from fastai.gen_doc.nbdoc import *
from fastai.vision.models.darknet import Darknet
from fastai.vision.models.wrn import wrn_22, WideResNet
On top of the models offered by torchivision, the fastai library has implementations for the following models:
models.unet
show_doc(Darknet, doc_string=False)
Create a Darknet with blocks of sizes given in num_blocks
, ending with num_classes
and using nf
initial features. Darknet53 uses num_blocks = [1,2,8,8,4]
.
show_doc(WideResNet, doc_string=False)
Create a wide resnet with blocks num_groups
groups, each containing blocks of size N
. k
is the width of the resnet, start_nf
the initial number of features. Dropout of drop_p
is applied at the end of each block.
show_doc(wrn_22)
wrn_22
[source]
wrn_22
()
Creates a wide resnet for CIFAR-10 with num_groups=3
, N=3
, k=6
and drop_p=0.
.