loader_tester.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "13693a96",
   "metadata": {},
   "outputs": [],
   "source": [
    "import time\n",
    "import os\n",
    "\n",
    "import IPython.display as ipd\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "import librosa\n",
    "from pathlib import Path,PureWindowsPath,PurePosixPath\n",
    "import torchaudio\n",
    "import torch\n",
    "import matplotlib.pyplot as plt\n",
    "import utils\n",
    "import librosa\n",
    "\n",
    "SAMPLE_RATE = 44100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "48c88185",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\manue\\AppData\\Local\\Temp\\ipykernel_7988\\1660309712.py:1: DtypeWarning: Columns (0,1,5,6,8,12,18,20,21,22,24,33,34,38,39,44,47,49) have mixed types. Specify dtype option on import or set low_memory=False.\n",
      "  tracks = pd.read_csv(\"./fma_metadata/tracks.csv\")\n"
     ]
    }
   ],
   "source": [
    "tracks = pd.read_csv(\"./fma_metadata/tracks.csv\")\n",
    "genres = pd.read_csv(\"./fma_metadata/genres.csv\")\n",
    "tracks.columns=tracks.iloc[0] \n",
    "tracks.columns.values[0] = \"track_id\"\n",
    "tracks.drop([0,1],inplace=True)\n",
    "tracks.track_id = tracks.track_id.astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "d95e540a",
   "metadata": {},
   "outputs": [],
   "source": [
    "def GetGenres(path,genre = 'genre_top'):\n",
    "    id_list = []\n",
    "    genre_list = []\n",
    "    for direc in list(path.iterdir()):\n",
    "        if not direc.is_file():\n",
    "            for file in (list(direc.iterdir())):\n",
    "                id_track = str(file)[-10:-4]\n",
    "                id_list.append(id_track)\n",
    "                genre_list.append(tracks.loc[tracks.track_id == int(id_track),genre].values[0])\n",
    "    return np.asarray(id_list),np.asarray(genre_list)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "741ee1ee",
   "metadata": {},
   "outputs": [],
   "source": [
    "path = Path(\"./fma_small\")\n",
    "id_list, genre_list = GetGenres(path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "282fb0af",
   "metadata": {},
   "outputs": [],
   "source": [
    "def CreateSpectrograms(load_path,save_path, transformation = \"MEL\"):\n",
    "    id_list = []\n",
    "    genre_list = []\n",
    "    if transformation == 'MEL':\n",
    "        transform = torchaudio.transforms.MelSpectrogram(SAMPLE_RATE,n_fft=2048,hop_length=512)\n",
    "    else:\n",
    "        transform = torchaudio.transforms.Spectrogram(SAMPLE_RATE,n_fft=2048,hop_length=512)\n",
    "    for direc in list(load_path.iterdir()):\n",
    "        if not direc.is_file():\n",
    "            for file in (list(direc.iterdir())):\n",
    "                id_track = str(file)[-10:-4]\n",
    "                try:\n",
    "                    waveform, sample_rate = torchaudio.load(file)\n",
    "                    if waveform.shape[0] > 1:\n",
    "                        waveform = (waveform[0] + waveform[1])/2\n",
    "                    spec = transform(waveform)\n",
    "                    torch.save(spec, save_path+\"/\"+id_track+\".pt\")\n",
    "                except:\n",
    "                    pass\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "ac28449e",
   "metadata": {},
   "outputs": [],
   "source": [
    "save = \"./Spectrograms\"\n",
    "CreateSpectrograms(path,save)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0e01ad78",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Prototype, not sure if it works\n",
    "class CustomSpectrogramDataset(Dataset):\n",
    "    def __init__(self, spectrogram,genre, transform=None):\n",
    "        self.x = spectrogram\n",
    "        self.target = genre\n",
    "        self.transform = transform\n",
    "\n",
    "    def __len__(self):\n",
    "        return len(self.img_labels)\n",
    "\n",
    "    def __getitem__(self, idx):\n",
    "        image = self.x[idx]\n",
    "        label = self.target.iloc[idx]\n",
    "        if (self.transform!=None):\n",
    "            image = self.transform(image)\n",
    "        return image, label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "09eba0b8",
   "metadata": {},
   "outputs": [],
   "source": [
    "#path = Path(\"./Spectrograms\")\n",
    "\n",
    "def ChargeDataset(path,id_list,genre_list):\n",
    "    images = []\n",
    "    labels = []\n",
    "    for spec in list(path.iterdir()):\n",
    "        id_track = str(spec)[13:-3]\n",
    "        labels.append(genre_list[np.argwhere(id_list == id_track)][0][0])\n",
    "        images.append(torch.load(spec))\n",
    "    return np.asarray(images),np.asarray(labels)\n",
    "\n",
    "#spectograms, genres = ChargeDataset(path,id_list,genre_list)    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "44e3241f",
   "metadata": {},
   "outputs": [],
   "source": [
    "#PLOT FUNCTIONS\n",
    "\n",
    "def plot_spectrogram(spec, title=None, ylabel=\"freq_bin\", aspect=\"auto\", xmax=None):\n",
    "    fig, axs = plt.subplots(1, 1)\n",
    "    axs.set_title(title or \"Spectrogram (db)\")\n",
    "    axs.set_ylabel(ylabel)\n",
    "    axs.set_xlabel(\"frame\")\n",
    "    im = axs.imshow(librosa.power_to_db(spec), origin=\"lower\", aspect=aspect)\n",
    "    if xmax:\n",
    "        axs.set_xlim((0, xmax))\n",
    "    fig.colorbar(im, ax=axs)\n",
    "    plt.show(block=False)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}