Augmentation with SpecAugment.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "59572dcc-0a1b-4c64-88a9-3fcc67937582",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Admin\\anaconda3\\Lib\\site-packages\\paramiko\\transport.py:219: CryptographyDeprecationWarning: Blowfish has been deprecated and will be removed in a future release\n",
      "  \"class\": algorithms.Blowfish,\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import librosa\n",
    "import IPython.display as ipd\n",
    "import tensorflow as tf\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01ab22ef-7a0c-41c4-a92f-05681c87c6b1",
   "metadata": {},
   "source": [
    "## Load The Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "7b0c32bf-26f4-460f-a75d-f6ce4b57391a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define the path to the extraction directory\n",
    "dataset_path = 'E:\\\\MS Data Science\\\\SIT 782 Capstone Team Project B\\\\project-echo-data-bucket-3'\n",
    "\n",
    "# Function to pad or truncate spectrograms to a fixed size\n",
    "def pad_spectrogram(spec, target_size):\n",
    "    if spec.shape[1] < target_size[1]:\n",
    "        pad_width = target_size[1] - spec.shape[1]\n",
    "        spec = np.pad(spec, ((0, 0), (0, pad_width)), mode='constant')\n",
    "    else:\n",
    "        spec = spec[:, :target_size[1]]\n",
    "    return spec\n",
    "\n",
    "# Function to extract label from file path\n",
    "def extract_label(file_path):\n",
    "    return os.path.basename(os.path.dirname(file_path))\n",
    "\n",
    "# Function to load dataset\n",
    "def load_dataset(data_dir, target_size):\n",
    "    data = []\n",
    "    labels = []\n",
    "    \n",
    "    # List all file paths in the data directory\n",
    "    for root, _, files in os.walk(data_dir):\n",
    "        for file_name in files:\n",
    "            if file_name.endswith('.mp3'):\n",
    "                file_path = os.path.join(root, file_name)\n",
    "                label = extract_label(file_path)\n",
    "                \n",
    "                # Load audio file\n",
    "                try:\n",
    "                    audio, sr = librosa.load(file_path, sr=None)\n",
    "                    if len(audio) == 0:\n",
    "                        print(f\"Warning: File {file_path} is empty.\")\n",
    "                        continue\n",
    "                    \n",
    "                    # Convert to mel spectrogram\n",
    "                    mel_spec = librosa.feature.melspectrogram(y=audio, sr=sr)\n",
    "                    mel_spec_db = librosa.power_to_db(mel_spec, ref=np.max)\n",
    "                    # Pad or truncate the spectrogram\n",
    "                    mel_spec_db = pad_spectrogram(mel_spec_db, target_size)\n",
    "                    data.append(mel_spec_db)\n",
    "                    labels.append(label)\n",
    "                except Exception as e:\n",
    "                    print(f\"Error processing file {file_path}: {e}\")  # Debug: Print file processing error\n",
    "                    pass  # Skip the file if there's an error\n",
    "    \n",
    "    return np.array(data), np.array(labels)\n",
    "\n",
    "# Define the target size for the spectrograms (height, width)\n",
    "target_size = (128, 128)\n",
    "\n",
    "# Load the dataset\n",
    "X, y_labels = load_dataset(dataset_path, target_size)\n",
    "\n",
    "# Ensure the data has the correct shape (samples, height, width, channels)\n",
    "if X.size == 0 or y_labels.size == 0:\n",
    "    raise ValueError(\"No data loaded. Exiting...\")\n",
    "\n",
    "X = np.expand_dims(X, -1)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ddabac0e-f78d-42e3-9ab9-4e275ac4ec63",
   "metadata": {},
   "source": [
    "## Preprocess the Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "5e6dba77-c850-48a5-9665-0439dad3b457",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Encode labels\n",
    "label_encoder = LabelEncoder()\n",
    "y_encoded = label_encoder.fit_transform(y_labels)\n",
    "\n",
    "# Split your data into training and validation sets\n",
    "X_train, X_val, y_train, y_val = train_test_split(X, y_encoded, test_size=0.2, random_state=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "83969f23-9ae0-4fd7-a0e4-6818ffdd9252",
   "metadata": {},
   "source": [
    "## Define and Apply SpecAugment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "f3fa9997-dbf7-4af4-a4ff-0c0e836809ea",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define SpecAugment functions\n",
    "def apply_spec_augment(spec, time_mask_val=10, freq_mask_val=10):\n",
    "    spec_augmented = spec.copy()\n",
    "    # Apply time mask\n",
    "    num_time_steps = spec_augmented.shape[1]\n",
    "    if num_time_steps > 0:\n",
    "        t = np.random.randint(0, min(time_mask_val, num_time_steps) + 1)\n",
    "        t0 = np.random.randint(0, num_time_steps - t + 1)\n",
    "        spec_augmented[:, t0:t0 + t] = 0\n",
    "    # Apply frequency mask\n",
    "    num_freq_bins = spec_augmented.shape[0]\n",
    "    if num_freq_bins > 0:\n",
    "        f = np.random.randint(0, min(freq_mask_val, num_freq_bins) + 1)\n",
    "        f0 = np.random.randint(0, num_freq_bins - f + 1)\n",
    "        spec_augmented[f0:f0 + f, :] = 0\n",
    "    return spec_augmented"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "64b3f256-ac1e-41b0-8b3e-35679185b421",
   "metadata": {},
   "source": [
    "## Train The Model With SpecAugment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "21cc8225-a0b3-432b-8cb7-11138cc48690",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Function to train the model with SpecAugment\n",
    "def train_model_with_spec_augment(X_train, y_train, X_val, y_val, time_mask_val=30, freq_mask_val=10, epochs=10):\n",
    "    num_classes = len(np.unique(y_train))\n",
    "    input_shape = (X_train.shape[1], X_train.shape[2], 1)\n",
    "    \n",
    "    model = tf.keras.models.Sequential([\n",
    "        tf.keras.layers.Input(shape=input_shape),\n",
    "        tf.keras.layers.Conv2D(32, (3, 3), activation='relu'),\n",
    "        tf.keras.layers.MaxPooling2D((2, 2)),\n",
    "        tf.keras.layers.Flatten(),\n",
    "        tf.keras.layers.Dense(64, activation='relu'),\n",
    "        tf.keras.layers.Dense(num_classes, activation='softmax')\n",
    "    ])\n",
    "    \n",
    "    model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])\n",
    "    \n",
    "    def augment_and_batch(X, y, batch_size=32): #Determine whether additional files should be cached, to have more data, or augmentations performed on the fly\n",
    "        while True:\n",
    "            idx = np.random.randint(0, len(X), batch_size)\n",
    "            X_batch = X[idx]\n",
    "            y_batch = y[idx]\n",
    "            X_batch_augmented = np.array([apply_spec_augment(spec, time_mask_val, freq_mask_val) for spec in X_batch])\n",
    "            X_batch_augmented = np.expand_dims(X_batch_augmented, -1)  # Add channel dimension once\n",
    "            yield X_batch_augmented, y_batch\n",
    "    \n",
    "    batch_size = 32\n",
    "    model.fit(augment_and_batch(X_train, y_train, batch_size), \n",
    "              steps_per_epoch=len(X_train) // batch_size, \n",
    "              validation_data=(X_val, y_val), \n",
    "              epochs=epochs)\n",
    "    \n",
    "    return model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4a6a9200-b821-44d2-948d-644ce435ea17",
   "metadata": {},
   "source": [
    "## Function To Play Augmented Audio"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "c42911d2-f35e-40fe-9839-aba923181ff7",
   "metadata": {},
   "outputs": [],
   "source": [
    "def play_augmented_audio(X): # Play an audio file that has been transformed with this augmentation: \n",
    "    spec = X[0]  # Use the first spectrogram for demonstration purposes\n",
    "    augmented_spec = apply_spec_augment(spec)\n",
    "    \n",
    "    if augmented_spec.size > 0 and np.any(augmented_spec):  # Check if the augmented spectrogram is not empty\n",
    "        try:\n",
    "            audio = librosa.feature.inverse.mel_to_audio(augmented_spec)\n",
    "            ipd.display(ipd.Audio(audio, rate=22050))\n",
    "        except Exception as e:\n",
    "            print(f\"Error converting spectrogram to audio: {e}\")\n",
    "    else:\n",
    "        print(\"Augmented spectrogram is empty, cannot convert to audio.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4766584e-53c5-4b67-ba49-a7f882225792",
   "metadata": {},
   "source": [
    "## Test SpecAugment Parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "9041a528-a45f-426d-b37d-1820c049862f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Testing parameters: time_mask_val=10, freq_mask_val=10\n",
      "Epoch 1/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m12s\u001b[0m 56ms/step - accuracy: 0.0544 - loss: 62.6439 - val_accuracy: 0.0829 - val_loss: 4.6651\n",
      "Epoch 2/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 53ms/step - accuracy: 0.0824 - loss: 4.6456 - val_accuracy: 0.1008 - val_loss: 4.5748\n",
      "Epoch 3/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 53ms/step - accuracy: 0.0788 - loss: 4.5600 - val_accuracy: 0.0829 - val_loss: 4.4920\n",
      "Epoch 4/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 53ms/step - accuracy: 0.0790 - loss: 4.4797 - val_accuracy: 0.0829 - val_loss: 4.4183\n",
      "Epoch 5/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 52ms/step - accuracy: 0.0819 - loss: 4.4062 - val_accuracy: 0.0829 - val_loss: 4.3524\n",
      "Epoch 6/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 52ms/step - accuracy: 0.0877 - loss: 4.3393 - val_accuracy: 0.0829 - val_loss: 4.2937\n",
      "Epoch 7/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 53ms/step - accuracy: 0.0844 - loss: 4.2906 - val_accuracy: 0.0829 - val_loss: 4.2420\n",
      "Epoch 8/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 55ms/step - accuracy: 0.0679 - loss: 4.2625 - val_accuracy: 0.0829 - val_loss: 4.1975\n",
      "Epoch 9/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 56ms/step - accuracy: 0.0810 - loss: 4.1973 - val_accuracy: 0.0829 - val_loss: 4.1566\n",
      "Epoch 10/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 57ms/step - accuracy: 0.0842 - loss: 4.1373 - val_accuracy: 0.0829 - val_loss: 4.1218\n",
      "\u001b[1m46/46\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 15ms/step - accuracy: 0.0645 - loss: 4.1154\n",
      "Testing parameters: time_mask_val=20, freq_mask_val=20\n",
      "Epoch 1/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m12s\u001b[0m 57ms/step - accuracy: 0.0591 - loss: 79.1902 - val_accuracy: 0.1008 - val_loss: 4.6649\n",
      "Epoch 2/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 60ms/step - accuracy: 0.0812 - loss: 4.8368 - val_accuracy: 0.1008 - val_loss: 4.5758\n",
      "Epoch 3/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 61ms/step - accuracy: 0.0894 - loss: 4.5553 - val_accuracy: 0.1008 - val_loss: 4.4928\n",
      "Epoch 4/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 59ms/step - accuracy: 0.0876 - loss: 4.4865 - val_accuracy: 0.1008 - val_loss: 4.4195\n",
      "Epoch 5/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 59ms/step - accuracy: 0.0855 - loss: 4.4132 - val_accuracy: 0.1008 - val_loss: 4.3539\n",
      "Epoch 6/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 61ms/step - accuracy: 0.0823 - loss: 4.3507 - val_accuracy: 0.1008 - val_loss: 4.2949\n",
      "Epoch 7/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 59ms/step - accuracy: 0.0920 - loss: 4.2887 - val_accuracy: 0.1008 - val_loss: 4.2428\n",
      "Epoch 8/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 59ms/step - accuracy: 0.0780 - loss: 4.2417 - val_accuracy: 0.1008 - val_loss: 4.1986\n",
      "Epoch 9/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 60ms/step - accuracy: 0.0867 - loss: 4.1942 - val_accuracy: 0.1008 - val_loss: 4.1596\n",
      "Epoch 10/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 59ms/step - accuracy: 0.0847 - loss: 4.1813 - val_accuracy: 0.1008 - val_loss: 4.1262\n",
      "\u001b[1m46/46\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 12ms/step - accuracy: 0.1050 - loss: 4.1186\n",
      "Testing parameters: time_mask_val=30, freq_mask_val=30\n",
      "Epoch 1/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m12s\u001b[0m 61ms/step - accuracy: 0.0629 - loss: 60.7498 - val_accuracy: 0.1008 - val_loss: 4.6659\n",
      "Epoch 2/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 60ms/step - accuracy: 0.0732 - loss: 4.6397 - val_accuracy: 0.0711 - val_loss: 4.5729\n",
      "Epoch 3/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 60ms/step - accuracy: 0.0753 - loss: 4.5573 - val_accuracy: 0.0711 - val_loss: 4.4903\n",
      "Epoch 4/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 62ms/step - accuracy: 0.0815 - loss: 4.4761 - val_accuracy: 0.1008 - val_loss: 4.4154\n",
      "Epoch 5/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 62ms/step - accuracy: 0.0980 - loss: 4.3891 - val_accuracy: 0.1008 - val_loss: 4.3462\n",
      "Epoch 6/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 62ms/step - accuracy: 0.0904 - loss: 4.3333 - val_accuracy: 0.1008 - val_loss: 4.2873\n",
      "Epoch 7/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 60ms/step - accuracy: 0.0847 - loss: 4.2909 - val_accuracy: 0.1008 - val_loss: 4.2374\n",
      "Epoch 8/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 62ms/step - accuracy: 0.0909 - loss: 4.2431 - val_accuracy: 0.1008 - val_loss: 4.1917\n",
      "Epoch 9/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 63ms/step - accuracy: 0.0910 - loss: 4.1902 - val_accuracy: 0.1008 - val_loss: 4.1524\n",
      "Epoch 10/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 62ms/step - accuracy: 0.0877 - loss: 4.1517 - val_accuracy: 0.1008 - val_loss: 4.1192\n",
      "\u001b[1m46/46\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 13ms/step - accuracy: 0.1050 - loss: 4.1098\n",
      "[(10, 10, 0.08287292718887329), (20, 20, 0.10082872956991196), (30, 30, 0.10082872956991196)]\n"
     ]
    }
   ],
   "source": [
    "# Define a function to test SpecAugment parameters\n",
    "# Determine what parameters work best\n",
    "def test_spec_augment_parameters(X_train, y_train, X_val, y_val): # Determine what parameters work best\n",
    "    params = [(10, 10), (20, 20), (30, 30)]  # Example parameter sets\n",
    "    results = []\n",
    "\n",
    "    for time_mask_val, freq_mask_val in params:\n",
    "        print(f\"Testing parameters: time_mask_val={time_mask_val}, freq_mask_val={freq_mask_val}\")\n",
    "        model = train_model_with_spec_augment(X_train, y_train, X_val, y_val, time_mask_val, freq_mask_val, epochs=10)\n",
    "        accuracy = model.evaluate(X_val, y_val)[1]\n",
    "        results.append((time_mask_val, freq_mask_val, accuracy))\n",
    "\n",
    "    return results\n",
    "\n",
    "# Test SpecAugment parameters\n",
    "results = test_spec_augment_parameters(X_train, y_train, X_val, y_val)\n",
    "print(results)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f49a4b46-0442-4610-b3c8-851595779751",
   "metadata": {},
   "source": [
    "## Visualize Augmentations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "5e41983a-b566-4758-ab24-c77cdc5153e6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def visualize_augmentations(X):\n",
    "    for spec in X[:1]:\n",
    "        augmented_spec = apply_spec_augment(spec)\n",
    "\n",
    "        plt.figure(figsize=(12, 6))\n",
    "        plt.subplot(1, 2, 1)\n",
    "        plt.title(\"Original Spectrogram\")\n",
    "        plt.imshow(spec, aspect='auto', origin='lower')\n",
    "\n",
    "        plt.subplot(1, 2, 2)\n",
    "        plt.title(\"Augmented Spectrogram\")\n",
    "        plt.imshow(augmented_spec, aspect='auto', origin='lower')\n",
    "\n",
    "        plt.show()\n",
    "# Example usage\n",
    "visualize_augmentations(X_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "826514a7-ba2a-4ecb-8141-0c6c3ab511c6",
   "metadata": {},
   "source": [
    "## Collect And Display Metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "b031cbdf-c8c5-4db4-ab14-868e1b67d285",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Testing parameters: time_mask_val=10, freq_mask_val=10\n",
      "Epoch 1/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m12s\u001b[0m 57ms/step - accuracy: 0.0619 - loss: 84.6649 - val_accuracy: 0.0829 - val_loss: 4.6721\n",
      "Epoch 2/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 58ms/step - accuracy: 0.0920 - loss: 4.6491 - val_accuracy: 0.0829 - val_loss: 4.5805\n",
      "Epoch 3/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 58ms/step - accuracy: 0.0723 - loss: 5.0506 - val_accuracy: 0.0829 - val_loss: 4.4975\n",
      "Epoch 4/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 59ms/step - accuracy: 0.0883 - loss: 4.4812 - val_accuracy: 0.0829 - val_loss: 4.4222\n",
      "Epoch 5/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 61ms/step - accuracy: 0.0804 - loss: 4.4049 - val_accuracy: 0.0829 - val_loss: 4.3557\n",
      "Epoch 6/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 63ms/step - accuracy: 0.0912 - loss: 4.3441 - val_accuracy: 0.0829 - val_loss: 4.2973\n",
      "Epoch 7/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 61ms/step - accuracy: 0.0875 - loss: 4.2877 - val_accuracy: 0.0829 - val_loss: 4.2454\n",
      "Epoch 8/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 62ms/step - accuracy: 0.0824 - loss: 4.2362 - val_accuracy: 0.0829 - val_loss: 4.1981\n",
      "Epoch 9/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 63ms/step - accuracy: 0.0834 - loss: 4.1999 - val_accuracy: 0.0829 - val_loss: 4.1578\n",
      "Epoch 10/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 62ms/step - accuracy: 0.0832 - loss: 4.1647 - val_accuracy: 0.0829 - val_loss: 4.1228\n",
      "\u001b[1m46/46\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 12ms/step - accuracy: 0.0645 - loss: 4.1154\n",
      "Testing parameters: time_mask_val=20, freq_mask_val=20\n",
      "Epoch 1/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m13s\u001b[0m 61ms/step - accuracy: 0.0731 - loss: 53.1329 - val_accuracy: 0.0829 - val_loss: 4.6667\n",
      "Epoch 2/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 61ms/step - accuracy: 0.0869 - loss: 4.6430 - val_accuracy: 0.0829 - val_loss: 4.5749\n",
      "Epoch 3/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 62ms/step - accuracy: 0.0766 - loss: 4.5566 - val_accuracy: 0.0829 - val_loss: 4.4917\n",
      "Epoch 4/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 62ms/step - accuracy: 0.0912 - loss: 4.4712 - val_accuracy: 0.0829 - val_loss: 4.4155\n",
      "Epoch 5/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 62ms/step - accuracy: 0.0838 - loss: 4.3972 - val_accuracy: 0.0829 - val_loss: 4.3486\n",
      "Epoch 6/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 62ms/step - accuracy: 0.0853 - loss: 4.3255 - val_accuracy: 0.0829 - val_loss: 4.2906\n",
      "Epoch 7/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 62ms/step - accuracy: 0.0800 - loss: 4.2987 - val_accuracy: 0.0829 - val_loss: 4.2399\n",
      "Epoch 8/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 62ms/step - accuracy: 0.0751 - loss: 4.2190 - val_accuracy: 0.0829 - val_loss: 4.1931\n",
      "Epoch 9/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 63ms/step - accuracy: 0.0835 - loss: 4.2038 - val_accuracy: 0.0711 - val_loss: 4.1548\n",
      "Epoch 10/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 63ms/step - accuracy: 0.0729 - loss: 4.1611 - val_accuracy: 0.0829 - val_loss: 4.1204\n",
      "\u001b[1m46/46\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 14ms/step - accuracy: 0.0645 - loss: 4.1094\n",
      "Testing parameters: time_mask_val=30, freq_mask_val=30\n",
      "Epoch 1/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m13s\u001b[0m 63ms/step - accuracy: 0.0605 - loss: 78.0261 - val_accuracy: 0.1008 - val_loss: 4.6631\n",
      "Epoch 2/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 63ms/step - accuracy: 0.0740 - loss: 4.6397 - val_accuracy: 0.1008 - val_loss: 4.5682\n",
      "Epoch 3/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m12s\u001b[0m 65ms/step - accuracy: 0.0826 - loss: 4.5502 - val_accuracy: 0.1008 - val_loss: 4.4848\n",
      "Epoch 4/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 64ms/step - accuracy: 0.0834 - loss: 4.4703 - val_accuracy: 0.1008 - val_loss: 4.4097\n",
      "Epoch 5/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 63ms/step - accuracy: 0.0780 - loss: 4.3949 - val_accuracy: 0.1008 - val_loss: 4.3430\n",
      "Epoch 6/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 64ms/step - accuracy: 0.0825 - loss: 4.3411 - val_accuracy: 0.0829 - val_loss: 4.2865\n",
      "Epoch 7/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m12s\u001b[0m 64ms/step - accuracy: 0.0775 - loss: 4.3008 - val_accuracy: 0.1008 - val_loss: 4.2366\n",
      "Epoch 8/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 61ms/step - accuracy: 0.0790 - loss: 4.2350 - val_accuracy: 0.0829 - val_loss: 4.1916\n",
      "Epoch 9/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 62ms/step - accuracy: 0.0774 - loss: 4.1855 - val_accuracy: 0.1008 - val_loss: 4.1537\n",
      "Epoch 10/10\n",
      "\u001b[1m180/180\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 64ms/step - accuracy: 0.0929 - loss: 4.1610 - val_accuracy: 0.1008 - val_loss: 4.1210\n",
      "\u001b[1m46/46\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 15ms/step - accuracy: 0.1050 - loss: 4.1133\n",
      "   Time Mask  Freq Mask  Accuracy\n",
      "0         10         10  0.082873\n",
      "1         20         20  0.082873\n",
      "2         30         30  0.100829\n"
     ]
    }
   ],
   "source": [
    "def collect_metrics(X_train, y_train, X_val, y_val): #Figures, stats, dataframes, etc. showing the performance metrics:\n",
    "    results = test_spec_augment_parameters(X_train, y_train, X_val, y_val)\n",
    "    df = pd.DataFrame(results, columns=['Time Mask', 'Freq Mask', 'Accuracy'])\n",
    "    return df\n",
    "\n",
    "# Example usage to collect and display metrics\n",
    "metrics_df = collect_metrics(X_train, y_train, X_val, y_val)\n",
    "print(metrics_df)"
   ]
  }
 ],
 "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": "3.11.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}