Comments (13)
@chenbolinstudent Hi!
There is an old issue talked about this problem: link
Following this modification, you can avoid this problem.
While for me, my loss started from about 180 and suddenly jumped to nan (seems it will never back).
@marcellacornia Could you give us an intro about how to train correctly?
Many thanks!
from sam.
Hi @chenbolinstudent,
thanks for downloading our code.
The data loading functions are compatible with the original release of the SALICON dataset. Which dataset are you using?
from sam.
I use the SALICON
from sam.
my configure is cuda9,python2.7,keras=1.1.0,theano=1.0.3,it that any configure require?thanks
from sam.
I do not know the function of ["I"] in the scipy
from sam.
The ["I"] is not a function of SciPy. For the original release of the SALICON dataset, authors provided matlab files with the fixation map for each image and the ["I"] is an attribute of those matlab files.
from sam.
can you give me the source of the original release of the SALICON dataset?thanks
from sam.
At the end of this page, you can find matlab files and saliency maps of the original release (i.e. the previous one) of the SALICON.
from sam.
thank you very much
from sam.
however,when I use the release salicon, i meet the same problem:
fix_map = scipy.io.loadmat(path)["I"]
KeyError: 'I'
from sam.
@eleboss Hi! I'm also experiencing a sudden drop to nan for loss. Did you figure out anything?
from sam.
@plin24
Sorry, seems everyone experienced the same issue.
I am waiting for the answer from author.
from sam.
@eleboss @plin24 Hi! Thanks for downloading our code and sorry for the delay.
In our experiments, we did not notice the same behavior during training and it's difficult to understand what caused the problem.
It could be helpful to understand if a specific metric drops to nan. Note that our model is trained by using a combination of three different saliency evaluation metrics (i.e. KL-Divergence, Normalized Scanpath Saliency, and Linear Correlation Coefficient). Could you please try to train the network by using a single metric at a time as loss function?
However, if you are interested, I have released the weights of our model (the ResNet version) also on the new release of the SALICON dataset. You can find them in the Pretrained Models Section of the README.
from sam.
Related Issues (20)
- raise NotImplementedError HOT 2
- 'NoneType' object is not subscriptable HOT 2
- Why dose the "loss" reduce first and then increase during the training? HOT 1
- Support in TX2 HOT 2
- where is the prediction HOT 1
- train model HOT 3
- On the CPU HOT 1
- ImportError: cannot import name get_from_module HOT 1
- Error in virtual env HOT 1
- I met this error about CorrMM images and kernel must have the same stack size HOT 1
- Retraining the models
- how to calculate this fix and map
- Getting this IOError HOT 1
- Adapting different testing image shapes? HOT 1
- fix_map = scipy.io.loadmat(path)["I"] HOT 1
- How to learn the prior maps in your method?
- Is there a Pytorch version
- Outputs must be theano variables or Out instances
- cannot import name 'RMSprop' from 'keras.optimizers' HOT 2
- how to visualize the Saliency Maps?
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 sam.