quiz bu lëkkale ak tegtal · Diggante Niveau
DenseNet ak Quiz lëkkaloo bu dëgër
Saytu màndarga-kàrtu boole, tolluwaayu màgg, jëfandikoowaat màndarga ci saasi, kompresioŋ tranzisioŋ, yooni gradient, ak lim chaine yi.
Laaj 1 de 8
Ban jëf la DenseNet di jëfandikoo ngir boole kàrtu màndarga yi bawoo ci diisaay yi njëkk?
Xoolaatal laaj bu nekk8 laaj
Sooy def quiz bi, tontu yi ak leeral yi dañuy feeñ.
- Ban jëf la DenseNet di jëfandikoo ngir boole kàrtu màndarga yi bawoo ci diisaay yi njëkk?
- lëkkaloo ci wàllu chaine
- Yokk ci wàllu elemen
- Max pooling ci diisaay yi
- Yokkateg elemen
- Ci DenseNet, lan mooy 'taux de croissance' k?
- Jadwalu tolluwaayu jàng
- Dayo palanteer pooling bi
- Limu kàrtu màndarga yu bees yi layer bu nekk di defar
- limu blok yu dëgër yi
- Ñaata lëkkaloo yu direct ñoo am ci digganté L layers ci benn block bu dëgër?
- L
- L(L+1)/2
- 2L
- L kaare
- Lan mooy jubluwaay bi gëna mag ci wàllu jàll ci DenseNet?
- Ngir yokk ay gaawaay yu des
- Yokk màgg gi
- ngir sàkku dropout kese
- Ngir wàññi misaal bi ak wàññi dayo kàrtu man-man bi ci diggante blok yu dëgër yi
- Njariñ li gëna mag ci lëkkaloo bu dëgër mooy, ban jafe-jafe tàggat yaram lay jàppale?
- Kulòor
- Dayo lote yuy yokk
- Jafe-jafe gradient buy réer
- Etiketu bruit
- Buñu ko méngale ak ResNet ci njubte bu noonu ci ImageNet, DenseNet-BC dafay faral di jëfandikoo:
- ay parametre yu néew lool
- Parametre yu bari
- Lu tollu ci benn parametre
- Amul benn convolusioŋ
- 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
- 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