Nhungamiro yehunyanzvi

Lasso uye Elastic Net Regression

Lasso regression inowedzera chirango cheL1 chinogona kudzikisa mamwe macoefficients akakwana kusvika zero, ichigadzira sparse linear modhi.

Teerera kugwaro iriAI-yakarondedzerwa · ~3 min

Pa peji ino4 min verenga
  1. Pfupiso
  2. Kudzika Kwakadzika
  3. Kuitwa muNyika yeChokwadi
  4. Ramangwana reLasso uye Elastic Net Regression
  5. Kuzvishandisa
  6. Ramba Uchiongorora
  7. Mibvunzo inowanzo bvunzwa

Pfupiso

Elastic net inosanganisa L1 neL2 zvirango, izvo zvinogona kubatsira kana mafungidziro akawanda kana akabatana, nepo nzira dzese dzichida kuyera nekusimbisa.

Kudzika Kwakadzika

Zvakajairika masikweya madiki anosarudza makobiri ekuderedza kukanganisa kwekufungidzira. Lasso inogadzirisa chinangwa ichocho nekuwedzera chirango chinoenzana nehwerengedzo yeabsolute coefficient magnitudes. Chirango chinokurudzira kudzikira, uye nekuti chirango chakazara-chako chine kona pa zero, mamwe macoefficients akaiswa anogona kuita zero chaizvo. Izvi zvinogadzira sparse modhi inogona kuve nyore kuchengeta kana kuongorora. Sparsity haisi humbowo hwekuti zvakachengetwa zvinofanotaura zvine causal kana zvakakosha zvakasiyana. Elastic net inosanganisa chirango cheL1 nechikara cheL2 chakapetwa. Yayo yekusanganisa paramende inodzora chiyero: iyo yakachena L1 endpoint is lasso, nepo iyo yakachena L2 endpoint iri ridge mune zvakajairika parameterizations. Iyo L2 chikamu chinogona kubatsira kugovera huremu pane zvakabatanidzwa kufanotaura pane kungochengeta imwe nekurasa zvimwe. Iyo yakazara simba parameter inodzora kuti yakawanda sei shrinkage inoiswa. Chaiyo software parameter magungano anosiyana, saka verenga raibhurari zvinyorwa. Chimbofunga imwe fungidziro modhi ine nhanho mbiri dzinoda kuchinjika dzekusvika. Lasso inogona kuchengetedza imwe uye zero imwe; shanduko diki kumuenzaniso wekudzidziswa dzinogona kushandura sarudzo iyoyo. Elastic net inogona kuchengetedza zvese neyakaderedzwa coefficients. Chero maitiro anogona kubatsira zvichienderana nekuti compactness, kufanotaura kugadzikana, kana kududzira ndicho chinangwa. Kana mhedzisiro yakayerwa pazvikero zvakasiyana zvakanyanya kana maficha anoshandisa mayuniti akasiyana, mhedzisiro yechirango haina kuenzana kunze kwekunge preprocessing yakakodzera. Sarudza marongero echirango uchishandisa kuchinjika-kusimbisa mukati meiyo data yekudzidziswa, uchichengeta bvunzo yekupedzisira isina kubatwa kusvika kusarudzwa kwemuenzaniso kwapera. Kuyera, kuteedzera uye kusarudzwa kwechimiro kunofanirwa kuiswa zvakaparadzana mukati mega rega rekudzidziswa kudzivirira kuvuza. Kune data rinoenderana nenguva, shandisa kupatsanurwa kunoremekedza kuverengwa kwenguva. Ongorora fungidziro yekuita uye ongorora kugadzikana kwecoefficient pane inonzwisisika resamples. Lasso uye elastic net maturusi enguva dzose mutsara kufanotaura; havagadzirise fomu risiri iro rinoshanda, data rakarerekera, kana kusawirirana pakati pehuwandu hwekudzidziswa uye chinangwa chekushandiswa.

Kuitwa muNyika yeChokwadi

A hypothetical nonprofit anofanotaura mari yezvipo kubva kune akawanda emushandirapamwe hunhu. Mushure mekuyera nhamba dzekufungidzira uye encoding zvikamu, kuchinjika-kusimbisa kunosarudza chirango chelasso; akati wandei macoefficients anova zero, kurerutsa iyo yakaiswa modhi yekuongorora.

Dhatabheti rinosanganisira zvakawanda zvine chekuita nemavara-anotorwa maficha. Lasso inogona kusarudza chimwe chimiro uye kudzvanya chimwe chinhu chakabatana, saka runyorwa rwakasarudzwa runogona kusiyanisa pamasampuri kunyangwe fungidziro yakafanana.

Muongorori anofananidza elastic-net kusanganisa marongero padyo ne zero uye imwe. Kurongeka pedyo ne zero kunosimbisa ridge-like shrinkage, nepo imwe inopindirana nekuguma kwe lasso; kushanda kunofananidzwa nenguva dzenguva dzakatarwa.

Chikwata chinoshandisa scikit-learn's Lasso mukati mepombi inomisa makoramu enhamba vachishandisa mapepa ekudzidzisa. Izvi zvinodzivirira ruzivo-kupeta ruzivo kubva mukufurira maparamita ekuyera.

Ramangwana reLasso uye Elastic Net Regression

Sparse modhi inogona kutsigira kuongororwa kana zvikwata zvinoda kunzwisisa kuti ndezvipi zvakarekodhwa zvinopesvedzera kufanotaura, asi coefficient zvinyorwa zvinofanirwa kuperekedzwa nekugadzikana uye macheki ekuita. Nhanho inoshanda inotevera ndeyekutaura kuti kangani chikamu chega chega chinosarudzwa pamasamples uye kuti chikanganiso chekufembera chinoshanduka sei pazvirango zvine musoro. Sezvo data masosi uye huwandu hwevanhu vanochinja, refita preprocessing uye ongorora zvakare maficha akasarudzwa. Makadhi emuenzaniso uye mapaipi anodhindwa anogona kurekodha kuyera, encoding, kupatsanura zano uye tuning sarudzo. Aya marekodhi anoita kuti ongororo iwedzere kongiri pasina kureva kuti compact linear modhi inotsanangura maitiro ese ari kumashure kwemhedzisiro.

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 Lasso and Elastic Net Regression 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 Lasso uye Elastic Net Regression?

Lasso regression inowedzera chirango cheL1 chinogona kudzikisa mamwe macoefficients akakwana kusvika zero, ichigadzira sparse linear modhi. Elastic net inosanganisa L1 neL2 zvirango, izvo zvinogona kubatsira kana mafungidziro akawanda kana akabatana, nepo nzira dzese dzichida kuyera nekusimbisa.

A lasso fit inopa chaiyo zero coefficient kune chimwe chinhu. Izvozvo zvinoratidzei mukati memuenzaniso uyu wakakodzera?

Zero yakarongedzerwa coefficient inoreva kuti chimiro hachipi chimwe chinhu kune iyi chaiyo mutsara kufanotaura pasi peiyo yakashongedzwa modhi.

Maviri akanyatso hukama akafanochinjana seayo anochengetwa nelasso mhiri kwemasamples. Izvi zvinoratidzei?

Lasso inogona kusarudza imwe yeakati wandei anoenderana fungidziro, ine yakasarudzwa chimiro ichichinja pasi pediki sampuli mutsauko.

Sei kuyera kuchifanira kudzidzwa mukati meimwe neimwe-yakasiyana-yekugonesa kudzidziswa kwakapetwa?

Kudzidza preprocessing pamitsara yese kunoita kuti ruzivo rwekusimbisa rwukanganise pombi yekudzidzisa uye inogona kuburitsa ruzivo.

Mumusangano wakajairwa musanganiswa-reshiyo, ndeipi magumo inoenderana nelasso yakachena?

Pamagumo eL1 akachena, elastic net inoderedza kune lasso; Mazita chaiwo eparameter anoenderana neraibhurari.

Dhatabheti rine zvirevo zvakayerwa mumadhora uye muzvikamu zvidiki. Sei scaling nyaya nokuda kurangwa regression?

Zvikamu zvakasiyana zvinoshandura coefficient magnitudes, saka chirango chisingaverengeki chinogona kubata vanofanotaura zvisina kuenzana.