Fix output_root -> self.output_root2

cfg/config.yaml

@@ -26,7 +26,7 @@ datamodule:
     _target_: remfx.datasets.VocalSet
     sample_rate: ${sample_rate}
     root: ${oc.env:DATASET_ROOT}
-    output_root: ${oc.env:OUTPUT_ROOT}
+    output_root: ${oc.env:OUTPUT_ROOT}/train
     chunk_size_in_sec: 6
     mode: "train"
     effect_types: ${effects.train_effects}
@@ -34,7 +34,7 @@ datamodule:
     _target_: remfx.datasets.VocalSet
     sample_rate: ${sample_rate}
     root: ${oc.env:DATASET_ROOT}
-    output_root: ${oc.env:OUTPUT_ROOT}
+    output_root: ${oc.env:OUTPUT_ROOT}/val
     chunk_size_in_sec: 6
     mode: "val"
     effect_types: ${effects.val_effects}

remfx/datasets.py

@@ -207,7 +207,7 @@ class VocalSet(Dataset):
         print("Processing files...")
         if render_files:
             if not self.output_root.exists():
-                self.output_root.mkdir()
+                self.output_root.mkdir(parents=True)
             if not output_mode_path.exists():
                 output_mode_path.mkdir()
             for i, audio_file in tqdm(enumerate(self.files)):
@@ -253,8 +253,8 @@ class VocalSet(Dataset):
 
     def __getitem__(self, idx):
         # Load audio
-        input_file = self.root / "processed" / self.mode / str(idx) / "input.wav"
-        target_file = self.root / "processed" / self.mode / str(idx) / "target.wav"
+        input_file = self.output_root / str(idx) / "input.wav"
+        target_file = self.output_root / str(idx) / "target.wav"
         input, sr = torchaudio.load(input_file)
         target, sr = torchaudio.load(target_file)
         return (input, target, "")