pretrained-models.pytorch
pretrained-models.pytorch copied to clipboard
Cadene
Pretrained ConvNets for pytorch: NASNet, ResNeXt, ResNet, InceptionV4, InceptionResnetV2, Xception, DPN, etc.
Is there a way to use bn-inception pretrained weights of places365 dataset from caffemodel trained using https://github.com/yjxiong/caffe . Thanks.
## Problems / Bugs 1. Data Parallel issue with types.MethodType #112 2. The 'num_classes' is not passed to torchvision models. All models are initiated for 1000 classes even if num_classes...
I found that when using nn.data_parallel along with this library, there will be issues during model forward on multi-GPU as when modifying torch vision network (such as in modify_resnets function)...
It's rasing issue in the **from . import utils** File "voc.py", line 13, in from . import utils ValueError: Attempted relative import in non-package How to solve this error?
Hello there, I'm trying to evaluate some models but it is giving 0.0 in acc. I'm using Python 3.6.5 and pytorch 0.4.0  I got...
This net is the current state-of-the-art on ImageNet and CIFAR-10: _GPipe: Efficient Training of Giant Neural Networks using Pipeline Parallelism_ https://arxiv.org/abs/1811.06965 Would be great to have a pretrained model. So...
Dear folks, I know this is not likely your bug, but I think it is good to at least share the result here so other people could avoid wasting time...
Traceback (most recent call last): File "clas_inc_res.py", line 299, in model_conv.classifier._modules["1"] = nn.Conv2d(512, num_of_output_classes, kernel_size=(1, 1)) File "/home/raman/anaconda2/envs/pyto3/lib/python3.6/site-packages/torch/nn/modules/module.py", line 535, in __getattr__ type(self).__name__, name)) AttributeError: 'InceptionResNetV2' object has no attribute...
this repo is a very good accuracy rate benchmark reference for all model on imagenet classification task, but when do real project deployment the model's inference time cost is also...
I am trying to pip install pretrainedmodels. I am getting an error as ImportError: The 'packaging.version' package is required; normally this is bundled with this package so if you get...
