pretrained-models.pytorch
pretrained-models.pytorch copied to clipboard
Pretrained ConvNets for pytorch: NASNet, ResNeXt, ResNet, InceptionV4, InceptionResnetV2, Xception, DPN, etc.
The location of the data changed, however the URL haven't been updated in the repository. E.g.: > /pretrainedmodels/models/resnext.py ``` pretrained_settings = { 'resnext101_32x4d': { 'imagenet': { 'url': 'http://data.lip6.fr/cadene/pretrainedmodels/resnext101_32x4d-29e315fa.pth', 'input_space': 'RGB',...
Removed a fully connected layer in the resnet models instead of installing None. Same in VGG.
I really don't understand such complex regular expressions. Can you explain it to me? re.compile( r'^(.*denselayer\d+\.(?:norm|relu|conv))\.((?:[12])\.(?:weight|bias|running_mean|running_var))$') Thanks!
RuntimeError: invalid hash value (expected "ce0d4300", got "a0a8c5ad919066a582f73746eb7f2ee742e282baab9130860ca03fe1f17ead75") the code I used was model = pretrainedmodels.__dict__['se_resnet50'](num_classes=1000, pretrained='imagenet')
Hi, Currently, most of the pre-trained models are still on Tensorflow. For me, I prefer pytorch, but I have to use Tensorflow. Could you add some common defense models in...
Use tensor.item() to convert a 0-dim tensor to a Python number
dont know why but i cant find attribut anywhere
Hi, I have tried to reproduce the result of ResNet-18 in your table which is: Model | Version | Acc@1 | Acc@5 -- | -- | -- | -- ResNet18...
My editor say Undefined variable 'normalize', so i want to know where normalize is defined ?
when i python setup.py install Processing dependencies for pretrainedmodels==0.7.4 error: torch 0.1.12.post2 is installed but torch==1.2.0 is required by {'torchvision'}