DenseNet iyo Kediska Isku Xidhka Cufan

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

1. Hawlgalkee ayaa DenseNet u isticmaashaa isku-darka khariidadaha sifada ee lakabyada hore?
   • Isku xidhka kanaalka-caqliga leh
   • Cunsurka-xikmad ku darid
   • Isku dhafka ugu badan ee lakabyada
   • Isku dhufashada curiyaha-xikmad leh
2. DeenseNet gudaheeda, waa maxay 'heerka kobaca' k kantaroolo?
   • Jadwalka heerka waxbarashada
   • Cabbirka daaqada is biirsiga
   • Tirada khariidadaha sifada cusub ee lakab kastaa soo saaro
   • Tirada blocks cufan
3. Immisa xiriir oo toos ah ayaa ka dhex jira lakabyada L ee ku yaal block cufan?
   • L
   • L(L+1)/2
   • 2L
   • L labajibbaaran
4. Waa maxay ujeedada ugu weyn ee lakabyada kala-guurka ee DenseNet?
   • Si loogu daro hadhaaga hadhaaga gaaban
   • Si loo kordhiyo heerka koritaanka
   • Si aad u codsato ka-tegidda kaliya
   • Si loo dhimo oo loo yareeyo cabbirka-map-ka ee u dhexeeya blocks cufan
5. Faa'iidada muhiimka ah ee isku xirnaanta cufan waa inay ka caawiso yaraynta dhibaatada tababarka?
   • Ku habboon midabka
   • Cabbirrada dufcada qarxinta
   • Dhibaatada sii liidata
   • Ku calaamadee buuqa
6. Marka la barbar dhigo ResNet isla saxnaanta ImageNet, DenseNet-BC waxay caadi ahaan isticmaashaa:
   • Qiyaaso aad u yar
   • Xuduudo aad u fog
   • Ku saabsan cabbirro isku mid ah
   • Wax qalalaase ah haba yaraatee
7. A late DenseNet layer needs an edge feature learned near the input. Why can it use that map directly?
   • Only the final layer can see edge features
   • DenseNet resets all maps after each layer
   • Transition layers reconstruct every earlier map from scratch
   • Earlier feature maps remain among its concatenated inputs
8. A dense block begins with m channels and three layers each add k new maps. How many channels emerge before a transition?
   • 3m + k
   • m + 3k
   • m times k cubed
   • Only k, because old maps are discarded