sg-nm/cgp-cnn-PyTorch
A Genetic Programming Approach to Designing CNN Architectures, In GECCO 2017 (oral presentation, Best Paper Award)
PythonMIT
Issues
- 0
Looking for help: The CGP-CNN (ConvSet) and CGP-CNN (ResSet) consist of two basic constituent unit modules in the text. However, when I ran the code I found that the two basic unit modules were mixed in one network at the same time. Is there a parameter variable in the code that can be controlled to get the results in the paper.
#13 opened by Xiaojiuwo168 - 0
How to find the generated architecture
#12 opened by GBitocchi - 0
- 0
the acc of retrain
#10 opened by Sunrisulfr - 0
- 5
RuntimeError: Expected a_in <= b_in to be true, but got false. (Could this error message be improved? If so, please report an enhancement request to PyTorch.)
#4 opened by yuxiaomu - 0
the acc of retrain
#7 opened by yuxiaomu - 0
- 5
- 0
ask for help about the result
#6 opened by yuxiaomu - 0
exp_main.py
#3 opened by yuxiaomu - 0
Hello, author, thank you very much for your work, but can you give us a more detailed explanation document on how to use the code?
#2 opened by yuxiaomu