adepierre / caffe_cyclegan Goto Github PK
View Code? Open in Web Editor NEWA Caffe/C++ implementation of Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
License: MIT License
A Caffe/C++ implementation of Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
License: MIT License
Hi, Caffe_CycleGAN.exe seems to crash (windows) on batch_size > 2.
Is there anything in the code that needs changing to increase batchsize?
Thanks
Hi, when you say code relies on caffe with modifications to solver.hpp, do you mean I have to make the changes to solver.hpp then re-Build/re-compile my whole installation of caffe?
thanks
Hi, @adepierre
I really appreciate your work~
But, i want to know how can i use it in Linux.
Could you please give me more details for running in Linux?
Many thanks!
Valentina
Hi, @adepierre ~
I found a weird problem when training.
It is weird that the program is killed unexpectedly when iteration is 62250.
But nobody kills the program.
There are 753 image pairs for training.
Loss on real A data: 0.0568343 on generated A data: 0.0570172
Loss on real B data: 0.0519266 on generated B data: 0.0524207
Cyclic loss on A data: 0.229246
Cyclic loss on B data: 0.164716
I0712 23:00:19.389642 45804 sgd_solver.cpp:112] Iteration 62250, lr = 0.0002
I0712 23:00:19.398865 45804 sgd_solver.cpp:112] Iteration 62250, lr = 0.0002
I0712 23:00:19.401080 45804 sgd_solver.cpp:112] Iteration 62250, lr = 0.0002
I0712 23:00:19.401469 45804 sgd_solver.cpp:112] Iteration 62250, lr = 0.0002
Killed
Could you please answer this question?
Many thanks!!
As far as I understand, the output of the discriminator outputs a scalar which represents whether the input comes from natural image set or generated image set. But I find that your discriminator outputs a 32x32 tensor. Could you explain the reason of that? Thx
Hi, Can you share your trained model? Thank you!
No bugs found, and the solver prototxt
is OK, any ideas?
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.