Pocket AutoML
1K+
Okudawunilodiwe
Wonke umuntu
info
Mayelana nalolu hlelo lokusebenza
• IPocket AutoML ivumela abathandi bobuhlakani bokufakelwa noma bengenaso isipiliyoni sokufunda semishini ngaphambi kokuqeqesha imodeli yokuhlukanisa isithombe ejulile (inethiwekhi ye-neural eguqukayo) khona kanye kumadivayisi abo futhi bayizame.
• Ingathumela amamodeli ngefomethi yeTensorFlow Lite ukuze umsebenzisi akwazi ukwenza uhlelo lokusebenza lwangokwezifiso lwe-Android ngokulandela okuhlinzekiwe okokufundisa .
• Umbono wekhompyutha noma ochwepheshe bokufunda abajulile bangayisebenzisa njengethuluzi lokwenza ubufakazi obusheshayo bomqondo wokudlulisa ukufunda emisebenzini yabo ngaphandle kolayini owodwa wekhodi.
• Iqeqesha imodeli ngqo kudivayisi yakho ngemizuzwana (yedathasethi enenqwaba yezithombe).
• Ihlonipha ubumfihlo bakho: izithombe zakho azikalayishwa noma yikuphi njengoba ukuqeqeshwa nokubikezela kwenzeka kudivayisi yakho.
• Akudingi ukuxhumana kwe-inthanethi ukuqeqeshwa nokubikezela.
• Zimbalwa nje izithombe ekilasini ngalinye ezenele ukuqeqesha imodeli ehlukanisa izinto ngokunembile (lokho kwaziwa njengokufunda okumbalwa).
• Ingathumela amamodeli ngefomethi yeTensorFlow Lite ukuze umsebenzisi akwazi ukwenza uhlelo lokusebenza lwangokwezifiso lwe-Android ngokulandela okuhlinzekiwe okokufundisa .
• Umbono wekhompyutha noma ochwepheshe bokufunda abajulile bangayisebenzisa njengethuluzi lokwenza ubufakazi obusheshayo bomqondo wokudlulisa ukufunda emisebenzini yabo ngaphandle kolayini owodwa wekhodi.
• Iqeqesha imodeli ngqo kudivayisi yakho ngemizuzwana (yedathasethi enenqwaba yezithombe).
• Ihlonipha ubumfihlo bakho: izithombe zakho azikalayishwa noma yikuphi njengoba ukuqeqeshwa nokubikezela kwenzeka kudivayisi yakho.
• Akudingi ukuxhumana kwe-inthanethi ukuqeqeshwa nokubikezela.
• Zimbalwa nje izithombe ekilasini ngalinye ezenele ukuqeqesha imodeli ehlukanisa izinto ngokunembile (lokho kwaziwa njengokufunda okumbalwa).
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
Le app ingaqoqa lezi zinhlobo zedatha
Umsebenzi wohlelo lokusebenza, Ulwazi lwe-app nokusebenza, ne-Idivayisi noma amanye ama-ID
Idatha ibetheliwe lapho ithunyelwa
Idatha ayikwazi ukusulwa
Yini entsha
• Improved model training speed.
• Added the Android app creation tutorial.
• Added model export in TensorFlow Lite format.
• Added model statistics.
• Fixed an issue with the app returning to the main screen, e.g. when taking or picking photos.
• Better photo quality.
• More hints for new users.
• Faster image import from device storage.
• Other bugfixes and improvements.
• Added the Android app creation tutorial.
• Added model export in TensorFlow Lite format.
• Added model statistics.
• Fixed an issue with the app returning to the main screen, e.g. when taking or picking photos.
• Better photo quality.
• More hints for new users.
• Faster image import from device storage.
• Other bugfixes and improvements.
Ukusekelwa kwe-app
Mayelana nonjiniyela
Evgeniy Mamchenko
evgeniy.mamchenko@gmail.com
Friedrichstraße 114A
10117 Berlin
Germany
undefined
