Keras 2.3 Docs
Iqukethe izikhangisoUkuthenga ngaphakathi nohlelo
1K+
Okudawunilodiwe
Wonke umuntu
info
Mayelana nalolu hlelo lokusebenza
I-Keras 2.3 Imibhalo
Ukufunda ngokujulile kwabantu.
IKeras iyi-API eyenzelwe abantu, hhayi imishini. UKeras ulandela izindlela ezinhle kakhulu zokunciphisa umthwalo wokuqonda: unikeza ama-API angaguquki nalula, unciphisa inani lezenzo zomsebenzisi ezidingekayo kumacala wokusetshenziswa okujwayelekile, futhi linikeza nemiyalezo yephutha ecacile nenokwenzeka. Iphinde ibe nemibhalo ebanzi nemihlahlandlela yonjiniyela.
Iterate ngejubane lomcabango.
UKeras uhlaka olusetshenziswa kakhulu lokufunda phakathi kwamaqembu aphumelela i-5 kuKaggle. Ngoba iKeras yenza kube lula ukuqhuba izivivinyo ezintsha, ikunikeza amandla okuzama imibono eminingi kunomncintiswano wakho, ngokushesha. Futhi unqoba kanjani.
Ukufunda ngomshini we-Exascale.
Yakhelwe ngaphezulu kweTensorFlow 2.0, iKeras uhlaka lwamandla ezimboni olungafinyelela kumaqoqo amakhulu we-GPUs noma yonke i-TPU pod. Akunakwenzeka kuphela; kulula.
Hambisa noma kuphi.
Thatha ithuba lamakhono wokuphakwa ngokugcwele kwepulatifomu yeTensorFlow. Ungathumela amamodeli weKeras kwiJavaScript ukuze usebenze ngokuqondile kwisiphequluli, ku-TF Lite ukuze usebenze ku-iOS, i-Android namadivayisi ashumekiwe. Kulula futhi ukusebenzisa amamodeli weKeras njenge-API yewebhu.
Isistimu enkulu yemvelo.
IKeras iyingxenye ephakathi ye-TensorFlow 2.0 ecosystem exhunywe ngokusezingeni eliphezulu, emboza zonke izinyathelo zokuhamba komsebenzi wokufunda komshini, kusukela ekuphathweni kwedatha kuya ekuqeqeshweni kwe-hyperparimita kuya kuzixazululo zokuhambisa.
Ucwaningo lwezwe.
I-Keras isetshenziswa yi-Cern, NASA, NIH, kanye nezinhlangano eziningi zesayensi emhlabeni jikelele (futhi yebo, iKeras isetshenziswa kwa-LHC). UKeras unokuvumelana nezinga eliphansi lokuqalisa ukusebenzisa imibono yokucwaninga kokuphikisayo ngenkathi enikeza izici zokuzikhethela ezisezingeni eliphakeme zokusheshisa imijikelezo yokuhlola.
Amandla amakhulu afinyelelekayo.
Ngenxa yokusebenzisa kalula futhi kugxila ekuhlangenwe nakho komsebenzisi, iKeras yisisombululo esijulile sokufunda ekukhetheni izifundo eziningi zasenyuvesi. Kunconywa kabanzi njengenye yezindlela ezinhle kakhulu zokufunda ukujula.
Ukufunda ngokujulile kwabantu.
IKeras iyi-API eyenzelwe abantu, hhayi imishini. UKeras ulandela izindlela ezinhle kakhulu zokunciphisa umthwalo wokuqonda: unikeza ama-API angaguquki nalula, unciphisa inani lezenzo zomsebenzisi ezidingekayo kumacala wokusetshenziswa okujwayelekile, futhi linikeza nemiyalezo yephutha ecacile nenokwenzeka. Iphinde ibe nemibhalo ebanzi nemihlahlandlela yonjiniyela.
Iterate ngejubane lomcabango.
UKeras uhlaka olusetshenziswa kakhulu lokufunda phakathi kwamaqembu aphumelela i-5 kuKaggle. Ngoba iKeras yenza kube lula ukuqhuba izivivinyo ezintsha, ikunikeza amandla okuzama imibono eminingi kunomncintiswano wakho, ngokushesha. Futhi unqoba kanjani.
Ukufunda ngomshini we-Exascale.
Yakhelwe ngaphezulu kweTensorFlow 2.0, iKeras uhlaka lwamandla ezimboni olungafinyelela kumaqoqo amakhulu we-GPUs noma yonke i-TPU pod. Akunakwenzeka kuphela; kulula.
Hambisa noma kuphi.
Thatha ithuba lamakhono wokuphakwa ngokugcwele kwepulatifomu yeTensorFlow. Ungathumela amamodeli weKeras kwiJavaScript ukuze usebenze ngokuqondile kwisiphequluli, ku-TF Lite ukuze usebenze ku-iOS, i-Android namadivayisi ashumekiwe. Kulula futhi ukusebenzisa amamodeli weKeras njenge-API yewebhu.
Isistimu enkulu yemvelo.
IKeras iyingxenye ephakathi ye-TensorFlow 2.0 ecosystem exhunywe ngokusezingeni eliphezulu, emboza zonke izinyathelo zokuhamba komsebenzi wokufunda komshini, kusukela ekuphathweni kwedatha kuya ekuqeqeshweni kwe-hyperparimita kuya kuzixazululo zokuhambisa.
Ucwaningo lwezwe.
I-Keras isetshenziswa yi-Cern, NASA, NIH, kanye nezinhlangano eziningi zesayensi emhlabeni jikelele (futhi yebo, iKeras isetshenziswa kwa-LHC). UKeras unokuvumelana nezinga eliphansi lokuqalisa ukusebenzisa imibono yokucwaninga kokuphikisayo ngenkathi enikeza izici zokuzikhethela ezisezingeni eliphakeme zokusheshisa imijikelezo yokuhlola.
Amandla amakhulu afinyelelekayo.
Ngenxa yokusebenzisa kalula futhi kugxila ekuhlangenwe nakho komsebenzisi, iKeras yisisombululo esijulile sokufunda ekukhetheni izifundo eziningi zasenyuvesi. Kunconywa kabanzi njengenye yezindlela ezinhle kakhulu zokufunda ukujula.
Kubuyekezwe ngo-
Ukuphepha kuqala ngokuqonda ukuthi onjiniyela baqoqa futhi babelane kanjani ngedatha yakho. Ubumfihlo bedatha nezinqubo zokuphepha zingahluka kuye ngokusebenzisa kwakho, isifunda, nobudala. Unjiniyela unikeze lolu lwazi futhi angalubuyekeza ngokuhamba kwesikhathi.
Ayikho idatha eyabiwe nezinkampani zangaphandle
Funda kabanzi mayelana nendlela onjiniyela abaveza ngayo ukwabelana
Ayikho idatha eqoqiwe
Funda kabanzi mayelana nokuthi onjiniyela bakuveza kanjani ukuqoqwa
Yini entsha
Keras 2.3 Documentation
Ukusekelwa kwe-app
Mayelana nonjiniyela
GWEE KENG SHENG
dictson@nextlabs.cc
905, Jalan Melor 2,
Taman Tangkak Jaya,
84900 Tangkak
Johor
Malaysia
undefined
