|
|
|
|
@ -2,25 +2,20 @@ |
|
|
|
|
|
|
|
|
|
## usage |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
```commandline |
|
|
|
|
python ./run.py |
|
|
|
|
``` |
|
|
|
|
## description |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
you can train the model by `run.py` |
|
|
|
|
|
|
|
|
|
## requirement |
|
|
|
|
|
|
|
|
|
TensorFlow==1.15.4 |
|
|
|
|
dppy==0.2.0 |
|
|
|
|
munkres |
|
|
|
|
scikit-learn |
|
|
|
|
cvxopt |
|
|
|
|
|
|
|
|
|
> `cvxopt` requires `Numpy-MKL`, you can get the windows binary [here](http://www.lfd.uci.edu/~gohlke/pythonlibs/#numpy). |
|
|
|
|
> And install the `cvxopt` binary from [here](http://www.lfd.uci.edu/~gohlke/pythonlibs/#cvxopt) |
|
|
|
|
|
|
|
|
|
## acknowledge |
|
|
|
|
Thanks for the projects **GAT：** [https://github.com/PetarV-/GAT](https://github.com/PetarV-/GAT) |
