uoguelph-mlrg / cutout Goto Github PK
View Code? Open in Web Editor NEW2.56%, 15.20%, 1.30% on CIFAR10, CIFAR100, and SVHN https://arxiv.org/abs/1708.04552
License: Other
2.56%, 15.20%, 1.30% on CIFAR10, CIFAR100, and SVHN https://arxiv.org/abs/1708.04552
License: Other
ValueError: Expected input batch_size (128) to match target batch_size (32), it happen all the time, the expexted size is 4 times batch size. The output of forward function return probabilities 4*batchsize x num_classes. The model used is wideresnet.
How about adding also a probability to vary when Cutout
is applied? For example something like this:
class Cutout(object):
"""Randomly mask out one or more patches from an image.
Args:
n_holes (int): Number of patches to cut out of each image.
length (int): The length (in pixels) of each square patch.
p1 (float): Probability of applying CutOut (default value 1, always applied).
"""
def __init__(self, n_holes, length, p1=1.0):
self.n_holes = n_holes
self.length = length
self.p1 = p1
def __call__(self, img):
"""
Args:
img (Tensor): Tensor image of size (C, H, W).
Returns:
Tensor: Image with n_holes of dimension length x length cut out of it.
"""
if torch.rand(1) > p1: #np.random.rand(1) > p1:
return img
# [...]
return img
So, when np.random.rand(1)
is larger than p1
the input image is returned unchanged (most of the time if p1
is small), if I got it right.
I have tried to run your code on cifar 100 with ResNet-18, but I can't get the same accuracy as you described in your paper.
My result is less than 70%, but yours is 77.54%
(with data augmentation, but without cutout. I just clone your code, and didn't do any change on your model or anything else)
So my question is: Is here any training tricks that didn't mention in your paper or code?
Thank you!
I only modify the global pooling according to STL10's image size. I also follow the implementation details you mentioned in the paper. But I cannot reproduce the result of STL10+ as 14:21 ± 0:29. My result is about 18. So I was wondering did you make some change in WRN?
My normalize parameters are mean as [0.44671097, 0.4398105 , 0.4066468 ], std as [0.2603405 , 0.25657743, 0.27126738]. Could you help me to reproduce the results?
@TDeVries This repository looks very useful thanks for putting it up!
Would you mind adding a license to this? Without a license it is impossible to legally clone or run this code. If you're not sure and would like a suggestion the Apache 2.0 license is a good option. A quick summary of apache 2.0 is available at tl;dr legal. This is the same license used by TensorFlow and it is compatible with the pytorch license. Here is the license text:
Apache License, Version 2.0 Apache License Version 2.0, January 2004 http://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction, and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all other entities that control, are controlled by, or are under common control with that entity. For the purposes of this definition, "control" means (i) the power, direct or indirect, to cause the direction or management of such entity, whether by contract or otherwise, or (ii) ownership of fifty percent (50%) or more of the outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications, including but not limited to software source code, documentation source, and configuration files.
"Object" form shall mean any form resulting from mechanical transformation or translation of a Source form, including but not limited to compiled object code, generated documentation, and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or Object form, made available under the License, as indicated by a copyright notice that is included in or attached to the work (an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object form, that is based on (or derived from) the Work and for which the editorial revisions, annotations, elaborations, or other modifications represent, as a whole, an original work of authorship. For the purposes of this License, Derivative Works shall not include works that remain separable from, or merely link (or bind by name) to the interfaces of, the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including the original version of the Work and any modifications or additions to that Work or Derivative Works thereof, that is intentionally submitted to Licensor for inclusion in the Work by the copyright owner or by an individual or Legal Entity authorized to submit on behalf of the copyright owner. For the purposes of this definition, "submitted" means any form of electronic, verbal, or written communication sent to the Licensor or its representatives, including but not limited to communication on electronic mailing lists, source code control systems, and issue tracking systems that are managed by, or on behalf of, the Licensor for the purpose of discussing and improving the Work, but excluding communication that is conspicuously marked or otherwise designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity on behalf of whom a Contribution has been received by Licensor and subsequently incorporated within the Work.
2. Grant of Copyright License.
Subject to the terms and conditions of this License, each Contributor hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable copyright license to reproduce, prepare Derivative Works of, publicly display, publicly perform, sublicense, and distribute the Work and such Derivative Works in Source or Object form.
3. Grant of Patent License.
Subject to the terms and conditions of this License, each Contributor hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable (except as stated in this section) patent license to make, have made, use, offer to sell, sell, import, and otherwise transfer the Work, where such license applies only to those patent claims licensable by such Contributor that are necessarily infringed by their Contribution(s) alone or by combination of their Contribution(s) with the Work to which such Contribution(s) was submitted. If You institute patent litigation against any entity (including a cross-claim or counterclaim in a lawsuit) alleging that the Work or a Contribution incorporated within the Work constitutes direct or contributory patent infringement, then any patent licenses granted to You under this License for that Work shall terminate as of the date such litigation is filed.
4. Redistribution.
You may reproduce and distribute copies of the Work or Derivative Works thereof in any medium, with or without modifications, and in Source or Object form, provided that You meet the following conditions:
You must give any other recipients of the Work or Derivative Works a copy of this License; and You must cause any modified files to carry prominent notices stating that You changed the files; and You must retain, in the Source form of any Derivative Works that You distribute, all copyright, patent, trademark, and attribution notices from the Source form of the Work, excluding those notices that do not pertain to any part of the Derivative Works; and If the Work includes a "NOTICE" text file as part of its distribution, then any Derivative Works that You distribute must include a readable copy of the attribution notices contained within such NOTICE file, excluding those notices that do not pertain to any part of the Derivative Works, in at least one of the following places: within a NOTICE text file distributed as part of the Derivative Works; within the Source form or documentation, if provided along with the Derivative Works; or, within a display generated by the Derivative Works, if and wherever such third-party notices normally appear. The contents of the NOTICE file are for informational purposes only and do not modify the License. You may add Your own attribution notices within Derivative Works that You distribute, alongside or as an addendum to the NOTICE text from the Work, provided that such additional attribution notices cannot be construed as modifying the License. You may add Your own copyright statement to Your modifications and may provide additional or different license terms and conditions for use, reproduction, or distribution of Your modifications, or for any such Derivative Works as a whole, provided Your use, reproduction, and distribution of the Work otherwise complies with the conditions stated in this License.
5. Submission of Contributions.
Unless You explicitly state otherwise, any Contribution intentionally submitted for inclusion in the Work by You to the Licensor shall be under the terms and conditions of this License, without any additional terms or conditions. Notwithstanding the above, nothing herein shall supersede or modify the terms of any separate license agreement you may have executed with Licensor regarding such Contributions.
6. Trademarks.
This License does not grant permission to use the trade names, trademarks, service marks, or product names of the Licensor, except as required for reasonable and customary use in describing the origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty.
Unless required by applicable law or agreed to in writing, Licensor provides the Work (and each Contributor provides its Contributions) on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied, including, without limitation, any warranties or conditions of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A PARTICULAR PURPOSE. You are solely responsible for determining the appropriateness of using or redistributing the Work and assume any risks associated with Your exercise of permissions under this License.
8. Limitation of Liability.
In no event and under no legal theory, whether in tort (including negligence), contract, or otherwise, unless required by applicable law (such as deliberate and grossly negligent acts) or agreed to in writing, shall any Contributor be liable to You for damages, including any direct, indirect, special, incidental, or consequential damages of any character arising as a result of this License or out of the use or inability to use the Work (including but not limited to damages for loss of goodwill, work stoppage, computer failure or malfunction, or any and all other commercial damages or losses), even if such Contributor has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability.
While redistributing the Work or Derivative Works thereof, You may choose to offer, and charge a fee for, acceptance of support, warranty, indemnity, or other liability obligations and/or rights consistent with this License. However, in accepting such obligations, You may act only on Your own behalf and on Your sole responsibility, not on behalf of any other Contributor, and only if You agree to indemnify, defend, and hold each Contributor harmless for any liability incurred by, or claims asserted against, such Contributor by reason of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
APPENDIX: How to apply the Apache License to your work
To apply the Apache License to your work, attach the following boilerplate notice, with the fields enclosed by brackets "[]" replaced with your own identifying information. (Don't include the brackets!) The text should be enclosed in the appropriate comment syntax for the file format. We also recommend that a file or class name and description of purpose be included on the same "printed page" as the copyright notice for easier identification within third-party archives.
Copyright [yyyy] [name of copyright owner]
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.
When executing python train.py --dataset svhn --model wideresnet --learning_rate 0.01 --epochs 160 --cutout --length 20
I get the following error:
Traceback (most recent call last):
File "train.py", line 149, in <module>
dropRate=0.4)
File "Cutout/model/wide_resnet.py", line 58, in __init__
self.block1 = NetworkBlock(n, nChannels[0], nChannels[1], block, 1, dropRate)
File "Cutout/model/wide_resnet.py", line 38, in __init__
self.layer = self._make_layer(block, in_planes, out_planes, nb_layers, stride, dropRate)
File "Cutout/model/wide_resnet.py", line 41, in _make_layer
for i in range(nb_layers):
TypeError: 'float' object cannot be interpreted as an integer
This error stems from the division here. For Python3.x, integer division must be forced like this: n = (depth - 4) // 6
Hi @TDeVries and @gwtaylor ! This paper is amazing(!!!!) and I've been using it's insights ever since it came out. Although I feel I am not getting the same benefits everyone else is getting when applying this regularization technique to my training process. I have a few questions to clear some cobwebs in my head regarding the technique:
How important is per channel normalization to getting cutout to work properly? Is dividing by 255 a bad way to normalize images? What is a go-to image normalization approach?
When should cutout be applied? Should it be applied before OR after image normalization?
If I have other augmentations happening to my images, should cutout be applied before OR after my other augmentations?
Thank you!!
In object detection, I would like to ask how to set the default parameter of cut out better? For example, num_holes and length, I want to know whether you has done any experiments when adding these parameters, or has anyone else set them? Thank you
Many thanks for your clean codes.
I wonder which version of SVHN dataset (it consists of two sub-versions, one including 73257 training images and 26032 test images, the other one named extra version containing 631131 images) is used to train?
I am trying to use cutout in my implementation, I keep getting the error below when I run the train code. Please, am I doing something wrong?
from cutout import Cutout
train_transform = transforms.Compose([
transforms.RandomCrop(32, padding=4, fill=128),
transforms.RandomHorizontalFlip(),
CIFAR10Policy(),
Cutout(n_holes=1, length=16),
transforms.ToTensor(),
transforms.Normalize((0.4914, 0.4822, 0.4465), (0.2023, 0.1994, 0.2010))
])
TypeError Traceback (most recent call last)
in
4 correct = 0.0
5 total = 0.0
----> 6 for i, data in enumerate(train_loader):
7 # get the inputs; data is a list of [inputs, labels]
8 inputs, labels = data
~/anaconda3/lib/python3.7/site-packages/torch/utils/data/dataloader.py in next(self)
817 else:
818 del self._task_info[idx]
--> 819 return self._process_data(data)
820
821 next = next # Python 2 compatibility
~/anaconda3/lib/python3.7/site-packages/torch/utils/data/dataloader.py in _process_data(self, data)
844 self._try_put_index()
845 if isinstance(data, ExceptionWrapper):
--> 846 data.reraise()
847 return data
848
~/anaconda3/lib/python3.7/site-packages/torch/_utils.py in reraise(self)
383 # (https://bugs.python.org/issue2651), so we work around it.
384 msg = KeyErrorMessage(msg)
--> 385 raise self.exc_type(msg)
TypeError: Caught TypeError in DataLoader worker process 0.
Original Traceback (most recent call last):
File "/home/enoch/anaconda3/lib/python3.7/site-packages/torch/utils/data/_utils/worker.py", line 178, in _worker_loop
data = fetcher.fetch(index)
File "/home/enoch/anaconda3/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 44, in fetch
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/home/enoch/anaconda3/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 44, in
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/home/enoch/anaconda3/lib/python3.7/site-packages/torchvision/datasets/cifar.py", line 125, in getitem
img = self.transform(img)
File "/home/enoch/anaconda3/lib/python3.7/site-packages/torchvision/transforms/transforms.py", line 61, in call
img = t(img)
File "/home/enoch/medicals/cutout.py", line 23, in call
h = img.size(1)
TypeError: 'tuple' object is not callable
I really like your idea, I think this is a very useful idea.
thank you!
Thanks for your work, really great. But I didn't see the experiment on ImageNet, what is the cutout region for ImageNet? Thanks.
First,thank you very much,You have done a very good job.
I don't unstand the following content in your paper
Figure 4: Magnitude of feature activations, sorted by descending value, and averaged over all test samples. A standard ResNet18 is compared with a ResNet18 trained with cutout at three different depths.
could your give me some tips how do you generate these figures?
and I don't know the means of the abscissa and ordinate values.
I need you help.thank you.
We added a generalized version of Cutout in Albumentains called CourseDropout
Generalized as it allows:
images
, segmentation masks
and key points
Feel free to check documentation
Could I ask if you have ever tried using Adam and AdaBelief? Have you had the opportunity to compare the performance of these two optimizers?
When I use torch.utils.data.DataLoader init with multi worker and also using Cutout for transform, I observed same transformed images were generated. This number of images are same as number of workers. I guess this is because use of numpy.random in Cutout source code. In torchvision.transforms, numpy seems to be initialized by the same seed in each workers.
This issue can be fixed by replacing numpy.random with torch.randint in Cutout source code.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.