Added GUI by Bas Curtiz

README.md

@@ -79,6 +79,7 @@ All inference parameters are [here](https://github.com/ZFTurbo/Music-Source-Sepa
 * `train.py` - main training code
 * `utils.py` - common functions used by train/valid
 * `valid.py` - validation of model with metrics
+* `ensemble.py` - useful script to ensemble results of different models to make results better (see [docs](docs/ensemble.md)).   
 
 ## Pre-trained models
 
@@ -142,6 +143,10 @@ Look here: [Dataset types](docs/dataset_types.md)
 
 Look here: [Augmentations](docs/augmentations.md)
 
+## Graphical user interface
+
+Look here: [GUI](docs/gui.md)
+
 ## Citation
 
 * [arxiv paper](https://arxiv.org/abs/2305.07489)

docs/gui.md

@@ -0,0 +1,24 @@
+## GUI for MSST code 
+
+GUI was prepared by **Bas Curtiz** and is based on [wxpython](https://en.wikipedia.org/wiki/WxPython) module.
+
+![Window example](gui/wx_msst_screen.png)
+
+### How to
+
+How to use GUI with ZFTurbo's Music Source Separation Universal Training Code:
+
+1. Install Python: https://www.python.org/ftp/python/3.11.6/python-3.11.6-amd64.exe
+2. Install Microsoft Visual C++ 2015-2022 (x64): https://aka.ms/vs/17/release/vc_redist.x64.exe
+3. Install Microsoft C++ Build Tools: https://visualstudio.microsoft.com/visual-cpp-build-tools/
+Select Desktop development with C++
+4. Install PyTorch: https://pytorch.org/get-started/locally/
+5. Download and unzip Music Source Separation Universal Training Code:
+https://github.com/ZFTurbo/Music-Source-Separation-Training
+6. Open up CMD inside the folder and enter: `pip install -r requirements.txt` 
+7. enter: `pip install wxpython` 
+8. enter: `python gui-wx.py`
+9. Download models - assign the config (.yaml) and checkpoint (.bin, .ckpt, or .th)
+10. Done.
+
+Video guide on [Youtube](https://youtu.be/gyxMCbDwfAw).

requirements.txt

@@ -28,4 +28,5 @@ wandb
 accelerate
 huggingface-hub>=0.23.0
 prodigyopt
-torch_log_wmse
+torch_log_wmse
+wxpython==4.2.2

train.py

@@ -153,10 +153,10 @@ def train_model(args):
 
     # wandb
     if args.wandb_key is None or args.wandb_key.strip() == '':
-        wandb.init(mode = 'disabled')
+        wandb.init(mode='disabled')
     else:
-        wandb.login(key = args.wandb_key)
-        wandb.init(project = 'msst', config = { 'config': config, 'args': args, 'device_ids': device_ids, 'batch_size': batch_size })
+        wandb.login(key=args.wandb_key)
+        wandb.init(project='msst', config={'config': config, 'args': args, 'device_ids': device_ids, 'batch_size': batch_size })
 
     trainset = MSSDataset(
         config,