Comments (2)
Thank you for your interest in our work and for these thoughtful questions.
Firstly, regarding the eps parameter setting in Batch Normalization, we indeed set it to 0.8 which is notably larger than the default value of 1e-5 in PyTorch. This setting is not arbitrary but directly borrowed from traditional GANs. You can find related code in this link: https://github.com/eriklindernoren/PyTorch-GAN/blob/master/implementations/dcgan/dcgan.py. This setting has been empirically found to be beneficial for the training of the generator.
As for your question about the correlation between the L1 norm and the authenticity of input images, our idea is that filters in the DNNs have been trained to extract intrinsic patterns in training data, and therefore feature maps tend to receive higher activation values if the input images are real rather than some random vectors. This notion is based on the paper "Interpretable convolutional neural networks" which suggests that activations in the neural network represent useful information in the input images. Therefore, we use the L1 norm as a way to measure the amount of useful information in the input images, and indirectly, their authenticity.
I hope these answers help. If you have any further questions, please don't hesitate to ask. Thank you again for your interest and support in our work!
from efficient-computing.
Wow! Thanks for your rapid answer. These info definitely solve my questions. Thanks a lot!
from efficient-computing.
Related Issues (20)
- missing labels and low evaluation HOT 3
- "_call_impl(self, *input, **kwargs)" forward abnormally。
- Assertion `index >= -sizes[i] && index < sizes[i] && "index out of bounds"` failed. For debugging consider passing CUDA_LAUNCH_BLOCKING=1. HOT 1
- 【Function get_box_metrics(self) 】Unabel to get repr for <class 'torch.Tensor'>
- question about number of layers HOT 1
- Feature map size Question? HOT 2
- The script of generating text for GPT4Image HOT 1
- hyper-parameter "--fuse_ab" in gold-yolo HOT 1
- Creating textual descriptions using BLIP-2 on Cifar-10/100 HOT 1
- Code for DAFL HOT 1
- when I train myself datasets ,a bad question was happened that my iou_loss dfl_loss and ap is zero HOT 2
- 蒸馏的时候问题:AttributeError: 'Trainer' object has no attribute 'loss_items' HOT 1
- 训练报错 HOT 1
- About parameter use_syncbn and fuse_ab in train.py HOT 1
- Gold-YOLO的运行 HOT 1
- gold_yolo怎么训练自己的数据集 HOT 1
- gold_yolo训练自己的数据集失败 HOT 1
- class Inferer 参数webcam, webcam_addr,调用时没赋值? HOT 1
- Gold-yolo推理时的精度非常差
- Does the model support OBB detection? HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from efficient-computing.