More cleanup

Author: Spijkervet

clmr/datasets/audio.py

@@ -1,9 +1,7 @@
 import os
-import subprocess
-import torchaudio
 from glob import glob
 from torch import Tensor
-from typing import Any, Tuple, Optional
+from typing import Tuple
 
 
 from clmr.datasets import Dataset
@@ -13,14 +11,9 @@ class AUDIO(Dataset):
     """Create a Dataset for any folder of audio files.
     Args:
         root (str): Path to the directory where the dataset is found or downloaded.
-        folder_in_archive (str, optional): The top-level directory of the dataset.
-        subset (str, optional): Which subset of the dataset to use.
-            One of ``"training"``, ``"validation"``, ``"testing"`` or ``None``.
-            If ``None``, the entire dataset is used. (default: ``None``).
+        src_ext_audio (str): The extension of the audio files to analyze. 
     """
 
-    _ext_audio = ".wav"
-
     def __init__(
         self,
         root: str,

config/config.yaml

@@ -8,7 +8,7 @@ dataset_dir: "./data"
 seed: 42
 batch_size: 48
 # max_epochs: 200
-dataset: "magnatagatune" # ["magnatagatune", "msd", "gtzan", "librispeech", "audio"]
+dataset: "magnatagatune" # ["magnatagatune", "msd", "gtzan", "audio"]
 supervised: 0 # train with supervised baseline
 
 # SimCLR model options

linear_evaluation.py

@@ -6,7 +6,6 @@
 from pytorch_lightning import Trainer
 from pytorch_lightning.callbacks import EarlyStopping
 from pytorch_lightning.loggers import TensorBoardLogger
-from tqdm import tqdm
 
 from clmr.datasets import get_dataset
 from clmr.data import ContrastiveDataset