Nhungamiro yehunyanzvi

Generalized Linear Models

Iyo generalized linear modhi (GLM) ine chekuita nekufanotaura kune mhedzisiro kuburikidza nemutsara wekufanotaura, kugovera mhinduro uye basa rekubatanidza.

Teerera kugwaro iriAI-yakarondedzerwa · ~3 min

Pa peji ino4 min verenga
  1. Pfupiso
  2. Kudzika Kwakadzika
  3. Kuitwa muNyika yeChokwadi
  4. Ramangwana reGeneralized Linear Models
  5. Kuzvishandisa
  6. Ramba Uchiongorora
  7. Mibvunzo inowanzo bvunzwa

Pfupiso

Iyi dhizaini inovhara zvakajairwa mutsara kudzokororwa, logistic regression uye Poisson regression nekushandura mhedzisiro mhuri uye mepu kubva kune zvinoreva kune vanofanotaura.

Kudzika Kwakadzika

A GLM ine zvidimbu zvitatu zvakabatana. Chikamu chisina kujairika chinosarudza mhinduro yekugovera kubva kune exponential-mhuri modhi, seGaussian, binomial, Poisson kana Gamma. Iyo yakarongeka chikamu inoumba mutsara wekufanotaura kubva kune zvinopinda, zvinowanzonyorwa eta = X beta. Basa rekubatanidza rinobatanidza mhinduro inotarisirwa mu kune iyo inofanotaura kuburikidza ne g(mu) = eta. Iyo modhi inoramba iine mutsara mune yayo coefficients pachiyero chekubatanidza, kunyangwe iyo inotarisirwa mhedzisiro pachayo isiri mutsara wakatwasuka mukufanotaura. Nekugovera kweGaussian uye chiziviso chinongedzo, chinomisikidzwa chinoreva mutsara wekufanotaura, kudzoreredza yakajairwa mutsara regression fomu. Binomial modhi ine logit yekubatanidza mhando log odds; kushandisa inverse logit kunopa mukana. Modhi yePoisson ine log link modhi iyo logarithm yeinotarisirwa kuverenga, saka zvayakafanotaurwa zvinoreva zvakanaka. Aya ma link anokodha zvipingaidzo uye hukama hwakakodzera kune akasiyana mhedzisiro mhando. Kune yekufungidzira mhedzisiro yebhanari, fungidzira mutsara wemodhi yekufungidzira i zero. Iyo inverse logit mepu zero kune zvingangoita 0.5. Kana iyo mutsara inofanotaura ichikwira, mukana unokwira asi unoramba uri pasi peumwe. Kusiyana neizvi, identity-link modhi inogona kufanotaura mhinduro yebhinari pazasi zero kana pamusoro peimwe, ndosaka yakajairika Gaussian mutsara modhi inogona kusaenderana nebasa iroro. GLM haivimbise kuti mhuri yakasarudzwa ndeyechokwadi; inoita kuti fungidziro dzijeke kuitira kuti dziongororwe. Mhuri inosarudza hukama-kusiyana-siyana pamwe nerutsigiro rwemhinduro. Poisson modhi dzinowanzo fungidzira musiyano wezvimiso zvakaenzana nechirevo chechimiro, mamiriro ayo chaiwo kuverenga data anogona kutyora. Binomial modhi inofanirwa kuratidza kuti miedzo nebudiriro zvinomiririrwa sei. Sarudzo dzekubatanidza dzinogona kumanikidzwa nemhuri uye software kuita. Ongorora zvakasara kana kutsauka kwekuongorora, kusava nechokwadi, kuenzanisa pazvinenge zvakakodzera, uye kuita-kunze kwekuita. Iyo GLM label haireve kuti marudzi ese emhedzisiro anogovera imwe kukanganisa kugovera kana dudziro imwechete yemakofi.

Kuitwa muNyika yeChokwadi

Muongorori wedzimba anoshandisa mhuri yeGaussian ine chitupa chinongedzo kumuenzaniso wepakati mutengo wekutengesa; iyo yakafanotaurwa yemamiriro ezvinhu ndiyo inofanotaura mutsara pachayo.

Yekufungidzira sevhisi timu inoenzanisira kana nyaya yagadziriswa muzuva rimwe chete nemhuri yebinomial uye logit link. The linear predictor inotsanangura log odds, iyo inverse inobatanidza mamepu kune mukana uri pakati paziro neimwe.

Anoronga zvekufambisa anofanotaura zvisingatarisirwi chiitiko chinotarisirwa kuverenga nemhuri yePoisson uye log link. Kuratidzira mutsara wekufungidzira kunopa yakanaka inotarisirwa kuverenga, nepo nguva yekuratidzwa inogona kumiririrwa kuburikidza neyakagadziriswa kana zvakakodzera.

Mutsvagiri anoenzanisira zvakanaka, zvakatsvetwa-kurudyi mari uchishandisa mhuri yeGamma uye chinongedzo chakakodzera. Ivo vanotarisa kana iyo mhuri yekupindura uye mutsauko chimiro chinoenderana nekuyera maitiro pane kusarudza GLM chete nekuti chinangwa chiri chakanaka.

Ramangwana reGeneralized Linear Models

GLM mafambiro ekushanda anogona kujeka kana mishumo ichiratidza mhuri, chinongedzo, zvikamu zvekupindura uye dudziro inokwana pamwe chete. Vaongorori vanogona kuenzanisa mhuri dzinonzwisisika pane data rinoenderana nehuwandu hwevanhu uye sarudzo, uchitarisa calibration kana mapeteni akasara anoenderana nemhedzisiro. Kana kusiyanisa kwekuverenga kuchipfuura fungidziro yePoisson modhi, inegative-binomial kana imwe modhi inogona kufanirwa kuongororwa. Sezvo maapplication achishanduka, zvinyorwa zvinofanirwa kuchengetedza zvigadziriso, uremu uye preprocessing padivi peakaiswa coefficients. Kushandisa zvirinani kunogona kubatsira macheki iwayo, asi sarudzo yemhuri inoramba iri sarudzo yemuenzaniso yakavakirwa pamaitirwo akaitwa zvakaonekwa.

Kuzvishandisa

Nei zvichikosha

  • Mutengo uye bhajeti. Zvisarudzo zvezvivakwa zvinotyaira kuita uye mutengo wekushandisa kwemakore.
  • Sarudzo dzakajeka. Dzidzo yehunyanzvi inobatsira zvikwata kusarudza murwi wakakodzera, kwete iwo mutsva chete.
  • Kudzora kwemhando yepamusoro. Sarudzo dzeinjiniya dziri nani dzinoderedza zviitiko zvekuvimbika mukugadzira.

Chenjera

  • Kugadzirisa imwe bhenji kunogona kuvanza yakafara system kushaya simba.
  • Infrastructure uye mari yekugadzirisa inowanzotarisirwa pasi.
  • Chengetedzo uye kucherechedzwa mapundu anogona kukura sezvo masisitimu anowedzera kuoma.

Matanho anotevera

  1. Tsanangura latency, mhando, uye mutengo zvinangwa usati waitwa.
  2. Benchmark pasi pechokwadi mutoro uye data mamiriro.
  3. Chishandiso chekutarisa zvikanganiso, kudonha, uye mushandisi maitiro.
  4. Gadzirira nzira dzekudzosera kumashure uye dzezviitiko usati wawedzera.

Ramba Uchiongorora

Tsamba yemahara

Tora iyo yemazuva ese AI pfupiso

Nhatu dzakasimbiswa dzeAI nyaya mangwanani ega ega pakati pevhiki, dzakanyorwa neChirungu chakajeka. Yemahara zvachose, hapana ads.

Imwe email zuva rega rega revhiki. Bvisa kunyoresa kamwe chete. Hatimbotengese kana kugovera kero yako.

Zviedze iwe pachako

Take the Generalized Linear Models quiz

Mhinduro yepakarepo pamhinduro yega yega, uye chitupa chinogovaniswa chine ID inobatika kana wapedza kosi.

Tanga mibvunzo

Tsigira dzidzo yemahara yeAI. AI Understanding ndeye 501(c)(3) isingabatsiri - hapana ads, hapana paywall, zvachose. Ita chipo

Mibvunzo inowanzo bvunzwa

Chii chinonzi Generalized Linear Models?

Iyo generalized linear modhi (GLM) ine chekuita nekufanotaura kune mhedzisiro kuburikidza nemutsara wekufanotaura, kugovera mhinduro uye basa rekubatanidza. Iyi dhizaini inovhara zvakajairwa mutsara kudzokororwa, logistic regression uye Poisson regression nekushandura mhedzisiro mhuri uye mepu kubva kune zvinoreva kune vanofanotaura.

Ndezvipi zvinhu zvitatu zvinotsanangura iyo yakakosha GLM chimiro chinotsanangurwa pano?

Iyo GLM inosanganisa mhuri inopindura, chirevo chekufungidzira uye chinongedzo chinobatanidza zvinoreva kune icho chinofanotaura.

MuGaussian GLM ine chitupa chinongedzo, zvinorehwa zvinorevei kune anofanotaura?

Iyo identity link inosiya zvinoreva pachiyero chakafanana neanofanotaura mutsara.

Binomial GLM inoshandisa chinongedzo. Ndechipi chikero chinotsanangurwa nemutsetse wayo?

Iyo logit link inoburitsa mukana wekuroga odds; iyo inverse link mepu kudzokera kune zvingangoita.

Sei iyo Poisson GLM ine log link kuburitsa zvakanaka zvinotarisirwa kuverenga?

Iyo inverse yelog link ibasa reexponential, rine maitiro akanaka.

Mhuri yemhinduro inotsanangudza chii kunze kwezvinotenderwa nemhedzisiro?

Mhuri inosarudza maitiro ekugovera kusanganisira hukama pakati pezvinorehwa zvinoreva uye musiyano.