search

MTG / DeepConvSep

Deep Convolutional Neural Networks for Musical Source Separation
GNU Affero General Public License v3.0
471 stars 110 forks source link

dear MTG #21

Closed hjzSky closed 6 years ago

hjzSky commented 6 years ago

I have a problem when run your codes. My command line like "python separate_dsd.py -i /home/hjz/test/1.wav -o /home/hjz/test/ -m /home/hjz/test/model_dsd_fft_1024.pkl". The 1.wav is a music file converted from a .mp3 file. Screenshot like that: 2018-03-23 21-38-41 The version of theano installed on my computer is 0.8.2 and Lasagne is 0.1. I run this code in linux17.04 Could you tell me why?

nkundiushuti commented 6 years ago

I think we use a newer version of lasagne. I suggest you use the bleeding edge version from github as per instructions in the readme.

hjzSky commented 6 years ago

The version of lasagne I using is a latest one, just the 0.8.2. It still not work. The terminal show me there not a module named BiasLayer in lasagne.layers. But in the Official website of Lasagne i find out BiasLayer in Special-purpose layers. 2018-03-26 15 52 54

hjzSky commented 6 years ago

Thank you very much. I have solved the problem. The reason of this issue is that I installed the lasagne using 'pip install lasagne', but there is not a newest version in 'pypi'. So I view the project of lasagne on the github, I find a method of installing by this command "pip install https://github.com/Lasagne/Lasagne/archive/master.zip". So the problem was solved.