i=4448&i=4480&i=4432&i=4460&i=4481&i=4443&i=4442&i=4456&i=4462&i=4447&i=4452&i=4440&i=4431&i=4467&i=4438&i=4478&i=4427&i=4461&i=4459&i=4434&i=4450&i=4466&i=4451&i=4479&i=4436&i=4475&i=4435&i=4472&i=4453&i=4470&i=4433&i=4429&i=4468&i=4482&i=4469&i=4463&i=4439&i=4428&i=4446&i=4483&i=4430&i=4476&i=4454&i=4437&i=4422&i=4424&i=4421&i=4423&i=4493&i=4496&i=4495&i=4497&i=4492&i=4498&i=4494&i=4484&i=4488&i=4485&i=4491&i=4487&i=4490&i=4489&i=4486&i=3436&i=603&i=589&i=599&i=584&i=585&i=590&i=2646&i=2648&i=2635&i=2638&i=2636&i=2649&i=2651&i=2637&i=2579&i=503&i=2584&i=2585&i=504&i=2580&i=2589&i=2582&i=2583&i=2586&i=2581&i=2587&i=2578&i=505&i=2588&i=2654&i=2652&i=2655&i=521&i=4317&i=2627&i=4322&i=522&i=2626&i=4323&i=2625&i=4320&i=524&i=519&i=516&i=4318&i=4319&i=4321&i=518&i=523&i=2634&i=2632&i=2633&i=4324&i=520&i=2617&i=2629&i=4316&i=2630&i=2616&i=2618&i=2619&i=512&i=2631&i=2623&i=511&i=2592&i=2598&i=2614&i=2597&i=509&i=2599&i=510&i=2605&i=4375&i=754&i=4376&i=752&i=747&i=4373&i=750&i=751&i=753&i=4374&i=755&i=749&i=745&i=741&i=723&i=742&i=737&i=4372&i=738&i=731&i=622&i=621&i=620&i=625&i=619&i=623&i=616&i=627&i=617&i=624&i=615&i=660&i=643&i=655&i=653&i=646&i=657&i=4348&i=651&i=641&i=4345&i=4339&i=4342&i=4335&i=629&i=4337&i=4334&i=4341&i=4343&i=639&i=4344&i=630&i=4338&i=4346&i=633&i=4347&i=4336&i=628&i=637&i=635&i=632&i=634&i=636&i=4340&i=631&i=662&i=669&i=4349&i=4350&i=668&i=674&i=666&i=667&i=663&i=664&i=608&i=614&i=607&i=611&i=613&i=612&i=609&i=610&i=606&i=694&i=4362&i=4355&i=4353&i=709&i=692&i=719&i=696&i=4367&i=706&i=713&i=4358&i=4360&i=4368&i=4371&i=4364&i=4356&i=4369&i=697&i=4354&i=712&i=700&i=718&i=4366&i=4370&i=4357&i=4359&i=711&i=4361&i=707&i=716&i=720&i=4365&i=4363&i=699&i=4528&i=4542&i=4532&i=4531&i=4548&i=4547&i=4549&i=4546&i=4556&i=4526&i=4530&i=4540&i=4533&i=4846&i=4847&i=4862&i=4845&i=4861&i=4858&i=4557&i=4558&i=4559&i=4843&i=4842&i=4844&i=4859&i=4848&i=4863&i=4860&i=4838&i=4840&i=4841&i=4839&i=4508&i=4509&i=4520&i=4518&i=4519&i=4517&i=4855&i=4854&i=4856&i=4857&i=4503&i=4506&i=4507&i=4511&i=4501&i=4502&i=4852&i=4851&i=4850&i=4849&i=4514&i=4513&i=4512&i=4853&i=546&i=545&i=544&i=3450&i=1809&i=595&i=593&i=594&i=1807&i=1808&i=1806&i=1839&i=596&i=1840&i=1844&i=1833&i=1831&i=1830&i=1829&i=1823&i=1826&i=1893&i=1890&i=1889&i=1878&i=1877&i=1882&i=1885&i=547&i=600&i=1868&i=1870&i=1873&i=1872&i=1871&i=1869&i=602&i=563&i=562&i=561&i=550&i=551&i=578&i=576&i=579&i=760&i=558&i=553&i=554&i=575&i=1780&i=1782&i=588&i=1788&i=1785&i=587&i=1787&i=1790&i=1786&i=1789&i=1791&i=1793&i=1798&i=1795&i=1800&i=1799&i=567&i=564&i=1913&i=1911&i=1912&i=1915&i=1910&i=1909&i=1914&i=1908&i=1917&i=1918&i=1920&i=1919&i=1916&i=761&i=762&i=764&i=763&i=4837&i=4836&i=4835&i=4834&i=190&i=130&i=173&i=154&i=153&i=108&i=166&i=172&i=177&i=165&i=183&i=193&i=101&i=136&i=191&i=152&i=170&i=187&i=184&i=143&i=118&i=157&i=175&i=156&i=106&i=146&i=128&i=185&i=196&i=124&i=140&i=188&i=192&i=137&i=150&i=103&i=138&i=186&i=104&i=134&i=114&i=144&i=162&i=127&i=70&i=67&i=68&i=66&i=64&i=65&i=59&i=57&i=63&i=61&i=62&i=56&i=58&i=60&i=93&i=91&i=92&i=89&i=47&i=48&i=50&i=45&i=49&i=34&i=35&i=33&i=36&i=41&i=44&i=43&i=42&i=54&i=55&i=53&i=39&i=38&i=22&i=23&i=24&i=25&i=20&i=21&i=18&i=19&i=7&i=29&i=4415&i=4414&i=4552&i=4553&i=4554&i=4416&i=4555&i=4560&i=4561&i=4562&i=4412&i=4399&i=4401&i=4409&i=4381&i=4384&i=4385&i=4411&i=4404&i=4394&i=4407&i=4378&i=4391&i=4386&i=4405&i=4379&i=4408&i=4396&i=4392&i=4383&i=4398&i=4377&i=4400&i=4410&i=4387&i=4393&i=4390&i=4389&i=4406&i=4380&i=4397&i=4403&i=4388&i=4395&i=4402&i=4382&i=1865&i=1861&i=1859&i=1864&i=1863&i=1804&i=1906&i=1905&i=1903&i=1904&i=2704&i=2699&i=2702&i=2705&i=2693&i=2694&i=768&i=765&i=766&i=767&i=4833&instance_filter=&submit=Show

It is better on the page (but please without "..") http://edacc2.informatik.uni-ulm.de/SATChallenge2012/experiment/18/results-by-instance?instance=4861

Please provide a way to distinguish the instances.

Cheers, Marius

ceari commented 12 years ago

On the results "by solver config & instance" and the "by instance" pages I also show the instance MD5 now. In the instance selection boxes it shows the full internal instance path now.

Let me know if you have suggestions on how to improve it.

MariusSchneider commented 12 years ago

I would prefer the full path (like you did at the "by instance" page) also on the "by solver config & instance" page, or at least as an option for the CSV export. Would it be possible?

ceari commented 12 years ago

It always uses the full instance path in the exported CSV now. Should be easy enough to remove the path with some script if someone only needs the instance name.

I'd like to keep the table on the page rather slim so it's easier to read for humans. Using the full path there would mean that either the column is either very high or wide, which wouldn't look so good I think.

MariusSchneider commented 12 years ago

ok. thank you very much