-
Notifications
You must be signed in to change notification settings - Fork 88
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add BN-Inception #2
base: master
Are you sure you want to change the base?
Conversation
Thank you for your PR! In a near futur, may I add it to https://github.com/Cadene/pretrained-models.pytorch ? |
Sure, with pleasure. |
As a side note, this BN-Inception model takes input in the range of (0~255) . So only mean subtraction should be used in preprocessing. This is different from ResNet models provided by Facebook guys, which take input in the range of (0~1). I think it is worth to note this when listing it. |
Yes sure. What is the exact transforms you used to evaluate your model ? Something like this? normalize = transforms.Normalize(mean=[127.5, 127.5, 127.5],
std=[1, 1, 1])
transforms.Compose([
transforms.RandomSizedCrop(224),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
normalize
]) And do you know the accuracy top1/top5 on imagenet valset ? Thanks |
Good point. Since it's from Caffe, it uses mean of |
I will give a try asap. Thanks ;) |
@yjxiong I encounter the following error when loading the pretrained parameters: Is it still working for you?
|
Hey there! Sorry to reply so late. I have identified the issue. It is related to the file extension limit of Azure web services. I have reconfigured the server. Now it should works as expected. Please try again installing the model. Thanks! |
Cool, I will give it a try tomorrow and will let you know ;) |
I just evaluated it on imagenet validation set, it works! What is the version of this implementation of BN Inception ? |
Hi @Cadene , Thanks for the testing. The model is implementing the one in Figure 5 of the paper. In Caffe we got 92.14% accuracy with it on the Val set. There might be some slight differences between the two frameworks. I guess the performance you got is quite reasonable. So please go ahead to release it to the community. |
Thanks very much for your great work. I have successfully used the pretrained model in my research. But I have found a problem, though this model can be successfully loaded by pytorch 0.2.0 and 0.3.0. But when we want to finetune the model, results under different version of Pytorch is very different. |
@yjxiong So, I want to ask you the version of python, torchvision and pytorch you used in tranferring the model from caffe. |
PS: Both version of pytorch can successfully load your model, I have see into the value fo the weight. Both are same, and the forward output is the same. |
do you have the architecture of BNInception? |
@poppingcode The architecture is the one described in the original batch normalization paper. |
@yjxiong when I test model with RGB, it appears error as below: Please give your pytorch version that you used to train |
Hi, |
This PR adds support for BN-Inception. The model architecture is from the paper
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift, Ioeff et. al.
Since Google didn't release trained weights of this architecture, we trained it by ourselves on ImageNet ILSVRC12 train data. It produces 92.1% top-5 accuracy on the single center crops, almost the save as reported in the original paper.
The model is not based on Tensorflow but translated from Caffe. But I feel it is still worth to provide it to other users.