audio_numericMeta.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "COMET WARNING: As you are running in a Jupyter environment, you will need to call `experiment.end()` when finished to ensure all metrics and code are logged before exiting.\n",
      "COMET INFO: Experiment is live on comet.ml https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/f3497400bbf440cfb39abfd2d1f90440\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# import comet_ml at the top of your file\n",
    "from comet_ml import Experiment\n",
    "experiment = Experiment(\n",
    "            project_name=\"music-genre-multiclass-classification\",\n",
    "            workspace=\"wodenwang820118\",\n",
    "        )\n",
    "import comet_ml\n",
    "import logging"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "logging.basicConfig(level=logging.INFO)\n",
    "LOGGER = logging.getLogger(\"comet_ml\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from tensorflow import keras\n",
    "\n",
    "%matplotlib inline\n",
    "sns.set_style('whitegrid')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# tensorflow 2.7 \n",
    "import tensorflow as tf \n",
    "from tensorflow.keras.layers import Input,Flatten,Dense,Dropout,BatchNormalization,Concatenate\n",
    "from tensorflow.keras.models import Model, Sequential"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# scale the numeric data\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "scaler = StandardScaler()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train_features = pd.read_csv('data/train_features.csv')\n",
    "df_train_labels = pd.read_csv('data/train_labels.csv')\n",
    "\n",
    "df_valid_features = pd.read_csv('data/valid_features.csv')\n",
    "df_valid_labels = pd.read_csv('data/valid_labels.csv')\n",
    "\n",
    "df_test_features = pd.read_csv('data/test_features.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>trackID</th>\n",
       "      <th>title</th>\n",
       "      <th>tags</th>\n",
       "      <th>loudness</th>\n",
       "      <th>tempo</th>\n",
       "      <th>time_signature</th>\n",
       "      <th>key</th>\n",
       "      <th>mode</th>\n",
       "      <th>duration</th>\n",
       "      <th>vect_1</th>\n",
       "      <th>...</th>\n",
       "      <th>vect_139</th>\n",
       "      <th>vect_140</th>\n",
       "      <th>vect_141</th>\n",
       "      <th>vect_142</th>\n",
       "      <th>vect_143</th>\n",
       "      <th>vect_144</th>\n",
       "      <th>vect_145</th>\n",
       "      <th>vect_146</th>\n",
       "      <th>vect_147</th>\n",
       "      <th>vect_148</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8424</td>\n",
       "      <td>The Calamity [Explicit]</td>\n",
       "      <td>the, you, to, a, not, in, of, that, are, we, w...</td>\n",
       "      <td>-2.824</td>\n",
       "      <td>131.969</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>298.13506</td>\n",
       "      <td>52.050795</td>\n",
       "      <td>...</td>\n",
       "      <td>0.001097</td>\n",
       "      <td>0.001161</td>\n",
       "      <td>0.001225</td>\n",
       "      <td>0.001251</td>\n",
       "      <td>0.001238</td>\n",
       "      <td>0.001202</td>\n",
       "      <td>0.00113</td>\n",
       "      <td>0.00104</td>\n",
       "      <td>0.104525</td>\n",
       "      <td>0.335818</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1 rows × 157 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   trackID                    title  \\\n",
       "0     8424  The Calamity [Explicit]   \n",
       "\n",
       "                                                tags  loudness    tempo  \\\n",
       "0  the, you, to, a, not, in, of, that, are, we, w...    -2.824  131.969   \n",
       "\n",
       "   time_signature  key  mode   duration     vect_1  ...  vect_139  vect_140  \\\n",
       "0               4    0     1  298.13506  52.050795  ...  0.001097  0.001161   \n",
       "\n",
       "   vect_141  vect_142  vect_143  vect_144  vect_145  vect_146  vect_147  \\\n",
       "0  0.001225  0.001251  0.001238  0.001202   0.00113   0.00104  0.104525   \n",
       "\n",
       "   vect_148  \n",
       "0  0.335818  \n",
       "\n",
       "[1 rows x 157 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train_features.head(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_train_data = df_train_features.iloc[:,9:]\n",
    "num_valid_data = df_valid_features.iloc[:,9:]\n",
    "num_test_data = df_test_features.iloc[:,9:]\n",
    "\n",
    "num_train_meta_data = df_train_features.iloc[:,3:5]\n",
    "num_valid_meta_data = df_valid_features.iloc[:,3:5]\n",
    "\n",
    "num_train_meta_data['duration'] = df_train_features['duration']\n",
    "num_valid_meta_data['duration'] = df_valid_features['duration']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_train = scaler.fit_transform(num_train_data)\n",
    "num_valid = scaler.fit_transform(num_valid_data)\n",
    "\n",
    "num_train_meta = scaler.fit_transform(num_train_meta_data)\n",
    "num_valid_meta = scaler.fit_transform(num_valid_meta_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_train_y = pd.get_dummies(df_train_labels['genre'])\n",
    "num_valid_y = pd.get_dummies(df_valid_labels['genre'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.utils import class_weight\n",
    "class_weight = class_weight.compute_class_weight(class_weight='balanced', classes=np.unique(df_train_labels['genre']),y=df_train_labels['genre'])\n",
    "class_weight = {i :class_weight[i] for i in range(8)}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.callbacks import EarlyStopping\n",
    "early_stop = EarlyStopping(\n",
    "    monitor='val_loss',\n",
    "    patience=5,\n",
    "    restore_best_weights=True,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "class AudioNumMetaClassifier:\n",
    "    def __init__(self,num_train,num_meta_train,num_train_y,num_valid,num_valid_meta,num_valid_y,class_weight,early_stop,experiment):\n",
    "        self.num_train = num_train\n",
    "        self.num_meta_train = num_meta_train    \n",
    "        self.num_train_y = num_train_y\n",
    "        self.num_valid = num_valid\n",
    "        self.num_valid_meta = num_valid_meta\n",
    "        self.num_valid_y = num_valid_y\n",
    "        self.class_weight = class_weight\n",
    "        self.early_stop = early_stop\n",
    "        self.experiment = experiment\n",
    "    \n",
    "    def build_model(self):\n",
    "        # Build the model\n",
    "        model = Sequential()\n",
    "        model.add(\n",
    "            Dense(\n",
    "                units=self.experiment.get_parameter(\"audio_layer_1\"),\n",
    "                activation='elu',\n",
    "                input_shape=(num_train.shape[1],)\n",
    "            )\n",
    "        )\n",
    "        model.add(Dropout(0.5))\n",
    "        model.add(BatchNormalization())\n",
    "\n",
    "        model.add(\n",
    "            Dense(\n",
    "                units=self.experiment.get_parameter(\"audio_layer_2\"),\n",
    "                activation='elu'))\n",
    "        model.add(Dropout(0.3))\n",
    "        model.add(BatchNormalization())\n",
    "\n",
    "        model_meta = Sequential()\n",
    "        model_meta.add(\n",
    "            Dense(\n",
    "                units=self.experiment.get_parameter(\"meta_layer_1\"),\n",
    "                activation='elu',\n",
    "                input_shape=(num_train_meta.shape[1],)\n",
    "            )\n",
    "        )\n",
    "        model_meta.add(Dropout(0.5))\n",
    "        model_meta.add(BatchNormalization())\n",
    "\n",
    "        model_meta.add(\n",
    "            Dense(\n",
    "                units=self.experiment.get_parameter(\"meta_layer_2\"),\n",
    "                activation='elu'))\n",
    "        model_meta.add(Dropout(0.3))\n",
    "        model_meta.add(BatchNormalization())\n",
    "\n",
    "\n",
    "        merged_out = Concatenate()([model.output,model_meta.output])\n",
    "        merged_out = Dense(\n",
    "            units=self.experiment.get_parameter(\"final_layer\"),\n",
    "            activation='elu')(merged_out)\n",
    "        merged_out = Dense(8,activation='softmax')(merged_out)\n",
    "\n",
    "        combined_model = Model([model.input] + [model_meta.input], [merged_out])\n",
    "        combined_model.compile(\n",
    "            optimizer='adam',\n",
    "            loss='categorical_crossentropy',\n",
    "            metrics=['accuracy']\n",
    "        )\n",
    "        return combined_model\n",
    "    \n",
    "    def train_model(self):\n",
    "        # Train the model\n",
    "        model = self.build_model()\n",
    "        model.fit(\n",
    "            [self.num_train,self.num_meta_train],\n",
    "            self.num_train_y,\n",
    "            batch_size=self.experiment.get_parameter(\"batch_size\"),\n",
    "            epochs=self.experiment.get_parameter(\"epochs\"),\n",
    "            validation_data=([self.num_valid,self.num_valid_meta],self.num_valid_y),\n",
    "            class_weight=self.class_weight,\n",
    "            callbacks=[self.early_stop],\n",
    "            shuffle=True\n",
    "        )\n",
    "        return model\n",
    "    \n",
    "    def evaluate_model(self):\n",
    "        # Evaluate the model\n",
    "        model = self.train_model()\n",
    "        score = model.evaluate([self.num_valid,self.num_valid_meta],self.num_valid_y)\n",
    "        LOGGER.info(f\"{ score }\")\n",
    "    \n",
    "    def grid_search(self, config_dict):\n",
    "        opt = comet_ml.Optimizer(config_dict)\n",
    "        for self.experiment in opt.get_experiments(project_name=\"music-genre-multiclass-classification\"):\n",
    "\n",
    "            self.build_model()\n",
    "            self.train_model()\n",
    "            self.evaluate_model()\n",
    "            self.experiment.end()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "classifier = AudioNumMetaClassifier(num_train,num_train_meta,num_train_y,num_valid,num_valid_meta,num_valid_y,class_weight,early_stop,experiment)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "COMET INFO: COMET_OPTIMIZER_ID=019075b278c649c7b9842acd5f3dcf6d\n",
      "COMET INFO: Using optimizer config: {'algorithm': 'bayes', 'configSpaceSize': 141120, 'endTime': None, 'id': '019075b278c649c7b9842acd5f3dcf6d', 'lastUpdateTime': None, 'maxCombo': 10, 'name': 'Optimize Music Classification Network', 'parameters': {'audio_layer_1': {'type': 'discrete', 'values': [400, 450, 500, 550, 600, 700, 900, 1000]}, 'audio_layer_2': {'type': 'discrete', 'values': [128, 256, 300, 400, 450, 500, 550]}, 'batch_size': {'type': 'discrete', 'values': [16, 32]}, 'epochs': {'type': 'discrete', 'values': [15, 20, 25, 30]}, 'final_layer': {'type': 'discrete', 'values': [20, 32, 64, 128, 256]}, 'meta_layer_1': {'type': 'discrete', 'values': [300, 400, 450, 500, 550, 600, 700, 900, 1000]}, 'meta_layer_2': {'type': 'discrete', 'values': [128, 256, 300, 400, 450, 500, 550]}}, 'predictor': None, 'spec': {'gridSize': 10, 'maxCombo': 10, 'metric': 'loss', 'minSampleSize': 100, 'objective': 'minimize', 'retryAssignLimit': 0, 'retryLimit': 1000}, 'startTime': 26934008029, 'state': {'mode': None, 'seed': None, 'sequence': [], 'sequence_i': 0, 'sequence_pid': None, 'sequence_retry': 0, 'sequence_retry_count': 0}, 'status': 'running', 'suggestion_count': 0, 'trials': 1, 'version': '2.0.1'}\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Comet.ml Experiment Summary\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO:   Data:\n",
      "COMET INFO:     display_summary_level : 1\n",
      "COMET INFO:     url                   : https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/f3497400bbf440cfb39abfd2d1f90440\n",
      "COMET INFO:   Parameters:\n",
      "COMET INFO:     copy      : True\n",
      "COMET INFO:     with_mean : True\n",
      "COMET INFO:     with_std  : True\n",
      "COMET INFO:   Uploads:\n",
      "COMET INFO:     environment details      : 1\n",
      "COMET INFO:     filename                 : 1\n",
      "COMET INFO:     git metadata             : 1\n",
      "COMET INFO:     git-patch (uncompressed) : 1 (826 bytes)\n",
      "COMET INFO:     installed packages       : 1\n",
      "COMET INFO:     notebook                 : 1\n",
      "COMET INFO:     source_code              : 1\n",
      "COMET INFO: ---------------------------\n",
      "COMET WARNING: As you are running in a Jupyter environment, you will need to call `experiment.end()` when finished to ensure all metrics and code are logged before exiting.\n",
      "COMET INFO: Experiment is live on comet.ml https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/80076188e4d94fb3ac9eef9665f17b31\n",
      "\n",
      "COMET INFO: Ignoring automatic log_parameter('verbose') because 'keras:verbose' is in COMET_LOGGING_PARAMETERS_IGNORE\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/15\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "COMET INFO: ignoring tensorflow summary log of metrics because of keras; set `comet_ml.loggers.tensorboard_logger.LOG_METRICS = True` to override\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "240/240 [==============================] - 3s 6ms/step - loss: 1.6898 - accuracy: 0.4066 - val_loss: 1.5073 - val_accuracy: 0.4533\n",
      "Epoch 2/15\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 1.3947 - accuracy: 0.4931 - val_loss: 1.5213 - val_accuracy: 0.4378\n",
      "Epoch 3/15\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 1.3293 - accuracy: 0.5059 - val_loss: 1.3605 - val_accuracy: 0.5267\n",
      "Epoch 4/15\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 1.2713 - accuracy: 0.5313 - val_loss: 1.3643 - val_accuracy: 0.5156\n",
      "Epoch 5/15\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 1.2181 - accuracy: 0.5376 - val_loss: 1.4668 - val_accuracy: 0.4467\n",
      "Epoch 6/15\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 1.1837 - accuracy: 0.5563 - val_loss: 1.3697 - val_accuracy: 0.4867\n",
      "Epoch 7/15\n",
      "240/240 [==============================] - 2s 8ms/step - loss: 1.1745 - accuracy: 0.5621 - val_loss: 1.3660 - val_accuracy: 0.5089\n",
      "Epoch 8/15\n",
      "240/240 [==============================] - 3s 12ms/step - loss: 1.1425 - accuracy: 0.5720 - val_loss: 1.4367 - val_accuracy: 0.4689\n",
      "Epoch 1/15\n",
      "240/240 [==============================] - 5s 14ms/step - loss: 1.6689 - accuracy: 0.4122 - val_loss: 1.4070 - val_accuracy: 0.4956\n",
      "Epoch 2/15\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.3994 - accuracy: 0.4829 - val_loss: 1.4742 - val_accuracy: 0.4889\n",
      "Epoch 3/15\n",
      "240/240 [==============================] - 3s 12ms/step - loss: 1.3158 - accuracy: 0.5134 - val_loss: 1.5158 - val_accuracy: 0.4733\n",
      "Epoch 4/15\n",
      "240/240 [==============================] - 3s 14ms/step - loss: 1.2652 - accuracy: 0.5294 - val_loss: 1.4270 - val_accuracy: 0.5133\n",
      "Epoch 5/15\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.2335 - accuracy: 0.5372 - val_loss: 1.3487 - val_accuracy: 0.5333\n",
      "Epoch 6/15\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.1994 - accuracy: 0.5513 - val_loss: 1.4538 - val_accuracy: 0.4733\n",
      "Epoch 7/15\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.1659 - accuracy: 0.5610 - val_loss: 1.4652 - val_accuracy: 0.4689\n",
      "Epoch 8/15\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.1357 - accuracy: 0.5715 - val_loss: 1.3494 - val_accuracy: 0.5311\n",
      "Epoch 9/15\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.1119 - accuracy: 0.5765 - val_loss: 1.2989 - val_accuracy: 0.5444\n",
      "Epoch 10/15\n",
      "240/240 [==============================] - 3s 12ms/step - loss: 1.0965 - accuracy: 0.5788 - val_loss: 1.4960 - val_accuracy: 0.4644\n",
      "Epoch 11/15\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.0849 - accuracy: 0.5830 - val_loss: 1.4930 - val_accuracy: 0.4844\n",
      "Epoch 12/15\n",
      "240/240 [==============================] - 3s 14ms/step - loss: 1.0614 - accuracy: 0.5909 - val_loss: 1.3595 - val_accuracy: 0.5289\n",
      "Epoch 13/15\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.0370 - accuracy: 0.6003 - val_loss: 1.3226 - val_accuracy: 0.5333\n",
      "Epoch 14/15\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.0046 - accuracy: 0.6081 - val_loss: 1.4162 - val_accuracy: 0.5067\n",
      "15/15 [==============================] - 0s 5ms/step - loss: 1.2989 - accuracy: 0.5444\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "COMET INFO: [1.2989434003829956, 0.5444444417953491]\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Comet.ml Experiment Summary\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO:   Data:\n",
      "COMET INFO:     display_summary_level : 1\n",
      "COMET INFO:     url                   : https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/80076188e4d94fb3ac9eef9665f17b31\n",
      "COMET INFO:   Metrics [count] (min, max):\n",
      "COMET INFO:     accuracy [22]                : (0.40661630034446716, 0.6081010699272156)\n",
      "COMET INFO:     batch_accuracy [528]         : (0.15625, 0.71875)\n",
      "COMET INFO:     batch_loss [528]             : (0.7330235242843628, 2.6585617065429688)\n",
      "COMET INFO:     epoch_duration [22]          : (1.1560000000026776, 4.76600000000326)\n",
      "COMET INFO:     loss [22]                    : (1.0046277046203613, 1.689835548400879)\n",
      "COMET INFO:     val_accuracy [22]            : (0.4377777874469757, 0.5444444417953491)\n",
      "COMET INFO:     val_loss [22]                : (1.2989434003829956, 1.5212737321853638)\n",
      "COMET INFO:     validate_batch_accuracy [44] : (0.3125, 0.53125)\n",
      "COMET INFO:     validate_batch_loss [44]     : (1.3105552196502686, 1.7266346216201782)\n",
      "COMET INFO:   Others:\n",
      "COMET INFO:     optimizer_count        : 1\n",
      "COMET INFO:     optimizer_id           : 019075b278c649c7b9842acd5f3dcf6d\n",
      "COMET INFO:     optimizer_metric       : loss\n",
      "COMET INFO:     optimizer_metric_value : 1.0046277046203613\n",
      "COMET INFO:     optimizer_objective    : minimum\n",
      "COMET INFO:     optimizer_parameters   : {\"audio_layer_1\": 700, \"audio_layer_2\": 128, \"batch_size\": 32, \"epochs\": 15, \"final_layer\": 32, \"meta_layer_1\": 1000, \"meta_layer_2\": 300}\n",
      "COMET INFO:     optimizer_pid          : a031f295c7725e16430e0b68eb8ef3a956b3ba72\n",
      "COMET INFO:     optimizer_process      : 13176\n",
      "COMET INFO:     optimizer_trial        : 1\n",
      "COMET INFO:     optimizer_version      : 2.0.1\n",
      "COMET INFO:     trainable_params       : 520832\n",
      "COMET INFO:   Parameters:\n",
      "COMET INFO:     Adam_amsgrad       : False\n",
      "COMET INFO:     Adam_beta_1        : 0.9\n",
      "COMET INFO:     Adam_beta_2        : 0.999\n",
      "COMET INFO:     Adam_decay         : 0.0\n",
      "COMET INFO:     Adam_epsilon       : 1e-07\n",
      "COMET INFO:     Adam_learning_rate : 0.001\n",
      "COMET INFO:     Adam_name          : Adam\n",
      "COMET INFO:     Optimizer          : Adam\n",
      "COMET INFO:     audio_layer_1      : 700\n",
      "COMET INFO:     audio_layer_2      : 128\n",
      "COMET INFO:     batch_size         : 32\n",
      "COMET INFO:     epochs             : 15\n",
      "COMET INFO:     final_layer        : 32\n",
      "COMET INFO:     meta_layer_1       : 1000\n",
      "COMET INFO:     meta_layer_2       : 300\n",
      "COMET INFO:     steps              : 240\n",
      "COMET INFO:   Uploads:\n",
      "COMET INFO:     environment details      : 1\n",
      "COMET INFO:     filename                 : 1\n",
      "COMET INFO:     git metadata             : 1\n",
      "COMET INFO:     git-patch (uncompressed) : 1 (826 bytes)\n",
      "COMET INFO:     installed packages       : 1\n",
      "COMET INFO:     model graph              : 1\n",
      "COMET INFO:     notebook                 : 1\n",
      "COMET INFO:     source_code              : 1\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Uploading metrics, params, and assets to Comet before program termination (may take several seconds)\n",
      "COMET INFO: The Python SDK has 3600 seconds to finish before aborting...\n",
      "COMET INFO: Waiting for completion of the file uploads (may take several seconds)\n",
      "COMET INFO: The Python SDK has 10800 seconds to finish before aborting...\n",
      "COMET INFO: Still uploading 2 file(s), remaining 16.23 KB/16.82 KB\n",
      "COMET WARNING: As you are running in a Jupyter environment, you will need to call `experiment.end()` when finished to ensure all metrics and code are logged before exiting.\n",
      "COMET INFO: Experiment is live on comet.ml https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/9a045464ff984548a40592d25148c1f1\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "240/240 [==============================] - 5s 14ms/step - loss: 1.6878 - accuracy: 0.4116 - val_loss: 1.6219 - val_accuracy: 0.3956\n",
      "Epoch 2/20\n",
      "240/240 [==============================] - 3s 12ms/step - loss: 1.4188 - accuracy: 0.4734 - val_loss: 1.4126 - val_accuracy: 0.5000\n",
      "Epoch 3/20\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.3347 - accuracy: 0.5060 - val_loss: 1.3865 - val_accuracy: 0.4933\n",
      "Epoch 4/20\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.2740 - accuracy: 0.5285 - val_loss: 1.3690 - val_accuracy: 0.4889\n",
      "Epoch 5/20\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.2297 - accuracy: 0.5379 - val_loss: 1.3468 - val_accuracy: 0.5067\n",
      "Epoch 6/20\n",
      "240/240 [==============================] - 3s 12ms/step - loss: 1.2033 - accuracy: 0.5498 - val_loss: 1.4357 - val_accuracy: 0.4756\n",
      "Epoch 7/20\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.1670 - accuracy: 0.5567 - val_loss: 1.3776 - val_accuracy: 0.5089\n",
      "Epoch 8/20\n",
      "240/240 [==============================] - 3s 12ms/step - loss: 1.1300 - accuracy: 0.5689 - val_loss: 1.3957 - val_accuracy: 0.5156\n",
      "Epoch 9/20\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.1198 - accuracy: 0.5662 - val_loss: 1.3636 - val_accuracy: 0.5222\n",
      "Epoch 10/20\n",
      "240/240 [==============================] - 3s 14ms/step - loss: 1.1020 - accuracy: 0.5804 - val_loss: 1.3811 - val_accuracy: 0.4978\n",
      "Epoch 1/20\n",
      "240/240 [==============================] - 5s 14ms/step - loss: 1.7106 - accuracy: 0.4163 - val_loss: 1.4114 - val_accuracy: 0.4756\n",
      "Epoch 2/20\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.3871 - accuracy: 0.4934 - val_loss: 1.3739 - val_accuracy: 0.5156\n",
      "Epoch 3/20\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.3160 - accuracy: 0.5178 - val_loss: 1.3629 - val_accuracy: 0.5133\n",
      "Epoch 4/20\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.2413 - accuracy: 0.5406 - val_loss: 1.4622 - val_accuracy: 0.5178\n",
      "Epoch 5/20\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.2193 - accuracy: 0.5447 - val_loss: 1.3477 - val_accuracy: 0.5200\n",
      "Epoch 6/20\n",
      "240/240 [==============================] - 3s 14ms/step - loss: 1.1656 - accuracy: 0.5574 - val_loss: 1.4462 - val_accuracy: 0.4933\n",
      "Epoch 7/20\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.1550 - accuracy: 0.5693 - val_loss: 1.4778 - val_accuracy: 0.5044\n",
      "Epoch 8/20\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.1200 - accuracy: 0.5768 - val_loss: 1.5492 - val_accuracy: 0.4889\n",
      "Epoch 9/20\n",
      "240/240 [==============================] - 2s 10ms/step - loss: 1.1024 - accuracy: 0.5788 - val_loss: 1.4186 - val_accuracy: 0.5000\n",
      "Epoch 10/20\n",
      "240/240 [==============================] - 3s 11ms/step - loss: 1.0607 - accuracy: 0.5914 - val_loss: 1.5060 - val_accuracy: 0.4800\n",
      "15/15 [==============================] - 0s 3ms/step - loss: 1.3477 - accuracy: 0.5200\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "COMET INFO: [1.3477267026901245, 0.5199999809265137]\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Comet.ml Experiment Summary\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO:   Data:\n",
      "COMET INFO:     display_summary_level : 1\n",
      "COMET INFO:     url                   : https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/9a045464ff984548a40592d25148c1f1\n",
      "COMET INFO:   Metrics [count] (min, max):\n",
      "COMET INFO:     accuracy [20]                : (0.41156551241874695, 0.5914300680160522)\n",
      "COMET INFO:     batch_accuracy [480]         : (0.09375, 0.71875)\n",
      "COMET INFO:     batch_loss [480]             : (0.8460208177566528, 3.0483016967773438)\n",
      "COMET INFO:     epoch_duration [20]          : (2.312999999994645, 5.0)\n",
      "COMET INFO:     loss [20]                    : (1.060660719871521, 1.710648775100708)\n",
      "COMET INFO:     val_accuracy [20]            : (0.3955555558204651, 0.5222222208976746)\n",
      "COMET INFO:     val_loss [20]                : (1.3467884063720703, 1.6219391822814941)\n",
      "COMET INFO:     validate_batch_accuracy [40] : (0.3125, 0.59375)\n",
      "COMET INFO:     validate_batch_loss [40]     : (1.3100355863571167, 1.6856769323349)\n",
      "COMET INFO:   Others:\n",
      "COMET INFO:     optimizer_count        : 2\n",
      "COMET INFO:     optimizer_id           : 019075b278c649c7b9842acd5f3dcf6d\n",
      "COMET INFO:     optimizer_metric       : loss\n",
      "COMET INFO:     optimizer_metric_value : 1.060660719871521\n",
      "COMET INFO:     optimizer_objective    : minimum\n",
      "COMET INFO:     optimizer_parameters   : {\"audio_layer_1\": 500, \"audio_layer_2\": 450, \"batch_size\": 32, \"epochs\": 20, \"final_layer\": 20, \"meta_layer_1\": 700, \"meta_layer_2\": 500}\n",
      "COMET INFO:     optimizer_pid          : 1a6c0cee744d416de3b2d41ba82ca5afc468cfaf\n",
      "COMET INFO:     optimizer_process      : 13176\n",
      "COMET INFO:     optimizer_trial        : 1\n",
      "COMET INFO:     optimizer_version      : 2.0.1\n",
      "COMET INFO:     trainable_params       : 681038\n",
      "COMET INFO:   Parameters:\n",
      "COMET INFO:     Adam_amsgrad       : False\n",
      "COMET INFO:     Adam_beta_1        : 0.9\n",
      "COMET INFO:     Adam_beta_2        : 0.999\n",
      "COMET INFO:     Adam_decay         : 0.0\n",
      "COMET INFO:     Adam_epsilon       : 1e-07\n",
      "COMET INFO:     Adam_learning_rate : 0.001\n",
      "COMET INFO:     Adam_name          : Adam\n",
      "COMET INFO:     Optimizer          : Adam\n",
      "COMET INFO:     audio_layer_1      : 500\n",
      "COMET INFO:     audio_layer_2      : 450\n",
      "COMET INFO:     batch_size         : 32\n",
      "COMET INFO:     epochs             : 20\n",
      "COMET INFO:     final_layer        : 20\n",
      "COMET INFO:     meta_layer_1       : 700\n",
      "COMET INFO:     meta_layer_2       : 500\n",
      "COMET INFO:     steps              : 240\n",
      "COMET INFO:   Uploads:\n",
      "COMET INFO:     environment details      : 1\n",
      "COMET INFO:     filename                 : 1\n",
      "COMET INFO:     git metadata             : 1\n",
      "COMET INFO:     git-patch (uncompressed) : 1 (826 bytes)\n",
      "COMET INFO:     installed packages       : 1\n",
      "COMET INFO:     model graph              : 1\n",
      "COMET INFO:     notebook                 : 1\n",
      "COMET INFO:     source_code              : 1\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Uploading 1 metrics, params and output messages\n",
      "COMET INFO: Waiting for completion of the file uploads (may take several seconds)\n",
      "COMET INFO: The Python SDK has 10800 seconds to finish before aborting...\n",
      "COMET INFO: Still uploading 2 file(s), remaining 16.23 KB/16.82 KB\n",
      "COMET WARNING: As you are running in a Jupyter environment, you will need to call `experiment.end()` when finished to ensure all metrics and code are logged before exiting.\n",
      "COMET INFO: Experiment is live on comet.ml https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/54fa21a4eee043c29f69f0b21acf8244\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "240/240 [==============================] - 2s 7ms/step - loss: 1.6806 - accuracy: 0.4127 - val_loss: 1.5244 - val_accuracy: 0.4644\n",
      "Epoch 2/20\n",
      "240/240 [==============================] - 1s 6ms/step - loss: 1.4118 - accuracy: 0.4799 - val_loss: 1.4139 - val_accuracy: 0.5178\n",
      "Epoch 3/20\n",
      "240/240 [==============================] - 1s 6ms/step - loss: 1.3222 - accuracy: 0.5090 - val_loss: 1.4189 - val_accuracy: 0.5067\n",
      "Epoch 4/20\n",
      "240/240 [==============================] - 1s 6ms/step - loss: 1.2713 - accuracy: 0.5245 - val_loss: 1.4601 - val_accuracy: 0.4867\n",
      "Epoch 5/20\n",
      "240/240 [==============================] - 2s 7ms/step - loss: 1.2078 - accuracy: 0.5465 - val_loss: 1.4458 - val_accuracy: 0.4822\n",
      "Epoch 6/20\n",
      "240/240 [==============================] - 1s 6ms/step - loss: 1.1576 - accuracy: 0.5503 - val_loss: 1.4891 - val_accuracy: 0.4756\n",
      "Epoch 7/20\n",
      "240/240 [==============================] - 1s 6ms/step - loss: 1.1509 - accuracy: 0.5680 - val_loss: 1.4009 - val_accuracy: 0.5044\n",
      "Epoch 8/20\n",
      "240/240 [==============================] - 1s 6ms/step - loss: 1.1256 - accuracy: 0.5780 - val_loss: 1.5089 - val_accuracy: 0.4689\n",
      "Epoch 9/20\n",
      "240/240 [==============================] - 2s 7ms/step - loss: 1.0978 - accuracy: 0.5843 - val_loss: 1.4226 - val_accuracy: 0.4911\n",
      "Epoch 10/20\n",
      "240/240 [==============================] - 2s 7ms/step - loss: 1.0581 - accuracy: 0.5934 - val_loss: 1.3621 - val_accuracy: 0.5244\n",
      "Epoch 11/20\n",
      "240/240 [==============================] - 3s 12ms/step - loss: 1.0461 - accuracy: 0.5968 - val_loss: 1.5120 - val_accuracy: 0.4733\n",
      "Epoch 12/20\n",
      "240/240 [==============================] - 3s 12ms/step - loss: 1.0336 - accuracy: 0.6045 - val_loss: 1.3449 - val_accuracy: 0.5311\n",
      "Epoch 13/20\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.0201 - accuracy: 0.6052 - val_loss: 1.4063 - val_accuracy: 0.5089\n",
      "Epoch 14/20\n",
      "240/240 [==============================] - 3s 12ms/step - loss: 0.9976 - accuracy: 0.6151 - val_loss: 1.4706 - val_accuracy: 0.4756\n",
      "Epoch 15/20\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 0.9917 - accuracy: 0.6157 - val_loss: 1.3615 - val_accuracy: 0.5067\n",
      "Epoch 16/20\n",
      "240/240 [==============================] - 3s 12ms/step - loss: 0.9767 - accuracy: 0.6193 - val_loss: 1.4334 - val_accuracy: 0.5022\n",
      "Epoch 17/20\n",
      "240/240 [==============================] - 3s 14ms/step - loss: 0.9555 - accuracy: 0.6198 - val_loss: 1.4567 - val_accuracy: 0.5022\n",
      "Epoch 1/20\n",
      "240/240 [==============================] - 4s 13ms/step - loss: 1.6645 - accuracy: 0.4117 - val_loss: 1.3850 - val_accuracy: 0.5311\n",
      "Epoch 2/20\n",
      "240/240 [==============================] - 3s 12ms/step - loss: 1.3958 - accuracy: 0.4883 - val_loss: 1.3806 - val_accuracy: 0.5022\n",
      "Epoch 3/20\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.3173 - accuracy: 0.5069 - val_loss: 1.2356 - val_accuracy: 0.5578\n",
      "Epoch 4/20\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.2368 - accuracy: 0.5362 - val_loss: 1.4589 - val_accuracy: 0.4867\n",
      "Epoch 5/20\n",
      "240/240 [==============================] - 3s 12ms/step - loss: 1.2082 - accuracy: 0.5430 - val_loss: 1.4152 - val_accuracy: 0.5133\n",
      "Epoch 6/20\n",
      "240/240 [==============================] - 3s 12ms/step - loss: 1.1862 - accuracy: 0.5473 - val_loss: 1.3385 - val_accuracy: 0.5267\n",
      "Epoch 7/20\n",
      "240/240 [==============================] - 3s 12ms/step - loss: 1.1605 - accuracy: 0.5600 - val_loss: 1.2720 - val_accuracy: 0.5444\n",
      "Epoch 8/20\n",
      "240/240 [==============================] - 3s 12ms/step - loss: 1.1256 - accuracy: 0.5706 - val_loss: 1.4641 - val_accuracy: 0.4844\n",
      "15/15 [==============================] - 0s 6ms/step - loss: 1.2356 - accuracy: 0.5578\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "COMET INFO: [1.23561429977417, 0.5577777624130249]\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Comet.ml Experiment Summary\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO:   Data:\n",
      "COMET INFO:     display_summary_level : 1\n",
      "COMET INFO:     url                   : https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/54fa21a4eee043c29f69f0b21acf8244\n",
      "COMET INFO:   Metrics [count] (min, max):\n",
      "COMET INFO:     accuracy [25]                : (0.4116957485675812, 0.6198228597640991)\n",
      "COMET INFO:     batch_accuracy [600]         : (0.0625, 0.78125)\n",
      "COMET INFO:     batch_loss [600]             : (0.5944318771362305, 2.8123703002929688)\n",
      "COMET INFO:     epoch_duration [25]          : (1.3440000000118744, 4.437999999994645)\n",
      "COMET INFO:     loss [25]                    : (0.9554587006568909, 1.6805789470672607)\n",
      "COMET INFO:     val_accuracy [25]            : (0.46444445848464966, 0.5577777624130249)\n",
      "COMET INFO:     val_loss [25]                : (1.23561429977417, 1.5243898630142212)\n",
      "COMET INFO:     validate_batch_accuracy [50] : (0.34375, 0.59375)\n",
      "COMET INFO:     validate_batch_loss [50]     : (1.1763099431991577, 1.7133269309997559)\n",
      "COMET INFO:   Others:\n",
      "COMET INFO:     optimizer_count        : 3\n",
      "COMET INFO:     optimizer_id           : 019075b278c649c7b9842acd5f3dcf6d\n",
      "COMET INFO:     optimizer_metric       : loss\n",
      "COMET INFO:     optimizer_metric_value : 0.9554587006568909\n",
      "COMET INFO:     optimizer_objective    : minimum\n",
      "COMET INFO:     optimizer_parameters   : {\"audio_layer_1\": 450, \"audio_layer_2\": 256, \"batch_size\": 32, \"epochs\": 20, \"final_layer\": 32, \"meta_layer_1\": 900, \"meta_layer_2\": 400}\n",
      "COMET INFO:     optimizer_pid          : e73bb8daad4732c7728f3b0be283a611411e7d18\n",
      "COMET INFO:     optimizer_process      : 13176\n",
      "COMET INFO:     optimizer_trial        : 1\n",
      "COMET INFO:     optimizer_version      : 2.0.1\n",
      "COMET INFO:     trainable_params       : 575818\n",
      "COMET INFO:   Parameters:\n",
      "COMET INFO:     Adam_amsgrad       : False\n",
      "COMET INFO:     Adam_beta_1        : 0.9\n",
      "COMET INFO:     Adam_beta_2        : 0.999\n",
      "COMET INFO:     Adam_decay         : 0.0\n",
      "COMET INFO:     Adam_epsilon       : 1e-07\n",
      "COMET INFO:     Adam_learning_rate : 0.001\n",
      "COMET INFO:     Adam_name          : Adam\n",
      "COMET INFO:     Optimizer          : Adam\n",
      "COMET INFO:     audio_layer_1      : 450\n",
      "COMET INFO:     audio_layer_2      : 256\n",
      "COMET INFO:     batch_size         : 32\n",
      "COMET INFO:     epochs             : 20\n",
      "COMET INFO:     final_layer        : 32\n",
      "COMET INFO:     meta_layer_1       : 900\n",
      "COMET INFO:     meta_layer_2       : 400\n",
      "COMET INFO:     steps              : 240\n",
      "COMET INFO:   Uploads:\n",
      "COMET INFO:     environment details      : 1\n",
      "COMET INFO:     filename                 : 1\n",
      "COMET INFO:     git metadata             : 1\n",
      "COMET INFO:     git-patch (uncompressed) : 1 (826 bytes)\n",
      "COMET INFO:     installed packages       : 1\n",
      "COMET INFO:     model graph              : 1\n",
      "COMET INFO:     notebook                 : 1\n",
      "COMET INFO:     source_code              : 1\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Uploading 1 metrics, params and output messages\n",
      "COMET INFO: Waiting for completion of the file uploads (may take several seconds)\n",
      "COMET INFO: The Python SDK has 10800 seconds to finish before aborting...\n",
      "COMET INFO: Still uploading 2 file(s), remaining 16.23 KB/16.82 KB\n",
      "COMET WARNING: As you are running in a Jupyter environment, you will need to call `experiment.end()` when finished to ensure all metrics and code are logged before exiting.\n",
      "COMET INFO: Experiment is live on comet.ml https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/65d35781be3c45dbab9ab59059959fd1\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/25\n",
      "480/480 [==============================] - 8s 13ms/step - loss: 1.6932 - accuracy: 0.4116 - val_loss: 1.4563 - val_accuracy: 0.4867\n",
      "Epoch 2/25\n",
      "480/480 [==============================] - 6s 12ms/step - loss: 1.4094 - accuracy: 0.4750 - val_loss: 1.5252 - val_accuracy: 0.4422\n",
      "Epoch 3/25\n",
      "480/480 [==============================] - 6s 12ms/step - loss: 1.3349 - accuracy: 0.5026 - val_loss: 1.4524 - val_accuracy: 0.5000\n",
      "Epoch 4/25\n",
      "480/480 [==============================] - 6s 12ms/step - loss: 1.3040 - accuracy: 0.5040 - val_loss: 1.3907 - val_accuracy: 0.5067\n",
      "Epoch 5/25\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.2579 - accuracy: 0.5238 - val_loss: 1.4341 - val_accuracy: 0.4978\n",
      "Epoch 6/25\n",
      "480/480 [==============================] - 6s 12ms/step - loss: 1.2279 - accuracy: 0.5378 - val_loss: 1.5357 - val_accuracy: 0.4844\n",
      "Epoch 7/25\n",
      "480/480 [==============================] - 6s 12ms/step - loss: 1.1995 - accuracy: 0.5431 - val_loss: 1.4304 - val_accuracy: 0.5089\n",
      "Epoch 8/25\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.1701 - accuracy: 0.5491 - val_loss: 1.5403 - val_accuracy: 0.4622\n",
      "Epoch 9/25\n",
      "480/480 [==============================] - 6s 12ms/step - loss: 1.1562 - accuracy: 0.5567 - val_loss: 1.5747 - val_accuracy: 0.4667\n",
      "Epoch 1/25\n",
      "480/480 [==============================] - 8s 14ms/step - loss: 1.6750 - accuracy: 0.4077 - val_loss: 1.4872 - val_accuracy: 0.4622\n",
      "Epoch 2/25\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.4470 - accuracy: 0.4667 - val_loss: 1.4061 - val_accuracy: 0.5044\n",
      "Epoch 3/25\n",
      "480/480 [==============================] - 6s 12ms/step - loss: 1.3335 - accuracy: 0.5048 - val_loss: 1.4181 - val_accuracy: 0.4889\n",
      "Epoch 4/25\n",
      "480/480 [==============================] - 6s 12ms/step - loss: 1.3029 - accuracy: 0.5148 - val_loss: 1.4426 - val_accuracy: 0.4778\n",
      "Epoch 5/25\n",
      "480/480 [==============================] - 6s 12ms/step - loss: 1.2576 - accuracy: 0.5236 - val_loss: 1.3115 - val_accuracy: 0.5289\n",
      "Epoch 6/25\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.2307 - accuracy: 0.5369 - val_loss: 1.3047 - val_accuracy: 0.5667\n",
      "Epoch 7/25\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.1849 - accuracy: 0.5518 - val_loss: 1.4914 - val_accuracy: 0.4778\n",
      "Epoch 8/25\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.1827 - accuracy: 0.5552 - val_loss: 1.4418 - val_accuracy: 0.4600\n",
      "Epoch 9/25\n",
      "480/480 [==============================] - 6s 12ms/step - loss: 1.1598 - accuracy: 0.5565 - val_loss: 1.4164 - val_accuracy: 0.5022\n",
      "Epoch 10/25\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.1388 - accuracy: 0.5619 - val_loss: 1.4737 - val_accuracy: 0.4578\n",
      "Epoch 11/25\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.1209 - accuracy: 0.5750 - val_loss: 1.4327 - val_accuracy: 0.4933\n",
      "15/15 [==============================] - 0s 6ms/step - loss: 1.3047 - accuracy: 0.5667\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "COMET INFO: [1.304710865020752, 0.5666666626930237]\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Comet.ml Experiment Summary\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO:   Data:\n",
      "COMET INFO:     display_summary_level : 1\n",
      "COMET INFO:     url                   : https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/65d35781be3c45dbab9ab59059959fd1\n",
      "COMET INFO:   Metrics [count] (min, max):\n",
      "COMET INFO:     accuracy [20]                : (0.40765824913978577, 0.5750195384025574)\n",
      "COMET INFO:     batch_accuracy [960]         : (0.0, 0.75)\n",
      "COMET INFO:     batch_loss [960]             : (0.6592506170272827, 4.429064750671387)\n",
      "COMET INFO:     epoch_duration [20]          : (5.7810000000026776, 8.10899999999674)\n",
      "COMET INFO:     loss [20]                    : (1.1209239959716797, 1.69320809841156)\n",
      "COMET INFO:     val_accuracy [20]            : (0.4422222077846527, 0.5666666626930237)\n",
      "COMET INFO:     val_loss [20]                : (1.304710865020752, 1.5747039318084717)\n",
      "COMET INFO:     validate_batch_accuracy [60] : (0.25, 0.5625)\n",
      "COMET INFO:     validate_batch_loss [60]     : (1.1437797546386719, 1.7475192546844482)\n",
      "COMET INFO:   Others:\n",
      "COMET INFO:     optimizer_count        : 4\n",
      "COMET INFO:     optimizer_id           : 019075b278c649c7b9842acd5f3dcf6d\n",
      "COMET INFO:     optimizer_metric       : loss\n",
      "COMET INFO:     optimizer_metric_value : 1.1209239959716797\n",
      "COMET INFO:     optimizer_objective    : minimum\n",
      "COMET INFO:     optimizer_parameters   : {\"audio_layer_1\": 1000, \"audio_layer_2\": 300, \"batch_size\": 16, \"epochs\": 25, \"final_layer\": 20, \"meta_layer_1\": 1000, \"meta_layer_2\": 256}\n",
      "COMET INFO:     optimizer_pid          : 62d2641341a3ebdef24af48f1072d317c5133340\n",
      "COMET INFO:     optimizer_process      : 13176\n",
      "COMET INFO:     optimizer_trial        : 1\n",
      "COMET INFO:     optimizer_version      : 2.0.1\n",
      "COMET INFO:     trainable_params       : 731088\n",
      "COMET INFO:   Parameters:\n",
      "COMET INFO:     Adam_amsgrad       : False\n",
      "COMET INFO:     Adam_beta_1        : 0.9\n",
      "COMET INFO:     Adam_beta_2        : 0.999\n",
      "COMET INFO:     Adam_decay         : 0.0\n",
      "COMET INFO:     Adam_epsilon       : 1e-07\n",
      "COMET INFO:     Adam_learning_rate : 0.001\n",
      "COMET INFO:     Adam_name          : Adam\n",
      "COMET INFO:     Optimizer          : Adam\n",
      "COMET INFO:     audio_layer_1      : 1000\n",
      "COMET INFO:     audio_layer_2      : 300\n",
      "COMET INFO:     batch_size         : 16\n",
      "COMET INFO:     epochs             : 25\n",
      "COMET INFO:     final_layer        : 20\n",
      "COMET INFO:     meta_layer_1       : 1000\n",
      "COMET INFO:     meta_layer_2       : 256\n",
      "COMET INFO:     steps              : 480\n",
      "COMET INFO:   Uploads:\n",
      "COMET INFO:     environment details      : 1\n",
      "COMET INFO:     filename                 : 1\n",
      "COMET INFO:     git metadata             : 1\n",
      "COMET INFO:     git-patch (uncompressed) : 1 (826 bytes)\n",
      "COMET INFO:     installed packages       : 1\n",
      "COMET INFO:     model graph              : 1\n",
      "COMET INFO:     notebook                 : 1\n",
      "COMET INFO:     source_code              : 1\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Uploading 1 metrics, params and output messages\n",
      "COMET INFO: Waiting for completion of the file uploads (may take several seconds)\n",
      "COMET INFO: The Python SDK has 10800 seconds to finish before aborting...\n",
      "COMET INFO: Still uploading 2 file(s), remaining 16.23 KB/16.82 KB\n",
      "COMET WARNING: As you are running in a Jupyter environment, you will need to call `experiment.end()` when finished to ensure all metrics and code are logged before exiting.\n",
      "COMET INFO: Experiment is live on comet.ml https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/2a66ce2af0354206a10d205a26b74a18\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/25\n",
      "480/480 [==============================] - 4s 6ms/step - loss: 1.8298 - accuracy: 0.3799 - val_loss: 1.6040 - val_accuracy: 0.4089\n",
      "Epoch 2/25\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.5378 - accuracy: 0.4398 - val_loss: 1.5954 - val_accuracy: 0.4400\n",
      "Epoch 3/25\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.4225 - accuracy: 0.4668 - val_loss: 1.4675 - val_accuracy: 0.4756\n",
      "Epoch 4/25\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.3669 - accuracy: 0.4954 - val_loss: 1.5102 - val_accuracy: 0.4267\n",
      "Epoch 5/25\n",
      "480/480 [==============================] - 3s 5ms/step - loss: 1.3128 - accuracy: 0.5156 - val_loss: 1.4330 - val_accuracy: 0.4689\n",
      "Epoch 6/25\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.2885 - accuracy: 0.5133 - val_loss: 1.4329 - val_accuracy: 0.4867\n",
      "Epoch 7/25\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.2706 - accuracy: 0.5220 - val_loss: 1.3771 - val_accuracy: 0.4956\n",
      "Epoch 8/25\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.2483 - accuracy: 0.5289 - val_loss: 1.3996 - val_accuracy: 0.5156\n",
      "Epoch 9/25\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.2109 - accuracy: 0.5388 - val_loss: 1.3837 - val_accuracy: 0.5222\n",
      "Epoch 10/25\n",
      "480/480 [==============================] - 4s 8ms/step - loss: 1.1979 - accuracy: 0.5409 - val_loss: 1.4601 - val_accuracy: 0.4867\n",
      "Epoch 11/25\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.2021 - accuracy: 0.5365 - val_loss: 1.5045 - val_accuracy: 0.4756\n",
      "Epoch 12/25\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.1790 - accuracy: 0.5488 - val_loss: 1.4888 - val_accuracy: 0.4778\n",
      "Epoch 1/25\n",
      "480/480 [==============================] - 8s 15ms/step - loss: 1.8218 - accuracy: 0.3802 - val_loss: 1.6120 - val_accuracy: 0.3844\n",
      "Epoch 2/25\n",
      "480/480 [==============================] - 7s 14ms/step - loss: 1.5191 - accuracy: 0.4466 - val_loss: 1.4781 - val_accuracy: 0.4489\n",
      "Epoch 3/25\n",
      "480/480 [==============================] - 7s 14ms/step - loss: 1.4237 - accuracy: 0.4780 - val_loss: 1.3418 - val_accuracy: 0.5089\n",
      "Epoch 4/25\n",
      "480/480 [==============================] - 7s 14ms/step - loss: 1.3632 - accuracy: 0.4911 - val_loss: 1.4250 - val_accuracy: 0.4844\n",
      "Epoch 5/25\n",
      "480/480 [==============================] - 7s 14ms/step - loss: 1.3303 - accuracy: 0.5017 - val_loss: 1.4306 - val_accuracy: 0.4778\n",
      "Epoch 6/25\n",
      "480/480 [==============================] - 7s 14ms/step - loss: 1.2814 - accuracy: 0.5185 - val_loss: 1.4023 - val_accuracy: 0.5000\n",
      "Epoch 7/25\n",
      "480/480 [==============================] - 7s 14ms/step - loss: 1.2729 - accuracy: 0.5264 - val_loss: 1.3885 - val_accuracy: 0.4800\n",
      "Epoch 8/25\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.2516 - accuracy: 0.5206 - val_loss: 1.3715 - val_accuracy: 0.5422\n",
      "15/15 [==============================] - 0s 5ms/step - loss: 1.3418 - accuracy: 0.5089\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "COMET INFO: [1.3417807817459106, 0.5088889002799988]\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Comet.ml Experiment Summary\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO:   Data:\n",
      "COMET INFO:     display_summary_level : 1\n",
      "COMET INFO:     url                   : https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/2a66ce2af0354206a10d205a26b74a18\n",
      "COMET INFO:   Metrics [count] (min, max):\n",
      "COMET INFO:     accuracy [20]                : (0.37991663813591003, 0.5488408207893372)\n",
      "COMET INFO:     batch_accuracy [960]         : (0.0, 0.6875)\n",
      "COMET INFO:     batch_loss [960]             : (0.7309568524360657, 2.561605453491211)\n",
      "COMET INFO:     epoch_duration [20]          : (2.3910000000032596, 8.39100000000326)\n",
      "COMET INFO:     loss [20]                    : (1.178981065750122, 1.8297556638717651)\n",
      "COMET INFO:     val_accuracy [20]            : (0.3844444453716278, 0.5422222018241882)\n",
      "COMET INFO:     val_loss [20]                : (1.3417832851409912, 1.612029790878296)\n",
      "COMET INFO:     validate_batch_accuracy [60] : (0.125, 0.625)\n",
      "COMET INFO:     validate_batch_loss [60]     : (1.1589874029159546, 1.9547268152236938)\n",
      "COMET INFO:   Others:\n",
      "COMET INFO:     optimizer_count        : 5\n",
      "COMET INFO:     optimizer_id           : 019075b278c649c7b9842acd5f3dcf6d\n",
      "COMET INFO:     optimizer_metric       : loss\n",
      "COMET INFO:     optimizer_metric_value : 1.178981065750122\n",
      "COMET INFO:     optimizer_objective    : minimum\n",
      "COMET INFO:     optimizer_parameters   : {\"audio_layer_1\": 450, \"audio_layer_2\": 128, \"batch_size\": 16, \"epochs\": 25, \"final_layer\": 256, \"meta_layer_1\": 450, \"meta_layer_2\": 300}\n",
      "COMET INFO:     optimizer_pid          : 1733f56b7c608e0e0f083c2e6a0a19406ef04709\n",
      "COMET INFO:     optimizer_process      : 13176\n",
      "COMET INFO:     optimizer_trial        : 1\n",
      "COMET INFO:     optimizer_version      : 2.0.1\n",
      "COMET INFO:     trainable_params       : 379070\n",
      "COMET INFO:   Parameters:\n",
      "COMET INFO:     Adam_amsgrad       : False\n",
      "COMET INFO:     Adam_beta_1        : 0.9\n",
      "COMET INFO:     Adam_beta_2        : 0.999\n",
      "COMET INFO:     Adam_decay         : 0.0\n",
      "COMET INFO:     Adam_epsilon       : 1e-07\n",
      "COMET INFO:     Adam_learning_rate : 0.001\n",
      "COMET INFO:     Adam_name          : Adam\n",
      "COMET INFO:     Optimizer          : Adam\n",
      "COMET INFO:     audio_layer_1      : 450\n",
      "COMET INFO:     audio_layer_2      : 128\n",
      "COMET INFO:     batch_size         : 16\n",
      "COMET INFO:     epochs             : 25\n",
      "COMET INFO:     final_layer        : 256\n",
      "COMET INFO:     meta_layer_1       : 450\n",
      "COMET INFO:     meta_layer_2       : 300\n",
      "COMET INFO:     steps              : 480\n",
      "COMET INFO:   Uploads:\n",
      "COMET INFO:     environment details      : 1\n",
      "COMET INFO:     filename                 : 1\n",
      "COMET INFO:     git metadata             : 1\n",
      "COMET INFO:     git-patch (uncompressed) : 1 (826 bytes)\n",
      "COMET INFO:     installed packages       : 1\n",
      "COMET INFO:     model graph              : 1\n",
      "COMET INFO:     notebook                 : 1\n",
      "COMET INFO:     source_code              : 1\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Uploading 1 metrics, params and output messages\n",
      "COMET INFO: Waiting for completion of the file uploads (may take several seconds)\n",
      "COMET INFO: The Python SDK has 10800 seconds to finish before aborting...\n",
      "COMET INFO: Still uploading 2 file(s), remaining 16.23 KB/16.82 KB\n",
      "COMET WARNING: As you are running in a Jupyter environment, you will need to call `experiment.end()` when finished to ensure all metrics and code are logged before exiting.\n",
      "COMET INFO: Experiment is live on comet.ml https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/4666dc2d408744d1b878296d3ffc25db\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/15\n",
      "480/480 [==============================] - 8s 13ms/step - loss: 1.6963 - accuracy: 0.4052 - val_loss: 1.3157 - val_accuracy: 0.5622\n",
      "Epoch 2/15\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.4392 - accuracy: 0.4642 - val_loss: 1.4364 - val_accuracy: 0.4756\n",
      "Epoch 3/15\n",
      "480/480 [==============================] - 6s 12ms/step - loss: 1.3588 - accuracy: 0.4782 - val_loss: 1.4608 - val_accuracy: 0.4800\n",
      "Epoch 4/15\n",
      "480/480 [==============================] - 6s 12ms/step - loss: 1.3162 - accuracy: 0.5027 - val_loss: 1.4786 - val_accuracy: 0.4844\n",
      "Epoch 5/15\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.2759 - accuracy: 0.5102 - val_loss: 1.3636 - val_accuracy: 0.5000\n",
      "Epoch 6/15\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.2420 - accuracy: 0.5297 - val_loss: 1.3446 - val_accuracy: 0.5311\n",
      "Epoch 1/15\n",
      "480/480 [==============================] - 8s 14ms/step - loss: 1.6926 - accuracy: 0.4112 - val_loss: 1.5593 - val_accuracy: 0.4089\n",
      "Epoch 2/15\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.4348 - accuracy: 0.4771 - val_loss: 1.4948 - val_accuracy: 0.4911\n",
      "Epoch 3/15\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.3648 - accuracy: 0.4904 - val_loss: 1.5572 - val_accuracy: 0.4311\n",
      "Epoch 4/15\n",
      "480/480 [==============================] - 7s 14ms/step - loss: 1.2852 - accuracy: 0.5223 - val_loss: 1.4379 - val_accuracy: 0.4711\n",
      "Epoch 5/15\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.2386 - accuracy: 0.5283 - val_loss: 1.4224 - val_accuracy: 0.4822\n",
      "Epoch 6/15\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.2292 - accuracy: 0.5380 - val_loss: 1.3610 - val_accuracy: 0.5044\n",
      "Epoch 7/15\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.1953 - accuracy: 0.5461 - val_loss: 1.4201 - val_accuracy: 0.4800\n",
      "Epoch 8/15\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.1549 - accuracy: 0.5565 - val_loss: 1.4840 - val_accuracy: 0.4956\n",
      "Epoch 9/15\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.1652 - accuracy: 0.5585 - val_loss: 1.5616 - val_accuracy: 0.4667\n",
      "Epoch 10/15\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.1382 - accuracy: 0.5675 - val_loss: 1.5572 - val_accuracy: 0.4689\n",
      "Epoch 11/15\n",
      "480/480 [==============================] - 6s 13ms/step - loss: 1.1256 - accuracy: 0.5732 - val_loss: 1.4400 - val_accuracy: 0.5156\n",
      "15/15 [==============================] - 0s 8ms/step - loss: 1.3610 - accuracy: 0.5044\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "COMET INFO: [1.3610267639160156, 0.504444420337677]\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Comet.ml Experiment Summary\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO:   Data:\n",
      "COMET INFO:     display_summary_level : 1\n",
      "COMET INFO:     url                   : https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/4666dc2d408744d1b878296d3ffc25db\n",
      "COMET INFO:   Metrics [count] (min, max):\n",
      "COMET INFO:     accuracy [17]                : (0.4051836431026459, 0.5731961727142334)\n",
      "COMET INFO:     batch_accuracy [816]         : (0.20454545319080353, 0.6875)\n",
      "COMET INFO:     batch_loss [816]             : (0.6490731835365295, 2.577146530151367)\n",
      "COMET INFO:     epoch_duration [17]          : (5.875, 7.906999999991967)\n",
      "COMET INFO:     loss [17]                    : (1.125561237335205, 1.696272611618042)\n",
      "COMET INFO:     val_accuracy [17]            : (0.40888887643814087, 0.5622222423553467)\n",
      "COMET INFO:     val_loss [17]                : (1.3157315254211426, 1.5615531206130981)\n",
      "COMET INFO:     validate_batch_accuracy [51] : (0.25, 0.5565476417541504)\n",
      "COMET INFO:     validate_batch_loss [51]     : (1.213361382484436, 1.6691054105758667)\n",
      "COMET INFO:   Others:\n",
      "COMET INFO:     optimizer_count        : 6\n",
      "COMET INFO:     optimizer_id           : 019075b278c649c7b9842acd5f3dcf6d\n",
      "COMET INFO:     optimizer_metric       : loss\n",
      "COMET INFO:     optimizer_metric_value : 1.125561237335205\n",
      "COMET INFO:     optimizer_objective    : minimum\n",
      "COMET INFO:     optimizer_parameters   : {\"audio_layer_1\": 450, \"audio_layer_2\": 400, \"batch_size\": 16, \"epochs\": 15, \"final_layer\": 32, \"meta_layer_1\": 300, \"meta_layer_2\": 300}\n",
      "COMET INFO:     optimizer_pid          : 08b5305e0c77983703bf12d6144bcd72af3a5f87\n",
      "COMET INFO:     optimizer_process      : 13176\n",
      "COMET INFO:     optimizer_trial        : 1\n",
      "COMET INFO:     optimizer_version      : 2.0.1\n",
      "COMET INFO:     trainable_params       : 367446\n",
      "COMET INFO:   Parameters:\n",
      "COMET INFO:     Adam_amsgrad       : False\n",
      "COMET INFO:     Adam_beta_1        : 0.9\n",
      "COMET INFO:     Adam_beta_2        : 0.999\n",
      "COMET INFO:     Adam_decay         : 0.0\n",
      "COMET INFO:     Adam_epsilon       : 1e-07\n",
      "COMET INFO:     Adam_learning_rate : 0.001\n",
      "COMET INFO:     Adam_name          : Adam\n",
      "COMET INFO:     Optimizer          : Adam\n",
      "COMET INFO:     audio_layer_1      : 450\n",
      "COMET INFO:     audio_layer_2      : 400\n",
      "COMET INFO:     batch_size         : 16\n",
      "COMET INFO:     epochs             : 15\n",
      "COMET INFO:     final_layer        : 32\n",
      "COMET INFO:     meta_layer_1       : 300\n",
      "COMET INFO:     meta_layer_2       : 300\n",
      "COMET INFO:     steps              : 480\n",
      "COMET INFO:   Uploads:\n",
      "COMET INFO:     environment details      : 1\n",
      "COMET INFO:     filename                 : 1\n",
      "COMET INFO:     git metadata             : 1\n",
      "COMET INFO:     git-patch (uncompressed) : 1 (826 bytes)\n",
      "COMET INFO:     installed packages       : 1\n",
      "COMET INFO:     model graph              : 1\n",
      "COMET INFO:     notebook                 : 1\n",
      "COMET INFO:     source_code              : 1\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Uploading 1 metrics, params and output messages\n",
      "COMET INFO: Waiting for completion of the file uploads (may take several seconds)\n",
      "COMET INFO: The Python SDK has 10800 seconds to finish before aborting...\n",
      "COMET INFO: Still uploading 2 file(s), remaining 16.23 KB/16.82 KB\n",
      "COMET WARNING: As you are running in a Jupyter environment, you will need to call `experiment.end()` when finished to ensure all metrics and code are logged before exiting.\n",
      "COMET INFO: Experiment is live on comet.ml https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/bfd3ca0c69a74f3c85d8dcf3b951b376\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/25\n",
      "240/240 [==============================] - 5s 15ms/step - loss: 1.7979 - accuracy: 0.4032 - val_loss: 1.4267 - val_accuracy: 0.4822\n",
      "Epoch 2/25\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.4607 - accuracy: 0.4721 - val_loss: 1.4065 - val_accuracy: 0.5444\n",
      "Epoch 3/25\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.3554 - accuracy: 0.4957 - val_loss: 1.4416 - val_accuracy: 0.5022\n",
      "Epoch 4/25\n",
      "240/240 [==============================] - 4s 15ms/step - loss: 1.2883 - accuracy: 0.5189 - val_loss: 1.4165 - val_accuracy: 0.4889\n",
      "Epoch 5/25\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.2308 - accuracy: 0.5388 - val_loss: 1.5219 - val_accuracy: 0.4644\n",
      "Epoch 6/25\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.1870 - accuracy: 0.5563 - val_loss: 1.4416 - val_accuracy: 0.5000\n",
      "Epoch 7/25\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.1517 - accuracy: 0.5642 - val_loss: 1.5717 - val_accuracy: 0.4356\n",
      "Epoch 1/25\n",
      "240/240 [==============================] - 5s 15ms/step - loss: 1.7081 - accuracy: 0.4144 - val_loss: 1.5499 - val_accuracy: 0.4444\n",
      "Epoch 2/25\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.4414 - accuracy: 0.4729 - val_loss: 1.4866 - val_accuracy: 0.4356\n",
      "Epoch 3/25\n",
      "240/240 [==============================] - 3s 14ms/step - loss: 1.3402 - accuracy: 0.5126 - val_loss: 1.3922 - val_accuracy: 0.4978\n",
      "Epoch 4/25\n",
      "240/240 [==============================] - 3s 14ms/step - loss: 1.2607 - accuracy: 0.5348 - val_loss: 1.3803 - val_accuracy: 0.5111\n",
      "Epoch 5/25\n",
      "240/240 [==============================] - 3s 12ms/step - loss: 1.2069 - accuracy: 0.5405 - val_loss: 1.4018 - val_accuracy: 0.5000\n",
      "Epoch 6/25\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.1652 - accuracy: 0.5514 - val_loss: 1.3778 - val_accuracy: 0.4978\n",
      "Epoch 7/25\n",
      "240/240 [==============================] - 3s 13ms/step - loss: 1.1418 - accuracy: 0.5623 - val_loss: 1.3664 - val_accuracy: 0.5400\n",
      "Epoch 8/25\n",
      "240/240 [==============================] - 2s 10ms/step - loss: 1.1075 - accuracy: 0.5757 - val_loss: 1.4716 - val_accuracy: 0.4867\n",
      "Epoch 9/25\n",
      "240/240 [==============================] - 2s 7ms/step - loss: 1.0782 - accuracy: 0.5882 - val_loss: 1.4766 - val_accuracy: 0.4867\n",
      "Epoch 10/25\n",
      "240/240 [==============================] - 1s 6ms/step - loss: 1.0735 - accuracy: 0.5917 - val_loss: 1.3813 - val_accuracy: 0.4978\n",
      "Epoch 11/25\n",
      "240/240 [==============================] - 2s 7ms/step - loss: 1.0307 - accuracy: 0.6017 - val_loss: 1.3974 - val_accuracy: 0.5133\n",
      "Epoch 12/25\n",
      "240/240 [==============================] - 2s 8ms/step - loss: 1.0182 - accuracy: 0.6069 - val_loss: 1.3946 - val_accuracy: 0.5333\n",
      "15/15 [==============================] - 0s 3ms/step - loss: 1.3664 - accuracy: 0.5400\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "COMET INFO: [1.366371989250183, 0.5400000214576721]\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Comet.ml Experiment Summary\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO:   Data:\n",
      "COMET INFO:     display_summary_level : 1\n",
      "COMET INFO:     url                   : https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/bfd3ca0c69a74f3c85d8dcf3b951b376\n",
      "COMET INFO:   Metrics [count] (min, max):\n",
      "COMET INFO:     accuracy [19]                : (0.4032300114631653, 0.6069288849830627)\n",
      "COMET INFO:     batch_accuracy [456]         : (0.03125, 0.78125)\n",
      "COMET INFO:     batch_loss [456]             : (0.7370853424072266, 2.974832057952881)\n",
      "COMET INFO:     epoch_duration [19]          : (1.4679999999934807, 5.4689999999973224)\n",
      "COMET INFO:     loss [19]                    : (1.018239140510559, 1.7978596687316895)\n",
      "COMET INFO:     val_accuracy [19]            : (0.4355555474758148, 0.5444444417953491)\n",
      "COMET INFO:     val_loss [19]                : (1.366371989250183, 1.571660041809082)\n",
      "COMET INFO:     validate_batch_accuracy [38] : (0.21875, 0.5625)\n",
      "COMET INFO:     validate_batch_loss [38]     : (1.3131895065307617, 1.6925764083862305)\n",
      "COMET INFO:   Others:\n",
      "COMET INFO:     optimizer_count        : 7\n",
      "COMET INFO:     optimizer_id           : 019075b278c649c7b9842acd5f3dcf6d\n",
      "COMET INFO:     optimizer_metric       : loss\n",
      "COMET INFO:     optimizer_metric_value : 1.018239140510559\n",
      "COMET INFO:     optimizer_objective    : minimum\n",
      "COMET INFO:     optimizer_parameters   : {\"audio_layer_1\": 400, \"audio_layer_2\": 256, \"batch_size\": 32, \"epochs\": 25, \"final_layer\": 128, \"meta_layer_1\": 550, \"meta_layer_2\": 400}\n",
      "COMET INFO:     optimizer_pid          : 825fcac910c770e38c088e597bd4841ca15adbac\n",
      "COMET INFO:     optimizer_process      : 13176\n",
      "COMET INFO:     optimizer_trial        : 1\n",
      "COMET INFO:     optimizer_version      : 2.0.1\n",
      "COMET INFO:     trainable_params       : 476408\n",
      "COMET INFO:   Parameters:\n",
      "COMET INFO:     Adam_amsgrad       : False\n",
      "COMET INFO:     Adam_beta_1        : 0.9\n",
      "COMET INFO:     Adam_beta_2        : 0.999\n",
      "COMET INFO:     Adam_decay         : 0.0\n",
      "COMET INFO:     Adam_epsilon       : 1e-07\n",
      "COMET INFO:     Adam_learning_rate : 0.001\n",
      "COMET INFO:     Adam_name          : Adam\n",
      "COMET INFO:     Optimizer          : Adam\n",
      "COMET INFO:     audio_layer_1      : 400\n",
      "COMET INFO:     audio_layer_2      : 256\n",
      "COMET INFO:     batch_size         : 32\n",
      "COMET INFO:     epochs             : 25\n",
      "COMET INFO:     final_layer        : 128\n",
      "COMET INFO:     meta_layer_1       : 550\n",
      "COMET INFO:     meta_layer_2       : 400\n",
      "COMET INFO:     steps              : 240\n",
      "COMET INFO:   Uploads:\n",
      "COMET INFO:     environment details      : 1\n",
      "COMET INFO:     filename                 : 1\n",
      "COMET INFO:     git metadata             : 1\n",
      "COMET INFO:     git-patch (uncompressed) : 1 (826 bytes)\n",
      "COMET INFO:     installed packages       : 1\n",
      "COMET INFO:     model graph              : 1\n",
      "COMET INFO:     notebook                 : 1\n",
      "COMET INFO:     source_code              : 1\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Uploading 1 metrics, params and output messages\n",
      "COMET INFO: Waiting for completion of the file uploads (may take several seconds)\n",
      "COMET INFO: The Python SDK has 10800 seconds to finish before aborting...\n",
      "COMET INFO: Still uploading 2 file(s), remaining 16.23 KB/16.82 KB\n",
      "COMET WARNING: As you are running in a Jupyter environment, you will need to call `experiment.end()` when finished to ensure all metrics and code are logged before exiting.\n",
      "COMET INFO: Experiment is live on comet.ml https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/a38ec5e9fb244a4cb89ddcaf04c3555e\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "240/240 [==============================] - 3s 8ms/step - loss: 1.6899 - accuracy: 0.4157 - val_loss: 1.5986 - val_accuracy: 0.4533\n",
      "Epoch 2/20\n",
      "240/240 [==============================] - 2s 7ms/step - loss: 1.3978 - accuracy: 0.4858 - val_loss: 1.6356 - val_accuracy: 0.4200\n",
      "Epoch 3/20\n",
      "240/240 [==============================] - 2s 7ms/step - loss: 1.3066 - accuracy: 0.5160 - val_loss: 1.3189 - val_accuracy: 0.5467\n",
      "Epoch 4/20\n",
      "240/240 [==============================] - 2s 8ms/step - loss: 1.2435 - accuracy: 0.5357 - val_loss: 1.4266 - val_accuracy: 0.5089\n",
      "Epoch 5/20\n",
      "240/240 [==============================] - 2s 7ms/step - loss: 1.1791 - accuracy: 0.5613 - val_loss: 1.4392 - val_accuracy: 0.4844\n",
      "Epoch 6/20\n",
      "240/240 [==============================] - 2s 7ms/step - loss: 1.1576 - accuracy: 0.5673 - val_loss: 1.6430 - val_accuracy: 0.4644\n",
      "Epoch 7/20\n",
      "240/240 [==============================] - 1s 6ms/step - loss: 1.0909 - accuracy: 0.5896 - val_loss: 1.3847 - val_accuracy: 0.5644\n",
      "Epoch 8/20\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 1.0798 - accuracy: 0.5845 - val_loss: 1.3090 - val_accuracy: 0.5511\n",
      "Epoch 9/20\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 1.0470 - accuracy: 0.5978 - val_loss: 1.5226 - val_accuracy: 0.5111\n",
      "Epoch 10/20\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 1.0380 - accuracy: 0.6063 - val_loss: 1.4831 - val_accuracy: 0.4756\n",
      "Epoch 11/20\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 0.9830 - accuracy: 0.6142 - val_loss: 1.3830 - val_accuracy: 0.5356\n",
      "Epoch 12/20\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 0.9708 - accuracy: 0.6224 - val_loss: 1.4968 - val_accuracy: 0.4889\n",
      "Epoch 13/20\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 0.9601 - accuracy: 0.6229 - val_loss: 1.4803 - val_accuracy: 0.5244\n",
      "Epoch 1/20\n",
      "240/240 [==============================] - 2s 6ms/step - loss: 1.6629 - accuracy: 0.4228 - val_loss: 1.4176 - val_accuracy: 0.5000\n",
      "Epoch 2/20\n",
      "240/240 [==============================] - 2s 8ms/step - loss: 1.4082 - accuracy: 0.4923 - val_loss: 1.3414 - val_accuracy: 0.5400\n",
      "Epoch 3/20\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 1.3288 - accuracy: 0.5182 - val_loss: 1.4449 - val_accuracy: 0.4978\n",
      "Epoch 4/20\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 1.2164 - accuracy: 0.5332 - val_loss: 1.5731 - val_accuracy: 0.4556\n",
      "Epoch 5/20\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 1.1746 - accuracy: 0.5621 - val_loss: 1.4686 - val_accuracy: 0.4667\n",
      "Epoch 6/20\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 1.1382 - accuracy: 0.5666 - val_loss: 1.4534 - val_accuracy: 0.4756\n",
      "Epoch 7/20\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 1.0964 - accuracy: 0.5817 - val_loss: 1.4194 - val_accuracy: 0.4911\n",
      "15/15 [==============================] - 0s 3ms/step - loss: 1.3414 - accuracy: 0.5400\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "COMET INFO: [1.3414344787597656, 0.5400000214576721]\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Comet.ml Experiment Summary\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO:   Data:\n",
      "COMET INFO:     display_summary_level : 1\n",
      "COMET INFO:     url                   : https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/a38ec5e9fb244a4cb89ddcaf04c3555e\n",
      "COMET INFO:   Metrics [count] (min, max):\n",
      "COMET INFO:     accuracy [20]                : (0.415733277797699, 0.6229487061500549)\n",
      "COMET INFO:     batch_accuracy [480]         : (0.0625, 0.78125)\n",
      "COMET INFO:     batch_loss [480]             : (0.7190606594085693, 2.235048294067383)\n",
      "COMET INFO:     epoch_duration [20]          : (1.2190000000118744, 2.812000000005355)\n",
      "COMET INFO:     loss [20]                    : (0.9601150751113892, 1.6898738145828247)\n",
      "COMET INFO:     val_accuracy [20]            : (0.41999998688697815, 0.5644444227218628)\n",
      "COMET INFO:     val_loss [20]                : (1.3089936971664429, 1.6430115699768066)\n",
      "COMET INFO:     validate_batch_accuracy [40] : (0.1875, 0.5625)\n",
      "COMET INFO:     validate_batch_loss [40]     : (1.2269785404205322, 1.6647276878356934)\n",
      "COMET INFO:   Others:\n",
      "COMET INFO:     optimizer_count        : 8\n",
      "COMET INFO:     optimizer_id           : 019075b278c649c7b9842acd5f3dcf6d\n",
      "COMET INFO:     optimizer_metric       : loss\n",
      "COMET INFO:     optimizer_metric_value : 0.9601150751113892\n",
      "COMET INFO:     optimizer_objective    : minimum\n",
      "COMET INFO:     optimizer_parameters   : {\"audio_layer_1\": 1000, \"audio_layer_2\": 256, \"batch_size\": 32, \"epochs\": 20, \"final_layer\": 256, \"meta_layer_1\": 700, \"meta_layer_2\": 128}\n",
      "COMET INFO:     optimizer_pid          : dd39b4dc0fb6bd63b51d79f1fcd9bf65b3e2dfc0\n",
      "COMET INFO:     optimizer_process      : 13176\n",
      "COMET INFO:     optimizer_trial        : 1\n",
      "COMET INFO:     optimizer_version      : 2.0.1\n",
      "COMET INFO:     trainable_params       : 606736\n",
      "COMET INFO:   Parameters:\n",
      "COMET INFO:     Adam_amsgrad       : False\n",
      "COMET INFO:     Adam_beta_1        : 0.9\n",
      "COMET INFO:     Adam_beta_2        : 0.999\n",
      "COMET INFO:     Adam_decay         : 0.0\n",
      "COMET INFO:     Adam_epsilon       : 1e-07\n",
      "COMET INFO:     Adam_learning_rate : 0.001\n",
      "COMET INFO:     Adam_name          : Adam\n",
      "COMET INFO:     Optimizer          : Adam\n",
      "COMET INFO:     audio_layer_1      : 1000\n",
      "COMET INFO:     audio_layer_2      : 256\n",
      "COMET INFO:     batch_size         : 32\n",
      "COMET INFO:     epochs             : 20\n",
      "COMET INFO:     final_layer        : 256\n",
      "COMET INFO:     meta_layer_1       : 700\n",
      "COMET INFO:     meta_layer_2       : 128\n",
      "COMET INFO:     steps              : 240\n",
      "COMET INFO:   Uploads:\n",
      "COMET INFO:     environment details      : 1\n",
      "COMET INFO:     filename                 : 1\n",
      "COMET INFO:     git metadata             : 1\n",
      "COMET INFO:     git-patch (uncompressed) : 1 (826 bytes)\n",
      "COMET INFO:     installed packages       : 1\n",
      "COMET INFO:     model graph              : 1\n",
      "COMET INFO:     notebook                 : 1\n",
      "COMET INFO:     source_code              : 1\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Uploading 1 metrics, params and output messages\n",
      "COMET INFO: Waiting for completion of the file uploads (may take several seconds)\n",
      "COMET INFO: The Python SDK has 10800 seconds to finish before aborting...\n",
      "COMET INFO: Still uploading 2 file(s), remaining 16.23 KB/16.82 KB\n",
      "COMET WARNING: As you are running in a Jupyter environment, you will need to call `experiment.end()` when finished to ensure all metrics and code are logged before exiting.\n",
      "COMET INFO: Experiment is live on comet.ml https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/94ff600ec47f4a6faf2685367cd5b814\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "480/480 [==============================] - 4s 7ms/step - loss: 1.7140 - accuracy: 0.4082 - val_loss: 1.4569 - val_accuracy: 0.4400\n",
      "Epoch 2/20\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.4450 - accuracy: 0.4667 - val_loss: 1.4453 - val_accuracy: 0.4933\n",
      "Epoch 3/20\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.3556 - accuracy: 0.4905 - val_loss: 1.5315 - val_accuracy: 0.4556\n",
      "Epoch 4/20\n",
      "480/480 [==============================] - 3s 6ms/step - loss: 1.2984 - accuracy: 0.5121 - val_loss: 1.4435 - val_accuracy: 0.4644\n",
      "Epoch 5/20\n",
      "480/480 [==============================] - 3s 6ms/step - loss: 1.2606 - accuracy: 0.5267 - val_loss: 1.5037 - val_accuracy: 0.4733\n",
      "Epoch 6/20\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.2298 - accuracy: 0.5371 - val_loss: 1.5495 - val_accuracy: 0.4622\n",
      "Epoch 7/20\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.2011 - accuracy: 0.5447 - val_loss: 1.4002 - val_accuracy: 0.5044\n",
      "Epoch 8/20\n",
      "480/480 [==============================] - 3s 5ms/step - loss: 1.1810 - accuracy: 0.5550 - val_loss: 1.4739 - val_accuracy: 0.4889\n",
      "Epoch 9/20\n",
      "480/480 [==============================] - 3s 7ms/step - loss: 1.1610 - accuracy: 0.5574 - val_loss: 1.5769 - val_accuracy: 0.4667\n",
      "Epoch 10/20\n",
      "480/480 [==============================] - 3s 5ms/step - loss: 1.1252 - accuracy: 0.5775 - val_loss: 1.3477 - val_accuracy: 0.5422\n",
      "Epoch 11/20\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.1293 - accuracy: 0.5714 - val_loss: 1.4709 - val_accuracy: 0.4933\n",
      "Epoch 12/20\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.1039 - accuracy: 0.5748 - val_loss: 1.6196 - val_accuracy: 0.4622\n",
      "Epoch 13/20\n",
      "480/480 [==============================] - 3s 6ms/step - loss: 1.0638 - accuracy: 0.5933 - val_loss: 1.3156 - val_accuracy: 0.5600\n",
      "Epoch 14/20\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.0565 - accuracy: 0.5901 - val_loss: 1.3584 - val_accuracy: 0.5467\n",
      "Epoch 15/20\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.0312 - accuracy: 0.6004 - val_loss: 1.5372 - val_accuracy: 0.4667\n",
      "Epoch 16/20\n",
      "480/480 [==============================] - 4s 7ms/step - loss: 1.0393 - accuracy: 0.5927 - val_loss: 1.3947 - val_accuracy: 0.5111\n",
      "Epoch 17/20\n",
      "480/480 [==============================] - 6s 12ms/step - loss: 1.0199 - accuracy: 0.6063 - val_loss: 1.4130 - val_accuracy: 0.5200\n",
      "Epoch 18/20\n",
      "480/480 [==============================] - 6s 12ms/step - loss: 0.9986 - accuracy: 0.6104 - val_loss: 1.4596 - val_accuracy: 0.5178\n",
      "Epoch 1/20\n",
      "480/480 [==============================] - 7s 13ms/step - loss: 1.7336 - accuracy: 0.3941 - val_loss: 1.6109 - val_accuracy: 0.3844\n",
      "Epoch 2/20\n",
      "480/480 [==============================] - 6s 12ms/step - loss: 1.4624 - accuracy: 0.4591 - val_loss: 1.4844 - val_accuracy: 0.4778\n",
      "Epoch 3/20\n",
      "480/480 [==============================] - 6s 12ms/step - loss: 1.3516 - accuracy: 0.4967 - val_loss: 1.4034 - val_accuracy: 0.4978\n",
      "Epoch 4/20\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.3180 - accuracy: 0.5063 - val_loss: 1.5294 - val_accuracy: 0.4644\n",
      "Epoch 5/20\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.2836 - accuracy: 0.5160 - val_loss: 1.3935 - val_accuracy: 0.5222\n",
      "Epoch 6/20\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.2194 - accuracy: 0.5341 - val_loss: 1.4530 - val_accuracy: 0.5067\n",
      "Epoch 7/20\n",
      "480/480 [==============================] - 3s 6ms/step - loss: 1.2024 - accuracy: 0.5346 - val_loss: 1.5182 - val_accuracy: 0.4644\n",
      "Epoch 8/20\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.1735 - accuracy: 0.5488 - val_loss: 1.5831 - val_accuracy: 0.4600\n",
      "Epoch 9/20\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.1684 - accuracy: 0.5560 - val_loss: 1.4372 - val_accuracy: 0.5089\n",
      "Epoch 10/20\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.1459 - accuracy: 0.5613 - val_loss: 1.3890 - val_accuracy: 0.5044\n",
      "Epoch 11/20\n",
      "480/480 [==============================] - 3s 6ms/step - loss: 1.1182 - accuracy: 0.5669 - val_loss: 1.4824 - val_accuracy: 0.4933\n",
      "Epoch 12/20\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.1012 - accuracy: 0.5778 - val_loss: 1.4145 - val_accuracy: 0.4956\n",
      "Epoch 13/20\n",
      "480/480 [==============================] - 3s 5ms/step - loss: 1.0987 - accuracy: 0.5716 - val_loss: 1.4219 - val_accuracy: 0.5222\n",
      "Epoch 14/20\n",
      "480/480 [==============================] - 3s 5ms/step - loss: 1.0638 - accuracy: 0.5923 - val_loss: 1.4884 - val_accuracy: 0.4667\n",
      "Epoch 15/20\n",
      "480/480 [==============================] - 5s 10ms/step - loss: 1.0538 - accuracy: 0.5888 - val_loss: 1.5441 - val_accuracy: 0.4556\n",
      "15/15 [==============================] - 0s 6ms/step - loss: 1.3890 - accuracy: 0.5044\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "COMET INFO: [1.3889837265014648, 0.504444420337677]\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Comet.ml Experiment Summary\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO:   Data:\n",
      "COMET INFO:     display_summary_level : 1\n",
      "COMET INFO:     url                   : https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/94ff600ec47f4a6faf2685367cd5b814\n",
      "COMET INFO:   Metrics [count] (min, max):\n",
      "COMET INFO:     accuracy [33]                : (0.39411306381225586, 0.6104454398155212)\n",
      "COMET INFO:     batch_accuracy [1584]        : (0.125, 0.8125)\n",
      "COMET INFO:     batch_loss [1584]            : (0.7375304698944092, 2.6997132301330566)\n",
      "COMET INFO:     epoch_duration [33]          : (2.3910000000032596, 7.35899999999674)\n",
      "COMET INFO:     loss [33]                    : (0.9985768795013428, 1.733643889427185)\n",
      "COMET INFO:     val_accuracy [33]            : (0.3844444453716278, 0.5600000023841858)\n",
      "COMET INFO:     val_loss [33]                : (1.315576195716858, 1.6196433305740356)\n",
      "COMET INFO:     validate_batch_accuracy [99] : (0.25, 0.6875)\n",
      "COMET INFO:     validate_batch_loss [99]     : (0.9737690687179565, 1.8780163526535034)\n",
      "COMET INFO:   Others:\n",
      "COMET INFO:     optimizer_count        : 9\n",
      "COMET INFO:     optimizer_id           : 019075b278c649c7b9842acd5f3dcf6d\n",
      "COMET INFO:     optimizer_metric       : loss\n",
      "COMET INFO:     optimizer_metric_value : 0.9985768795013428\n",
      "COMET INFO:     optimizer_objective    : minimum\n",
      "COMET INFO:     optimizer_parameters   : {\"audio_layer_1\": 1000, \"audio_layer_2\": 450, \"batch_size\": 16, \"epochs\": 20, \"final_layer\": 20, \"meta_layer_1\": 500, \"meta_layer_2\": 450}\n",
      "COMET INFO:     optimizer_pid          : ec7c3786368f7a1f36db795f7644975cc115ef9a\n",
      "COMET INFO:     optimizer_process      : 13176\n",
      "COMET INFO:     optimizer_trial        : 1\n",
      "COMET INFO:     optimizer_version      : 2.0.1\n",
      "COMET INFO:     trainable_params       : 854688\n",
      "COMET INFO:   Parameters:\n",
      "COMET INFO:     Adam_amsgrad       : False\n",
      "COMET INFO:     Adam_beta_1        : 0.9\n",
      "COMET INFO:     Adam_beta_2        : 0.999\n",
      "COMET INFO:     Adam_decay         : 0.0\n",
      "COMET INFO:     Adam_epsilon       : 1e-07\n",
      "COMET INFO:     Adam_learning_rate : 0.001\n",
      "COMET INFO:     Adam_name          : Adam\n",
      "COMET INFO:     Optimizer          : Adam\n",
      "COMET INFO:     audio_layer_1      : 1000\n",
      "COMET INFO:     audio_layer_2      : 450\n",
      "COMET INFO:     batch_size         : 16\n",
      "COMET INFO:     epochs             : 20\n",
      "COMET INFO:     final_layer        : 20\n",
      "COMET INFO:     meta_layer_1       : 500\n",
      "COMET INFO:     meta_layer_2       : 450\n",
      "COMET INFO:     steps              : 480\n",
      "COMET INFO:   Uploads:\n",
      "COMET INFO:     environment details      : 1\n",
      "COMET INFO:     filename                 : 1\n",
      "COMET INFO:     git metadata             : 1\n",
      "COMET INFO:     git-patch (uncompressed) : 1 (826 bytes)\n",
      "COMET INFO:     installed packages       : 1\n",
      "COMET INFO:     model graph              : 1\n",
      "COMET INFO:     notebook                 : 1\n",
      "COMET INFO:     source_code              : 1\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Uploading 1 metrics, params and output messages\n",
      "COMET INFO: Waiting for completion of the file uploads (may take several seconds)\n",
      "COMET INFO: The Python SDK has 10800 seconds to finish before aborting...\n",
      "COMET INFO: Still uploading 2 file(s), remaining 16.23 KB/16.82 KB\n",
      "COMET WARNING: As you are running in a Jupyter environment, you will need to call `experiment.end()` when finished to ensure all metrics and code are logged before exiting.\n",
      "COMET INFO: Experiment is live on comet.ml https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/51cc732f39a041c3a8153883bb169f4f\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/30\n",
      "480/480 [==============================] - 7s 12ms/step - loss: 1.7990 - accuracy: 0.3967 - val_loss: 1.7092 - val_accuracy: 0.4311\n",
      "Epoch 2/30\n",
      "480/480 [==============================] - 5s 11ms/step - loss: 1.4704 - accuracy: 0.4575 - val_loss: 1.5348 - val_accuracy: 0.4778\n",
      "Epoch 3/30\n",
      "480/480 [==============================] - 5s 11ms/step - loss: 1.3730 - accuracy: 0.4932 - val_loss: 1.5214 - val_accuracy: 0.4556\n",
      "Epoch 4/30\n",
      "480/480 [==============================] - 3s 5ms/step - loss: 1.2862 - accuracy: 0.5120 - val_loss: 1.4260 - val_accuracy: 0.4978\n",
      "Epoch 5/30\n",
      "480/480 [==============================] - 3s 5ms/step - loss: 1.2613 - accuracy: 0.5322 - val_loss: 1.3659 - val_accuracy: 0.5133\n",
      "Epoch 6/30\n",
      "480/480 [==============================] - 3s 6ms/step - loss: 1.2090 - accuracy: 0.5402 - val_loss: 1.5193 - val_accuracy: 0.4844\n",
      "Epoch 7/30\n",
      "480/480 [==============================] - 3s 7ms/step - loss: 1.1767 - accuracy: 0.5547 - val_loss: 1.3259 - val_accuracy: 0.5289\n",
      "Epoch 8/30\n",
      "480/480 [==============================] - 3s 5ms/step - loss: 1.1661 - accuracy: 0.5593 - val_loss: 1.5037 - val_accuracy: 0.4822\n",
      "Epoch 9/30\n",
      "480/480 [==============================] - 3s 5ms/step - loss: 1.1442 - accuracy: 0.5630 - val_loss: 1.5739 - val_accuracy: 0.4378\n",
      "Epoch 10/30\n",
      "480/480 [==============================] - 3s 6ms/step - loss: 1.1051 - accuracy: 0.5722 - val_loss: 1.4339 - val_accuracy: 0.4956\n",
      "Epoch 11/30\n",
      "480/480 [==============================] - 3s 7ms/step - loss: 1.0899 - accuracy: 0.5852 - val_loss: 1.6826 - val_accuracy: 0.4489\n",
      "Epoch 12/30\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.0649 - accuracy: 0.5822 - val_loss: 1.4208 - val_accuracy: 0.5333\n",
      "Epoch 1/30\n",
      "480/480 [==============================] - 3s 5ms/step - loss: 1.7790 - accuracy: 0.3989 - val_loss: 1.6225 - val_accuracy: 0.4156\n",
      "Epoch 2/30\n",
      "480/480 [==============================] - 3s 6ms/step - loss: 1.4756 - accuracy: 0.4642 - val_loss: 1.5273 - val_accuracy: 0.4689\n",
      "Epoch 3/30\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.3568 - accuracy: 0.4948 - val_loss: 1.5568 - val_accuracy: 0.4467\n",
      "Epoch 4/30\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.3007 - accuracy: 0.5154 - val_loss: 1.3634 - val_accuracy: 0.5244\n",
      "Epoch 5/30\n",
      "480/480 [==============================] - 3s 5ms/step - loss: 1.2364 - accuracy: 0.5361 - val_loss: 1.4653 - val_accuracy: 0.5000\n",
      "Epoch 6/30\n",
      "480/480 [==============================] - 3s 6ms/step - loss: 1.2224 - accuracy: 0.5414 - val_loss: 1.5036 - val_accuracy: 0.4733\n",
      "Epoch 7/30\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.1762 - accuracy: 0.5570 - val_loss: 1.6731 - val_accuracy: 0.4200\n",
      "Epoch 8/30\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.1428 - accuracy: 0.5645 - val_loss: 1.5723 - val_accuracy: 0.4311\n",
      "Epoch 9/30\n",
      "480/480 [==============================] - 2s 5ms/step - loss: 1.1357 - accuracy: 0.5712 - val_loss: 1.4208 - val_accuracy: 0.5067\n",
      "15/15 [==============================] - 0s 3ms/step - loss: 1.3634 - accuracy: 0.5244\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "COMET INFO: [1.3634496927261353, 0.5244444608688354]\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Comet.ml Experiment Summary\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO:   Data:\n",
      "COMET INFO:     display_summary_level : 1\n",
      "COMET INFO:     url                   : https://www.comet.ml/wodenwang820118/music-genre-multiclass-classification/51cc732f39a041c3a8153883bb169f4f\n",
      "COMET INFO:   Metrics [count] (min, max):\n",
      "COMET INFO:     accuracy [21]                : (0.39671790599823, 0.5851784348487854)\n",
      "COMET INFO:     batch_accuracy [1008]        : (0.0, 0.75)\n",
      "COMET INFO:     batch_loss [1008]            : (0.8131793737411499, 2.8120861053466797)\n",
      "COMET INFO:     epoch_duration [21]          : (2.452999999994063, 7.188000000009197)\n",
      "COMET INFO:     loss [21]                    : (1.0649036169052124, 1.7990400791168213)\n",
      "COMET INFO:     val_accuracy [21]            : (0.41555556654930115, 0.5333333611488342)\n",
      "COMET INFO:     val_loss [21]                : (1.3258519172668457, 1.7091999053955078)\n",
      "COMET INFO:     validate_batch_accuracy [63] : (0.25, 0.625)\n",
      "COMET INFO:     validate_batch_loss [63]     : (1.1157007217407227, 1.9270386695861816)\n",
      "COMET INFO:   Others:\n",
      "COMET INFO:     optimizer_count        : 10\n",
      "COMET INFO:     optimizer_id           : 019075b278c649c7b9842acd5f3dcf6d\n",
      "COMET INFO:     optimizer_metric       : loss\n",
      "COMET INFO:     optimizer_metric_value : 1.0649036169052124\n",
      "COMET INFO:     optimizer_objective    : minimum\n",
      "COMET INFO:     optimizer_parameters   : {\"audio_layer_1\": 550, \"audio_layer_2\": 400, \"batch_size\": 16, \"epochs\": 30, \"final_layer\": 64, \"meta_layer_1\": 300, \"meta_layer_2\": 450}\n",
      "COMET INFO:     optimizer_pid          : 9f3788f2b31b4d8b65b5555a9174c71878c8dd81\n",
      "COMET INFO:     optimizer_process      : 13176\n",
      "COMET INFO:     optimizer_trial        : 1\n",
      "COMET INFO:     optimizer_version      : 2.0.1\n",
      "COMET INFO:     trainable_params       : 500784\n",
      "COMET INFO:   Parameters:\n",
      "COMET INFO:     Adam_amsgrad       : False\n",
      "COMET INFO:     Adam_beta_1        : 0.9\n",
      "COMET INFO:     Adam_beta_2        : 0.999\n",
      "COMET INFO:     Adam_decay         : 0.0\n",
      "COMET INFO:     Adam_epsilon       : 1e-07\n",
      "COMET INFO:     Adam_learning_rate : 0.001\n",
      "COMET INFO:     Adam_name          : Adam\n",
      "COMET INFO:     Optimizer          : Adam\n",
      "COMET INFO:     audio_layer_1      : 550\n",
      "COMET INFO:     audio_layer_2      : 400\n",
      "COMET INFO:     batch_size         : 16\n",
      "COMET INFO:     epochs             : 30\n",
      "COMET INFO:     final_layer        : 64\n",
      "COMET INFO:     meta_layer_1       : 300\n",
      "COMET INFO:     meta_layer_2       : 450\n",
      "COMET INFO:     steps              : 480\n",
      "COMET INFO:   Uploads:\n",
      "COMET INFO:     environment details      : 1\n",
      "COMET INFO:     filename                 : 1\n",
      "COMET INFO:     git metadata             : 1\n",
      "COMET INFO:     git-patch (uncompressed) : 1 (826 bytes)\n",
      "COMET INFO:     installed packages       : 1\n",
      "COMET INFO:     model graph              : 1\n",
      "COMET INFO:     notebook                 : 1\n",
      "COMET INFO:     source_code              : 1\n",
      "COMET INFO: ---------------------------\n",
      "COMET INFO: Uploading 1 metrics, params and output messages\n",
      "COMET INFO: Waiting for completion of the file uploads (may take several seconds)\n",
      "COMET INFO: The Python SDK has 10800 seconds to finish before aborting...\n",
      "COMET INFO: Still uploading 2 file(s), remaining 16.23 KB/16.82 KB\n",
      "COMET INFO: Optimizer search 019075b278c649c7b9842acd5f3dcf6d has completed\n"
     ]
    }
   ],
   "source": [
    "classifier.grid_search({\n",
    "    \"algorithm\": \"bayes\",\n",
    "    \"name\": \"Optimize Music Classification Network\",\n",
    "    \"spec\": {\"maxCombo\": 10, \"objective\": \"minimize\", \"metric\": \"loss\"},\n",
    "    \"parameters\": {\n",
    "        \"audio_layer_1\": {\"type\": \"discrete\", \"values\": [400,450,500,550,600,700,900,1000]},\n",
    "        \"audio_layer_2\": {\"type\": \"discrete\", \"values\": [128,256,300,400,450,500,550]},\n",
    "        \"meta_layer_1\": {\"type\": \"discrete\", \"values\": [300,400,450,500,550,600,700,900,1000]},\n",
    "        \"meta_layer_2\": {\"type\": \"discrete\", \"values\": [128,256,300,400,450,500,550]},\n",
    "        \"final_layer\": {\"type\": \"discrete\", \"values\": [20,32,64,128,256]},\n",
    "        \"batch_size\": {\"type\": \"discrete\", \"values\": [16,32]},\n",
    "        \"epochs\": {\"type\": \"discrete\", \"values\": [15,20,25,30]},\n",
    "    },\n",
    "    \"trials\": 1,\n",
    "})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Build the model\n",
    "model = Sequential()\n",
    "model.add(\n",
    "    Dense(\n",
    "        units=450,\n",
    "        activation='elu',\n",
    "        input_shape=(num_train.shape[1],)\n",
    "    )\n",
    ")\n",
    "model.add(Dropout(0.5))\n",
    "model.add(BatchNormalization())\n",
    "\n",
    "model.add(\n",
    "    Dense(\n",
    "        units=256,\n",
    "        activation='elu'))\n",
    "model.add(Dropout(0.3))\n",
    "model.add(BatchNormalization())\n",
    "\n",
    "model_meta = Sequential()\n",
    "model_meta.add(\n",
    "    Dense(\n",
    "        units=900,\n",
    "        activation='elu',\n",
    "        input_shape=(num_train_meta.shape[1],)\n",
    "    )\n",
    ")\n",
    "model_meta.add(Dropout(0.5))\n",
    "model_meta.add(BatchNormalization())\n",
    "\n",
    "model_meta.add(\n",
    "    Dense(\n",
    "        units=400,\n",
    "        activation='elu'))\n",
    "model_meta.add(Dropout(0.3))\n",
    "model_meta.add(BatchNormalization())\n",
    "\n",
    "\n",
    "merged_out = Concatenate()([model.output,model_meta.output])\n",
    "merged_out = Dense(\n",
    "    units=32,\n",
    "    activation='elu')(merged_out)\n",
    "merged_out = Dense(8,activation='softmax')(merged_out)\n",
    "\n",
    "combined_model = Model([model.input] + [model_meta.input], [merged_out])\n",
    "combined_model.compile(\n",
    "    optimizer='adam',\n",
    "    loss='categorical_crossentropy',\n",
    "    metrics=['accuracy']\n",
    "        )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "__________________________________________________________________________________________________\n",
      " Layer (type)                   Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      " dense_input (InputLayer)       [(None, 148)]        0           []                               \n",
      "                                                                                                  \n",
      " dense_2_input (InputLayer)     [(None, 3)]          0           []                               \n",
      "                                                                                                  \n",
      " dense (Dense)                  (None, 450)          67050       ['dense_input[0][0]']            \n",
      "                                                                                                  \n",
      " dense_2 (Dense)                (None, 900)          3600        ['dense_2_input[0][0]']          \n",
      "                                                                                                  \n",
      " dropout (Dropout)              (None, 450)          0           ['dense[0][0]']                  \n",
      "                                                                                                  \n",
      " dropout_2 (Dropout)            (None, 900)          0           ['dense_2[0][0]']                \n",
      "                                                                                                  \n",
      " batch_normalization (BatchNorm  (None, 450)         1800        ['dropout[0][0]']                \n",
      " alization)                                                                                       \n",
      "                                                                                                  \n",
      " batch_normalization_2 (BatchNo  (None, 900)         3600        ['dropout_2[0][0]']              \n",
      " rmalization)                                                                                     \n",
      "                                                                                                  \n",
      " dense_1 (Dense)                (None, 256)          115456      ['batch_normalization[0][0]']    \n",
      "                                                                                                  \n",
      " dense_3 (Dense)                (None, 400)          360400      ['batch_normalization_2[0][0]']  \n",
      "                                                                                                  \n",
      " dropout_1 (Dropout)            (None, 256)          0           ['dense_1[0][0]']                \n",
      "                                                                                                  \n",
      " dropout_3 (Dropout)            (None, 400)          0           ['dense_3[0][0]']                \n",
      "                                                                                                  \n",
      " batch_normalization_1 (BatchNo  (None, 256)         1024        ['dropout_1[0][0]']              \n",
      " rmalization)                                                                                     \n",
      "                                                                                                  \n",
      " batch_normalization_3 (BatchNo  (None, 400)         1600        ['dropout_3[0][0]']              \n",
      " rmalization)                                                                                     \n",
      "                                                                                                  \n",
      " concatenate (Concatenate)      (None, 656)          0           ['batch_normalization_1[0][0]',  \n",
      "                                                                  'batch_normalization_3[0][0]']  \n",
      "                                                                                                  \n",
      " dense_4 (Dense)                (None, 32)           21024       ['concatenate[0][0]']            \n",
      "                                                                                                  \n",
      " dense_5 (Dense)                (None, 8)            264         ['dense_4[0][0]']                \n",
      "                                                                                                  \n",
      "==================================================================================================\n",
      "Total params: 575,818\n",
      "Trainable params: 571,806\n",
      "Non-trainable params: 4,012\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "combined_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "240/240 [==============================] - 3s 8ms/step - loss: 1.7233 - accuracy: 0.4064 - val_loss: 1.5251 - val_accuracy: 0.4111\n",
      "Epoch 2/20\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 1.4096 - accuracy: 0.4828 - val_loss: 1.3281 - val_accuracy: 0.5356\n",
      "Epoch 3/20\n",
      "240/240 [==============================] - 1s 6ms/step - loss: 1.3172 - accuracy: 0.5091 - val_loss: 1.4720 - val_accuracy: 0.4822\n",
      "Epoch 4/20\n",
      "240/240 [==============================] - 1s 6ms/step - loss: 1.2772 - accuracy: 0.5198 - val_loss: 1.3004 - val_accuracy: 0.5244\n",
      "Epoch 5/20\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 1.2109 - accuracy: 0.5483 - val_loss: 1.3472 - val_accuracy: 0.5378\n",
      "Epoch 6/20\n",
      "240/240 [==============================] - 1s 6ms/step - loss: 1.1852 - accuracy: 0.5520 - val_loss: 1.3688 - val_accuracy: 0.5289\n",
      "Epoch 7/20\n",
      "240/240 [==============================] - 2s 9ms/step - loss: 1.1598 - accuracy: 0.5525 - val_loss: 1.2768 - val_accuracy: 0.5844\n",
      "Epoch 8/20\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 1.1330 - accuracy: 0.5802 - val_loss: 1.3220 - val_accuracy: 0.5311\n",
      "Epoch 9/20\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 1.1153 - accuracy: 0.5788 - val_loss: 1.3040 - val_accuracy: 0.5333\n",
      "Epoch 10/20\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 1.0647 - accuracy: 0.5886 - val_loss: 1.3613 - val_accuracy: 0.5444\n",
      "Epoch 11/20\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 1.0612 - accuracy: 0.5964 - val_loss: 1.3673 - val_accuracy: 0.5244\n",
      "Epoch 12/20\n",
      "240/240 [==============================] - 1s 5ms/step - loss: 1.0575 - accuracy: 0.5940 - val_loss: 1.4162 - val_accuracy: 0.5267\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1ff28600c70>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combined_model.fit([num_train,num_train_meta],num_train_y,validation_data=([num_valid,num_valid_meta],num_valid_y),epochs=20,batch_size=32,shuffle=True,class_weight=class_weight,callbacks=[early_stop])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_test_data = df_test_features.iloc[:,9:]\n",
    "num_test_meta_data = df_test_features.iloc[:,3:5]\n",
    "num_test_meta_data['duration'] = df_test_features['duration']\n",
    "\n",
    "num_test = scaler.fit_transform(num_test_data)\n",
    "num_test_meta = scaler.fit_transform(num_test_meta_data)\n",
    "\n",
    "df_test_label = pd.read_csv('test_labels.csv')\n",
    "num_test_y = pd.get_dummies(df_test_label['genre'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                       precision    recall  f1-score   support\n",
      "\n",
      " classic pop and rock       0.46      0.65      0.54        55\n",
      "dance and electronica       0.62      0.40      0.49        45\n",
      "                 folk       0.45      0.56      0.50        64\n",
      "       jazz and blues       0.55      0.50      0.52        44\n",
      "                metal       0.82      0.68      0.74        66\n",
      "                  pop       0.63      0.54      0.58        74\n",
      "                 punk       0.46      0.55      0.50        44\n",
      "      soul and reggae       0.79      0.72      0.76        58\n",
      "\n",
      "             accuracy                           0.58       450\n",
      "            macro avg       0.60      0.58      0.58       450\n",
      "         weighted avg       0.61      0.58      0.59       450\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "validation = combined_model.predict([num_valid,num_valid_meta])\n",
    "validation = validation.argmax(axis=1)\n",
    "classes_mapping = {\n",
    "    0: 'classic pop and rock',\n",
    "    1: 'dance and electronica',\n",
    "    2: 'folk',\n",
    "    3: 'jazz and blues',\n",
    "    4: 'metal',\n",
    "    5: 'pop',\n",
    "    6: 'punk',\n",
    "    7: 'soul and reggae',\n",
    "}\n",
    "predict_label_array = np.vectorize(classes_mapping.get)(validation)\n",
    "correct_labels = df_valid_labels['genre'].values\n",
    "print(classification_report(correct_labels,predict_label_array))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "14/14 [==============================] - 0s 2ms/step - loss: 1.7532 - accuracy: 0.4276\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[1.753165602684021, 0.427570104598999]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combined_model.evaluate([num_test,num_test_meta],num_test_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "984944ce4eefbed249e2a3501e72ee0777f56e66d10a0a53283275062444400f"
  },
  "kernelspec": {
   "display_name": "Python 3.9.6 64-bit",
   "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": "3.9.6"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}