NonlinearModelFit[]<\/a> command. So, he created a completely noise free dataset directly from the model itself<\/p>\ndata = Table[{t, dPWO2[t, 33, 25, 55]}, {t, 0, 6, 6.\/75}];<\/pre>\nand then, pretending that he didn’t know the parameters that formed that dataset, he asked NonlinearModelFit[] to find them:<\/p>\n
fittedModel = NonlinearModelFit[data, dPWO2[t, v, qL, q], {v, qL, q}, t]; \/\/ Timing\r\nparams = fittedModel[\"BestFitParameters\"]<\/pre>\nOn my machine this calculation completes in 11.21 seconds and gives the result we expect: <\/p>\n
\r\n{11.21, Null} \r\n{v -> 33., qL-> 25., q -> 55.} <\/pre>\n 11.21 seconds is too slow though since he wants to evaluate this on real data many hundreds of times. How can we make it any faster?<\/p>\n
I started off by trying to make NIntegrate go faster since this will be evaluated potentially hundreds of times by the optimizer. Turning off Symbolic pre-processing did the trick nicely in this case. To turn off Symbolic pre-processing you just add <\/p>\n
Method -> {Automatic, \"SymbolicProcessing\" -> 0}<\/pre>\n to the end of the NIntegrate command. So, we re-define his model function as<\/p>\n
\r\ndPWO2[t_?NumericQ, v_?NumericQ, qL_?NumericQ, q_?NumericQ] := \r\n PBH2O*((v\/(v + qL))*dFIO2*(1 - E^((-(v + qL))*t)) + \r\n q*NIntegrate[E^((v + qL)*(\\[Tau] - t))*dCVO2[\\[Tau]], {\\[Tau], 0, t}\r\n,Method -> {Automatic, \"SymbolicProcessing\" -> 0}])\r\n<\/pre>\nThis speeds things up almost by a factor of 2.<\/p>\n
fittedModel = NonlinearModelFit[data, dPWO2[t, v, qL, q], {v, qL, q}, t]; \/\/ Timing\r\nparams = fittedModel[\"BestFitParameters\"]\r\n\r\n{6.13, Null}\r\n{v -> 33., qL -> 25., q -> 55.}\r\n<\/pre>\nNot bad but can we do better?<\/p>\n
Giving Mathematica a helping hand.<\/h2>\n
Like most optimizers, NonlinearModelFit[] will work a lot faster if you give it a decent starting guess. For example if we provide a close-ish guess at the starting parameters then we get a little more speed <\/p>\n
\r\nfittedModel = \r\n NonlinearModelFit[data, \r\n dPWO2[t, v, qL, q], {{v, 25}, {qL, 20}, {q, 40}}, \r\n t]; \/\/ Timing\r\nparams = fittedModel[\"BestFitParameters\"]\r\n\r\n{5.78, Null}\r\n{v -> 33., qL -> 25., q -> 55.}\r\n<\/pre>\ngive an even better starting guess and we get yet more speed<\/p>\n
\r\nfittedModel = \r\n NonlinearModelFit[data, \r\n dPWO2[t, v, qL, q], {{v, 31}, {qL, 24}, {q, 54}}, \r\n t]; \/\/ Timing\r\nparams = fittedModel[\"BestFitParameters\"]\r\n\r\n{4.27, Null}\r\n{v -> 33., qL -> 25., q -> 55.}\r\n<\/pre>\nAsk the audience<\/h2>\n
So, that’s where we are up to so far. Neither of us have much experience with this particular part of Mathematica so we are hoping that someone out there can speed this calculation up even further. Comments are welcomed.<\/p>\n","protected":false},"excerpt":{"rendered":"
A Mathematica user recently asked me if I could help him speed up the following Mathematica code. He has some data and he wants to model it using the following model dFIO2 = 0.8; PBH2O = 732; dCVO2[t_] := 0.017 Exp[-4 Exp[-3 t]]; dPWO2[t_?NumericQ, v_?NumericQ, qL_?NumericQ, q_?NumericQ] := PBH2O*((v\/(v + qL))*dFIO2*(1 – E^((-(v + qL))*t)) […]<\/p>\n","protected":false},"author":3,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"jetpack_post_was_ever_published":false,"footnotes":"","jetpack_publicize_message":"","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":false,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2}},"categories":[4,8,7],"tags":[],"class_list":["post-2416","post","type-post","status-publish","format-standard","hentry","category-math-software","category-mathematica","category-programming"],"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_shortlink":"https:\/\/wp.me\/p3swhs-CY","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/walkingrandomly.com\/index.php?rest_route=\/wp\/v2\/posts\/2416","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/walkingrandomly.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/walkingrandomly.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/walkingrandomly.com\/index.php?rest_route=\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/walkingrandomly.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=2416"}],"version-history":[{"count":29,"href":"https:\/\/walkingrandomly.com\/index.php?rest_route=\/wp\/v2\/posts\/2416\/revisions"}],"predecessor-version":[{"id":2445,"href":"https:\/\/walkingrandomly.com\/index.php?rest_route=\/wp\/v2\/posts\/2416\/revisions\/2445"}],"wp:attachment":[{"href":"https:\/\/walkingrandomly.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=2416"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/walkingrandomly.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=2416"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/walkingrandomly.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=2416"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}