Motivation
This is one step forward from Data transport between python and matlab, since sometimes you not only want to transport data, but want to share some code in python or matlab. How can we do so?
Using Python in Matlab
Matlab has official support for python and you can set your python environment within and then code python in matlab.
Caveat: python env is set in a matlab session, you cannot reset it!
But it’s handy if you would like to have some io function or some programming tricks in python but don’t want to open a new python for it.
Handling Python modules
import in matlab means aliasing, not the same as import in python. Here is a matlab example.
import matlab.io.hdf4.sd.readChunk
data = readChunk(ID,[n m]);
For real importing you can do py.importlib.import_module('lucent'); or just type py.lucent and the package will load for you.
To import arbitrary package or module you need to modify path in the matlab python environment like this.
syspath = py.sys.path(); % add the official stylegan2 repo.
syspath.append("E:\DL_Projects\Vision\stylegan2-pytorch");
py.importlib.import_module('model');
Direct append like py.sys.path.append does not work! py.sys.path = syspath; will create something worse!
Note: Currently, matlab can import pytorch 1.3.1 but not 1.5.0. It will throw an error when importing latter. But it seems everything we used are compatible with 1.3.1 .
conda install pytorch=1.3.1 torchvision cudatoolkit=10.1 -c pytorch
Using Matlab in Python
I feel there is less case for this but it’s doable. If you really need it, check this out!
import matlab.engine
import time
# 第一步，初始化Matlab的Runtime
t1 = time.time()
eng = matlab.engine.start_matlab()
t2 = time.time()
# 第二步，调用Matlab函数
res1 = eng.sqrt(16.0)
t3 = time.time()
res2 = eng.abs(-8.6)
t4 = time.time()
# 第三步，退出Runtime
eng.quit()
t5 = time.time()
import matlab.engine
A = matlab.int8([1, 2, 3, 4, 5])
print(type(A), A.size, A)
Plotting is supported as well!