I-NVIDIA inobukhona obubonakalayo emkhakheni wamanethiwekhi we-neural nobuhlakani bokufakelwa ngehardware yayo. Ubufakazi balokhu yibhodi layo elihle le-NVIDIA UJetson Nano ukuthuthukisa amaphrojekthi asuselwa kumanethiwekhi we-neural artificial ne-AI, noma ngabe ungumfundi, noma uma ufuna ukuthuthukisa okuthile okubucayi.
NgoJetson Nano ngeke kudingeke usebenzise imali eningi ukuthola i-Hardware yalolu hlobo lwephrojekthi, kepha une ibhodi lokuthuthukisa okungaphezudlwana kwe- € 100. Intengo ephansi uma ubheka ukuthi amanye amasistimu afanayo abiza ini nayo yonke into enikezwa yileli bhodi ...
Uyini uJetson Nano?
NVIDIA Jetson Nano iphrojekthi evumela ukuthuthukiswa kwezinhlelo ze-AI ngenani elingabizi nosayizi owehlisiwe. Ngamanye amagama, yi "Arduino" yomhlaba wezinethiwekhi zokufakelwa ze-neural. Yingakho ithola ukuthandwa futhi sekuvele kunamaphrojekthi amaningi asuselwa kuwo.
Ngaleli bhodi lokuthuthuka ungakha eziningana amaphrojekthi, zombili izinhlelo zokusebenza ze-IOT, njengokunye okususelwa kumarobhothi, amasistimu ahlakaniphile enza ngokufanelekile ukuhlola uchungechunge lwezimo, ukufunda okujulile, ukuqashelwa kwento nombono wokufakelwa, njll. Futhi konke kune-PCB encane yamasentimitha ambalwa kuphela ...
Impela yakho ikhithi yokuthuthukisa Iyahambisana neLinux, ngakho-ke ungathuthuka nge-NVIDIA Jetson Nano kusuka ku-distro yakho oyithandayo. Ngaphezu kwalokho, izosebenzisa amaphrojekthi adumile njengomthombo ovulekile wePyTorch neTensorFlow.
Izici zobuchwepheshe
I-NVIDIA Jetson Nano inezinye izici zobuchwepheshe Kuyamangaza, ngamandla angama-472 GFLOPs, ukusebenza okungacabangi ngosayizi wayo omncane nentengo futhi enamandla ngokwanele ukusebenzisa inani elikhulu lama-algorithms we-AI ngokushelelayo. Ungaqalisa amanethiwekhi amaningi we-neural ngasikhathi sinye ngezinhloso ezahlukahlukene.
Okunye kwe- izici Okufanele ukwazi nge-NVIDIA Jetson Nano yile:
- I-NVIDIA Maxwell GPU enezinhlamvu ze-128 CUDA.
- I-ARM Cortex-A57 QuadCore CPU
- I-4GB LPDDR4 RAM
- Isitoreji sangaphakathi esingu-16GB se-eMMC 5.1
- Izimbobo ze-MIPI CSI-2 zekhamera, iGigabit Ethrnet (RJ-45), i-HDMI 2.0 noma i-DP 1.2, i-DSI, i-PCIe, i-USB 3.0 ne-2.0, i-SDIO, i-SPI, i-I2C, i-I2S, ne-GPIO.
- Ukusetshenziswa 5-10w, amandla aphansi ngokusebenza kwawo.
- Usayizi 69.6x45mm
Imininingwane engaphezulu - NVIDIA Jetson Nano