Demonstration.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "from os.path import basename\n",
    "from multiprocessing.dummy import Pool as ThreadPool\n",
    "import subprocess\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import pickle\n",
    "\n",
    "import glob\n",
    "from natsort import natsorted\n",
    "import tqdm\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from keras.layers import Input, Dense, concatenate\n",
    "from keras.models import Sequential, Model\n",
    "from os import environ\n",
    "import tensorflow as tf\n",
    "from keras.backend.tensorflow_backend import set_session\n",
    "\n",
    "# Libraries for data Processing\n",
    "from pydub import AudioSegment # Requires ffmpeg. You can do \"apt install ffmpeg\" if you don't have it\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# README"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "In this demonstration, we use our trained model to predict emotion from songs that we can find online.\n",
    "For model training, please refer to xxx.py\n",
    "\"\"\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Processing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 1: Converting mp3 to wav"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'You can use any music files you like. In this example, I use the first 1000 most popular songs from FMA'"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\"\"\" \n",
    "You can use any music files you like. \n",
    "In this example, I use the first 1000 most popular songs from FMA\n",
    "https://freemusicarchive.org/music/charts/all\n",
    "\n",
    "I also filtered out the songs that is less than 60 seconds and longer than 10 minutes.\n",
    "In the end, there are only 953 songs\n",
    "\"\"\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "inpath = './Z1-1000/' # Path to your mp3 files\n",
    "output_path = './Z1-1000wav/' # Path to save your wav files results \n",
    "# (opensmile only supports wav files, so if your music files are not this format, you need to convert it to wav first)\n",
    "# if your music files are already wav files, then you can skip this step\n",
    "mp3s = os.listdir(inpath)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def mp3_2_wav(mp3,inpath=inpath,output_path=output_path):\n",
    "    \"This function is for converting mp3 files to wav files, since openSMILE only works for wav files\"\n",
    "    sound = AudioSegment.from_mp3(inpath+mp3)\n",
    "    \n",
    "    if os.path.exists(output_path)==False: # Check if the output folder exist\n",
    "        os.makedirs(output_path) # If not, create the folder\n",
    "    else:\n",
    "        pass        \n",
    "    sound.export(os.path.join(output_path,mp3[:-3]+'wav'), format=\"wav\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None,\n",
       " None]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Processing data with multi-threads\n",
    "pool = ThreadPool(8)\n",
    "pool.map(mp3_2_wav, mp3s)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 2: Extracting features using opensmile"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# In this step, you need to install openSMILE to extract features from wav files.\n",
    "# For me, I put the complied software in the same folder as ./SMILExtract\n",
    "# The software can be downloaded here https://www.audeering.com/opensmile/\n",
    "\n",
    "# After you have openSMILE installed, we also need a configuration file to tell the software what features to extract\n",
    "# I have put the configuration file in the same folder as ./IS13_ComParE_lld-func.conf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "inpath = './Z1-1000wav/'\n",
    "outpath = './features/'\n",
    "\n",
    "def extractFeatures(inpath, outpath):\n",
    "    \"\"\"Extracting features from wav files by calling SMILExtract\"\"\"\n",
    "    \n",
    "    def runcommand(command):\n",
    "        subprocess.call(command, shell=True)    \n",
    "    commands = []\n",
    "    audios = os.listdir(inpath)\n",
    "    pool = ThreadPool(40)\n",
    "    \n",
    "    if os.path.exists(outpath)==False: # Check if the output folder exist\n",
    "        os.makedirs(outpath) # If not, create the folder\n",
    "    else:\n",
    "        pass        \n",
    "    \n",
    "    for audio in audios:\n",
    "        command1 = \"./SMILExtract -C IS13_ComParE_lld-func.conf -I \" + str(inpath) + audio + \" -O \" + str(outpath) + audio[:-3] + \"csv\"\n",
    "        commands.append(command1)\n",
    "    pool.map(runcommand, commands)\n",
    "\n",
    "\n",
    "extractFeatures(inpath,outpath)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 3: Loading openSMILE features as numpy arrays"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 953/953 [00:39<00:00, 24.21it/s]\n"
     ]
    }
   ],
   "source": [
    "length_list = []\n",
    "X_avg = [] # Create a list for storing the processed features\n",
    "filenames = []\n",
    "for file in tqdm.tqdm(natsorted(glob.glob('./features/*.csv'))):\n",
    "    \n",
    "    data = pd.read_csv(file,sep=';').drop('frameTime',axis=1)\n",
    "    audio_len = data.shape[0]\n",
    "    if audio_len < 600 and audio_len > 60: # filtering out songs that is more than 5 mins long and shorter than 30s\n",
    "        X_avg.append(data.values.mean(0))\n",
    "        filenames.append(os.path.basename(file[:-4]))\n",
    "    else:\n",
    "        pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_avg = np.array(X_avg) # Converting the list to numpy array."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 4: Data Normalization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_scaler = StandardScaler()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_avg_norm = X_scaler.fit_transform(X_avg)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# TNN model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 1: Defining the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "environ[\"CUDA_VISIBLE_DEVICES\"]='0' # Selecting GPU\n",
    "\n",
    "config= tf.ConfigProto()\n",
    "config.gpu_options.per_process_gpu_memory_fraction=0.2 # Use 20% of the GPU memory\n",
    "set_session(tf.Session(config=config))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /opt/conda/lib/python3.6/site-packages/tensorflow/python/ops/resource_variable_ops.py:435: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Colocations handled automatically by placer.\n"
     ]
    }
   ],
   "source": [
    "def create_base_network(in_dims, out_dims):\n",
    "    \"\"\"\n",
    "    Base network to be shared.\n",
    "    \"\"\"\n",
    "    model = Sequential()\n",
    "    model.add(Dense(out_dims, input_shape=(in_dims,),activation='relu'))\n",
    "    # model.add(Dense(1024, input_shape=(in_dims,),activation='relu'))\n",
    "    # model.add(Dense(512,activation='relu'))    \n",
    "    # model.add(Dense(256,activation='relu'))       \n",
    "    # model.add(Dense(out_dims,activation='relu')) \n",
    "    return model\n",
    "\n",
    "input_dim = 260\n",
    "anchor_input = Input((input_dim, ), name='anchor_input')\n",
    "positive_input = Input((input_dim, ), name='positive_input')\n",
    "negative_input = Input((input_dim, ), name='negative_input')\n",
    "\n",
    "# Shared embedding layer for positive and negative items\n",
    "Shared_DNN = create_base_network(input_dim, 100)\n",
    "\n",
    "\n",
    "encoded_anchor = Shared_DNN(anchor_input)\n",
    "\n",
    "TNN_model = Model(inputs=anchor_input, outputs=encoded_anchor)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_emotion(X_avg_norm, TNN_model, GMB_model, weight_path):\n",
    "    TNN_model.load_weights(weight_path)\n",
    "    transformed_X = TNN_model.predict(X_avg_norm)\n",
    "    y_pred = GMB_model.predict(transformed_X)\n",
    "    \n",
    "    return y_pred"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Arousal prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "GMB_model = pickle.load(open('./Weight/grad_model_arousal', 'rb'))\n",
    "weight_path = r'./Weight/arousal_task1_triplet_Dense100_88nd_r2-0.6875515232584031.hdf5'\n",
    "\n",
    "y_pred_arousal = compute_emotion(X_avg_norm, TNN_model, GMB_model, weight_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The arousal values for the songs are:\n",
      " [0.60029277 0.75608682 0.48365754 0.52802984 0.75206128 0.3475996\n",
      " 0.51095784 0.27620656 0.29008374 0.13615872 0.32529241 0.66620417\n",
      " 0.64644912 0.46950495 0.60356943 0.65952872 0.1946222  0.80575451\n",
      " 0.5298027  0.29695747 0.47394559 0.37818233 0.62971806 0.64373108\n",
      " 0.70583541 0.66552795 0.47221956 0.53286219 0.24297111 0.33618326\n",
      " 0.36949148 0.21926042 0.28591247 0.2973582  0.52341909 0.5405343\n",
      " 0.30432724 0.26231588 0.81359715 0.30879977 0.13935107 0.57243357\n",
      " 0.63311534 0.67250607 0.45848479 0.39766893 0.33935878 0.72301415\n",
      " 0.67852965 0.38985347 0.74912316 0.20864409 0.57520065 0.47782632\n",
      " 0.28876813 0.32495356 0.2213834  0.17504804 0.62580453 0.57077766\n",
      " 0.60059221 0.55541219 0.106008   0.37087794 0.49943433 0.50780272\n",
      " 0.39619975 0.37101101 0.30098521 0.70979963 0.63556836 0.32271404\n",
      " 0.47139024 0.50650232 0.48301373 0.6474467  0.45045513 0.32571971\n",
      " 0.61505276 0.50814057 0.72149024 0.22879308 0.58726062 0.49512384\n",
      " 0.68773702 0.59123877 0.39393083 0.85835934 0.59203072 0.75482622\n",
      " 0.14377093 0.61906311 0.67421512 0.21566951 0.2711515  0.73617375\n",
      " 0.40473147 0.5772045  0.50765073 0.50723537 0.30740514 0.50516503\n",
      " 0.78625576 0.23475227 0.56058096 0.24859916 0.74130199 0.30706876\n",
      " 0.79268495 0.48659126 0.21638347 0.6983599  0.24035965 0.67903426\n",
      " 0.73323234 0.62124386 0.61532711 0.11654285 0.63159075 0.4029224\n",
      " 0.42007099 0.61225439 0.67532964 0.57977472 0.73568076 0.55412517\n",
      " 0.68686893 0.6912405  0.69559256 0.7459359  0.75187339 0.78793422\n",
      " 0.77493577 0.30985339 0.60073777 0.21276453 0.53758988 0.70804399\n",
      " 0.21075795 0.22224465 0.69166781 0.52581362 0.34369435 0.85216152\n",
      " 0.4307073  0.40843978 0.34620545 0.6002379  0.61480864 0.53053482\n",
      " 0.59803599 0.35435924 0.31937513 0.76821905 0.69844855 0.56254117\n",
      " 0.30388385 0.4280528  0.32196517 0.67897969 0.35667416 0.56517303\n",
      " 0.411469   0.84013193 0.56428932 0.46873883 0.68708001 0.24468091\n",
      " 0.52538852 0.81450013 0.53041093 0.84053533 0.24066604 0.78590255\n",
      " 0.35577719 0.10351447 0.65425853 0.61928609 0.29693719 0.46699261\n",
      " 0.81581537 0.45878976 0.37787486 0.71618488 0.53105921 0.32134319\n",
      " 0.73647982 0.85405137 0.64725922 0.19216741 0.42766126 0.27436373\n",
      " 0.20045515 0.55338236 0.28543455 0.47896245 0.30560393 0.55214647\n",
      " 0.41027081 0.6605239  0.19765479 0.48697951 0.28195316 0.66580158\n",
      " 0.62034985 0.24598986 0.17352198 0.32052285 0.49657276 0.47808945\n",
      " 0.2338316  0.74915513 0.14920459 0.53133751 0.50094182 0.54869384\n",
      " 0.64109168 0.64249401 0.38411158 0.27599489 0.44077347 0.49526942\n",
      " 0.27285412 0.32565647 0.21438379 0.70189624 0.51666201 0.14329843\n",
      " 0.68976689 0.49462345 0.61295031 0.12583087 0.24875384 0.18851034\n",
      " 0.17020323 0.48860579 0.21187234 0.47397479 0.61422629 0.59254114\n",
      " 0.19575521 0.4755757  0.35850293 0.82057844 0.69643636 0.80959161\n",
      " 0.28394608 0.58004367 0.44033858 0.65011478 0.6984858  0.7706042\n",
      " 0.49574065 0.75994807 0.74389045 0.52210614 0.33560852 0.79121424\n",
      " 0.6832583  0.52335585 0.47999272 0.62924177 0.58523887 0.467522\n",
      " 0.4246039  0.4872959  0.15665484 0.53052815 0.27350704 0.59723075\n",
      " 0.66533843 0.28157106 0.37780461 0.45067817 0.38447456 0.80978431\n",
      " 0.34976762 0.7804273  0.19001754 0.60940239 0.72397851 0.29818965\n",
      " 0.26831142 0.37971027 0.33188817 0.85420557 0.13513925 0.6385\n",
      " 0.50034858 0.44553087 0.41507104 0.4173812  0.18618908 0.61645083\n",
      " 0.26279286 0.31682502 0.54923165 0.47532159 0.20113581 0.80168665\n",
      " 0.56322623 0.24953748 0.36337995 0.42815077 0.24202981 0.73879901\n",
      " 0.3157013  0.69675179 0.61495938 0.57693548 0.56414754 0.37816821\n",
      " 0.53111923 0.76538069 0.34834368 0.82628483 0.52015319 0.50760896\n",
      " 0.52141476 0.32393367 0.5079394  0.65433154 0.61232522 0.32852051\n",
      " 0.52571121 0.55972824 0.30966176 0.61685544 0.57319544 0.54636346\n",
      " 0.34089954 0.26694712 0.27532626 0.38712186 0.31042496 0.535511\n",
      " 0.52615687 0.28246062 0.8213899  0.89356947 0.24216416 0.28913784\n",
      " 0.45693258 0.6921449  0.55513396 0.51173418 0.80113554 0.72360866\n",
      " 0.63192922 0.29757127 0.84928774 0.36998337 0.4434874  0.604746\n",
      " 0.79180653 0.20503099 0.34472548 0.7190735  0.35116306 0.67794735\n",
      " 0.36916203 0.26322891 0.35817872 0.19308079 0.29530675 0.387933\n",
      " 0.22475048 0.33545085 0.6536011  0.5379405  0.36037484 0.73405073\n",
      " 0.29199453 0.82047071 0.64101235 0.71566816 0.88519336 0.37600641\n",
      " 0.44180308 0.51366372 0.52575809 0.7547061  0.57047527 0.53756929\n",
      " 0.584416   0.7302262  0.68362538 0.24601924 0.42300225 0.47666001\n",
      " 0.71209662 0.54190783 0.42630776 0.45401298 0.55803414 0.78029173\n",
      " 0.19210145 0.35792774 0.37675607 0.2063811  0.33199314 0.36954503\n",
      " 0.28061266 0.64200051 0.58859071 0.55761424 0.79680957 0.46350095\n",
      " 0.24223093 0.66512942 0.4177329  0.71360189 0.96415662 0.77612766\n",
      " 0.58478336 0.71260878 0.65659167 0.61519319 0.80120258 0.49474242\n",
      " 0.17688977 0.4670861  0.31446446 0.49963056 0.5926374  0.26524302\n",
      " 0.81889265 0.24636917 0.37816821 0.58932193 0.9818039  0.6571292\n",
      " 0.5161202  0.28110592 0.13569203 0.47747084 0.39060432 0.605967\n",
      " 0.23880905 0.50589319 0.47309113 0.37569408 0.43048586 0.60912023\n",
      " 0.70604028 0.38177874 0.64933826 0.81844669 0.64655628 0.6674457\n",
      " 0.52860046 0.21261712 0.52620092 0.17921223 0.39825461 0.49477642\n",
      " 0.44341265 0.50128896 0.56909272 0.38305821 0.29935302 0.3469064\n",
      " 0.51577298 0.45362269 0.82946457 0.33478231 0.45377767 0.47319106\n",
      " 0.57617764 0.74833396 0.71559617 0.44010682 0.61356683 0.82166232\n",
      " 0.88451623 0.52437731 0.67045654 0.6185132  0.45138488 0.2954085\n",
      " 0.6257991  0.5630268  0.6328169  0.49115573 0.36160152 0.46106347\n",
      " 0.6258285  0.68279185 0.68696564 0.47111508 0.64536058 0.35781356\n",
      " 0.36454251 0.53840524 0.44434395 0.52863046 0.82626168 0.51770385\n",
      " 0.61725449 0.53156023 0.3934493  0.55898292 0.77632055 0.71509809\n",
      " 0.33761731 0.46725097 0.14541053 0.71481106 0.29284631 0.62012607\n",
      " 0.35184925 0.20403932 0.60415934 0.45780457 0.5162024  0.38334617\n",
      " 0.53925797 0.94580963 0.25080552 0.576545   0.66298578 0.42711784\n",
      " 0.55100913 0.42383322 0.56140347 0.26792291 0.32291471 0.14913745\n",
      " 0.77333883 0.41980524 0.60317548 0.45640135 0.7810979  0.54959925\n",
      " 0.3775357  0.52224378 0.6377985  0.23055803 0.36684739 0.48639577\n",
      " 0.34254794 0.39536724 0.62291581 0.61101702 0.58553518 0.6830539\n",
      " 0.17098194 0.40684211 0.57359648 0.59831549 0.62380525 0.54721843\n",
      " 0.39622231 0.81832935 0.56931757 0.1631873  0.7165385  0.54455185\n",
      " 0.22295631 0.31275605 0.41088431 0.16883031 0.59424318 0.6982857\n",
      " 0.59221923 0.20534518 0.36630651 0.3568257  0.6149412  0.73440169\n",
      " 0.45010858 0.63121111 0.39270081 0.8530378  0.34589157 0.15596683\n",
      " 0.16527649 0.2201587  0.37424392 0.54846644 0.33415981 0.48335669\n",
      " 0.81851033 0.17412919 0.6644257  0.50634245 0.54721843 0.50937132\n",
      " 0.42911231 0.68752641 0.78201415 0.28518556 0.84394805 0.50329127\n",
      " 0.62310855 0.45317787 0.44131911 0.21855935 0.49554935 0.17042841\n",
      " 0.46770847 0.39824742 0.22537805 0.59278869 0.47117413 0.3049696\n",
      " 0.64911171 0.39929283 0.63663212 0.55626201 0.32281242 0.53868617\n",
      " 0.7301528  0.51816598 0.30636004 0.42754646 0.75247757 0.36616874\n",
      " 0.69030652 0.679975   0.78786363 0.57397408 0.52641742 0.47825315\n",
      " 0.46137439 0.33618124 0.36205693 0.14514148 0.2385538  0.27045408\n",
      " 0.73110875 0.64438533 0.45030507 0.54915102 0.70776192 0.23239763\n",
      " 0.27671251 0.56211995 0.52005379 0.49433605 0.59799744 0.19160061\n",
      " 0.50659124 0.23902425 0.63028988 0.66326176 0.78614745 0.43303653\n",
      " 0.47686287 0.37447337 0.1847668  0.31115304 0.58154322 0.19175249\n",
      " 0.41314741 0.34350359 0.59357368 0.26663809 0.6768105  0.68881131\n",
      " 0.71408474 0.64515249 0.48036291 0.65597319 0.67438726 0.68108932\n",
      " 0.22539035 0.15750969 0.56156794 0.42303511 0.45431669 0.58107739\n",
      " 0.23090222 0.31907561 0.64959313 0.62453006 0.38942006 0.38775145\n",
      " 0.73808661 0.49855577 0.51680337 0.47177563 0.22702729 0.46174611\n",
      " 0.67793522 0.63113808 0.32356545 0.165775   0.55435795 0.89179575\n",
      " 0.20770836 0.38855588 0.80788282 0.23191837 0.2777128  0.46044772\n",
      " 0.43024217 0.77176015 0.63795219 0.41993261 0.30382973 0.6567298\n",
      " 0.76962061 0.45953102 0.68846524 0.19714768 0.24999299 0.73538813\n",
      " 0.66066331 0.2525394  0.76657597 0.44939966 0.56809595 0.80170426\n",
      " 0.48237516 0.28682483 0.60468172 0.5954979  0.46510253 0.1718972\n",
      " 0.64724068 0.50584647 0.57483041 0.57517457 0.17890457 0.73918746\n",
      " 0.45025999 0.45236695 0.49027124 0.66071056 0.53136418 0.81093844\n",
      " 0.34359977 0.42426245 0.20326461 0.43444222 0.53339311 0.2138327\n",
      " 0.48911708 0.28846936 0.38566743 0.60943136 0.74652655 0.40682268\n",
      " 0.3683275  0.76842893 0.32991002 0.59188111]\n"
     ]
    }
   ],
   "source": [
    "print(f'The arousal values for the songs are:\\n {y_pred_arousal}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Valence prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "GMB_model = pickle.load(open('./Weight/grad_model_valence', 'rb'))\n",
    "weight_path = r'./Weight/task1_triplet_Dense260_89nd_r2-0.4087313673537267.hdf5'\n",
    "\n",
    "y_pred_valence= compute_emotion(X_avg_norm, TNN_model, GMB_model, weight_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The valence values for the songs are:\n",
      " [0.61284673 0.55987466 0.61039216 0.63504526 0.62195907 0.41294177\n",
      " 0.50887903 0.37833308 0.39447406 0.35428787 0.47628572 0.6012664\n",
      " 0.51563354 0.41060388 0.66865629 0.59711693 0.34419925 0.80636219\n",
      " 0.51879195 0.48136129 0.46053479 0.43156815 0.56270055 0.60619746\n",
      " 0.52463666 0.58827846 0.59823674 0.47956109 0.39212563 0.38358031\n",
      " 0.47350869 0.40988583 0.25500796 0.37659802 0.52600823 0.53353706\n",
      " 0.50426789 0.30919018 0.72445928 0.44130365 0.41083807 0.61404291\n",
      " 0.62543171 0.59928944 0.58396309 0.41472011 0.48756765 0.60779669\n",
      " 0.68166955 0.4813725  0.70605016 0.3468544  0.63888881 0.52800861\n",
      " 0.34246267 0.30003524 0.3789602  0.3186506  0.50371071 0.34391229\n",
      " 0.72562252 0.53288033 0.33927886 0.51468118 0.57683575 0.58422168\n",
      " 0.43605867 0.47802803 0.36216748 0.74665967 0.6208892  0.44474943\n",
      " 0.45214415 0.54780542 0.36698577 0.52049896 0.71769348 0.35246355\n",
      " 0.57774187 0.43315485 0.66729912 0.38208052 0.59229741 0.70264124\n",
      " 0.65446355 0.60058338 0.44620642 0.59416066 0.59275846 0.590397\n",
      " 0.29938821 0.56815063 0.7338961  0.31763791 0.49030556 0.67794945\n",
      " 0.38727702 0.55712748 0.51864023 0.58816268 0.33832222 0.42765915\n",
      " 0.67887006 0.38669901 0.61201125 0.53056953 0.59699767 0.42516845\n",
      " 0.63769219 0.58954445 0.42625037 0.4483056  0.36269638 0.63651752\n",
      " 0.64054581 0.41278388 0.61811181 0.28679734 0.67045301 0.46934492\n",
      " 0.61325465 0.59010568 0.53938942 0.40594347 0.66007371 0.43523131\n",
      " 0.57057783 0.66323534 0.60075337 0.6933351  0.70451663 0.59032719\n",
      " 0.5499836  0.48315469 0.59676721 0.47751983 0.45283624 0.68757501\n",
      " 0.31046495 0.30369188 0.64031375 0.32929624 0.54547666 0.48266238\n",
      " 0.50500593 0.64270673 0.41068323 0.52243602 0.65322813 0.56519503\n",
      " 0.52514643 0.54712247 0.53002337 0.69376063 0.65314643 0.52978619\n",
      " 0.30730134 0.46703773 0.40117341 0.60874407 0.51083298 0.56917734\n",
      " 0.44105026 0.6431468  0.61570386 0.56345735 0.71284947 0.39842865\n",
      " 0.41035812 0.82203869 0.51304821 0.64242384 0.36703818 0.65467176\n",
      " 0.30277398 0.47473098 0.69277557 0.59366245 0.43075699 0.64189312\n",
      " 0.72237911 0.50517768 0.39166955 0.64910598 0.40697017 0.42613069\n",
      " 0.7402626  0.50672209 0.65905834 0.43404006 0.45891926 0.38494902\n",
      " 0.36914655 0.47791917 0.52050416 0.36052837 0.36314244 0.62149571\n",
      " 0.6062941  0.51766199 0.35128382 0.4586048  0.47736849 0.66739082\n",
      " 0.58686401 0.32923642 0.30066682 0.34780216 0.62806577 0.43037916\n",
      " 0.42278359 0.5120682  0.3462992  0.47904674 0.40458627 0.58169169\n",
      " 0.66471215 0.51350899 0.39058075 0.44580087 0.48452724 0.54215296\n",
      " 0.28473133 0.47736835 0.30937861 0.71962348 0.51451227 0.3514626\n",
      " 0.65456943 0.65016724 0.54565001 0.46910486 0.44486485 0.4384244\n",
      " 0.3014452  0.61913117 0.38636072 0.4468261  0.69570818 0.42997945\n",
      " 0.33771981 0.54080948 0.52937874 0.64155886 0.71901379 0.72271958\n",
      " 0.41769148 0.63024579 0.59919278 0.5803785  0.56644853 0.66663244\n",
      " 0.49806393 0.57864119 0.65460655 0.46962719 0.42759637 0.66939877\n",
      " 0.614066   0.61735537 0.56516329 0.60148418 0.62106272 0.54751586\n",
      " 0.5257361  0.66559517 0.39230475 0.37941407 0.33731889 0.62978558\n",
      " 0.61252822 0.3252017  0.55059319 0.47389876 0.591259   0.64454046\n",
      " 0.41165261 0.7510845  0.42140325 0.59676699 0.60560094 0.4087233\n",
      " 0.49142208 0.51276853 0.41863976 0.72321797 0.40891207 0.70798133\n",
      " 0.69735842 0.37996638 0.47180427 0.63983038 0.4076409  0.63382119\n",
      " 0.37156173 0.33076964 0.5860153  0.40871461 0.28007229 0.66551066\n",
      " 0.44277168 0.42432123 0.41481027 0.46268829 0.40963911 0.60699959\n",
      " 0.32858266 0.73432253 0.44799493 0.55646759 0.54216009 0.50012464\n",
      " 0.62960413 0.59860839 0.40411856 0.73494441 0.58085478 0.38879651\n",
      " 0.57879182 0.33234211 0.55365933 0.69409076 0.62606947 0.46901857\n",
      " 0.68591018 0.6025272  0.39452473 0.58468721 0.4704862  0.58879897\n",
      " 0.42100949 0.42540753 0.50556212 0.61380374 0.40388457 0.52396586\n",
      " 0.55919637 0.42151899 0.62352725 0.76325022 0.31497545 0.31268906\n",
      " 0.56380692 0.5355829  0.55748719 0.54785163 0.58149175 0.71363148\n",
      " 0.77213219 0.47170693 0.81992779 0.39146148 0.52284486 0.58345428\n",
      " 0.80966846 0.40390524 0.53670686 0.62769932 0.36644244 0.74344637\n",
      " 0.54734304 0.30060151 0.4358522  0.42359782 0.31013768 0.41872222\n",
      " 0.39247193 0.38432998 0.73233844 0.48507939 0.55445672 0.69399936\n",
      " 0.4261379  0.67305732 0.65429983 0.72065264 0.83325903 0.43113041\n",
      " 0.56134353 0.44551568 0.60638328 0.68039063 0.55762529 0.53226448\n",
      " 0.74913357 0.59516968 0.61320735 0.38469313 0.5541483  0.44770038\n",
      " 0.74818333 0.49504028 0.46843237 0.53377821 0.65004068 0.68945079\n",
      " 0.34382695 0.58940262 0.55754459 0.34864981 0.52012898 0.32728644\n",
      " 0.35600077 0.55875018 0.60357225 0.46195208 0.64482516 0.54770324\n",
      " 0.37807544 0.54654885 0.35466865 0.64023863 0.75625838 0.6754392\n",
      " 0.55300052 0.67916319 0.78909453 0.66947225 0.7504541  0.52749613\n",
      " 0.40527236 0.59653143 0.38654686 0.51339875 0.64087163 0.39374771\n",
      " 0.71574046 0.40887512 0.50012464 0.62883265 0.73009498 0.57808263\n",
      " 0.5399669  0.3453646  0.36684225 0.71926062 0.52176972 0.50165476\n",
      " 0.28868428 0.54928513 0.39628028 0.46537007 0.48146624 0.50644802\n",
      " 0.66874192 0.44222375 0.48976398 0.58924808 0.5501619  0.62997421\n",
      " 0.61307441 0.30996414 0.58261103 0.44171647 0.52922362 0.50722437\n",
      " 0.47721433 0.63799955 0.62877358 0.59246688 0.48471648 0.3477734\n",
      " 0.49292033 0.49746414 0.77042764 0.50358988 0.43048103 0.44820548\n",
      " 0.61767093 0.7065919  0.71651426 0.36072157 0.42384792 0.62853387\n",
      " 0.71875724 0.64936956 0.6198372  0.57584891 0.65138906 0.38543041\n",
      " 0.68725811 0.50690773 0.66822321 0.6572564  0.34432826 0.44197895\n",
      " 0.61894592 0.65639176 0.62389969 0.61174007 0.60654085 0.53078439\n",
      " 0.45484814 0.63055021 0.63636842 0.55701118 0.73607422 0.45313523\n",
      " 0.51599201 0.62509537 0.52191224 0.49868169 0.51458761 0.64177611\n",
      " 0.44524293 0.37660253 0.33053747 0.65537493 0.4755882  0.59408122\n",
      " 0.41832137 0.43081016 0.55232109 0.49268743 0.44252603 0.52260999\n",
      " 0.76611642 0.80674917 0.36470636 0.54443551 0.49050246 0.40984578\n",
      " 0.55534307 0.36916246 0.57848379 0.39758401 0.4967003  0.28032882\n",
      " 0.52762495 0.51683867 0.56289515 0.56129036 0.53244163 0.61420052\n",
      " 0.46232944 0.57680992 0.53353841 0.35243049 0.45375691 0.47464407\n",
      " 0.44510531 0.4406697  0.55502508 0.51319782 0.46783596 0.60566784\n",
      " 0.36638735 0.45976091 0.52062734 0.57776754 0.62061174 0.47708467\n",
      " 0.47665975 0.79075891 0.65556466 0.25917307 0.64680634 0.52635311\n",
      " 0.34575417 0.49146251 0.51437321 0.18075033 0.65273337 0.68689428\n",
      " 0.4364738  0.36080913 0.50011175 0.45162531 0.50707021 0.7011226\n",
      " 0.44035153 0.60680007 0.535635   0.69272332 0.46411676 0.41391746\n",
      " 0.30651532 0.28982008 0.53390024 0.55367559 0.29916235 0.51639835\n",
      " 0.7170054  0.32499675 0.65354177 0.69709847 0.47708467 0.57372506\n",
      " 0.39310251 0.53501794 0.51266443 0.4732438  0.80493874 0.50376909\n",
      " 0.52908705 0.40672395 0.48551872 0.4103795  0.48971844 0.30108418\n",
      " 0.36020113 0.54624796 0.35548225 0.55378501 0.4568565  0.48379663\n",
      " 0.64114328 0.49370292 0.65393376 0.54999776 0.31314056 0.60792185\n",
      " 0.69015595 0.61842535 0.33755771 0.481553   0.7587454  0.38964685\n",
      " 0.69569294 0.70505283 0.6701305  0.60516442 0.39793699 0.64654764\n",
      " 0.4352905  0.38920738 0.42641214 0.41935722 0.31579882 0.38863737\n",
      " 0.80026907 0.61275032 0.36558211 0.57292343 0.61580827 0.37373814\n",
      " 0.37189637 0.57333111 0.58892978 0.54853322 0.661229   0.30191528\n",
      " 0.47525528 0.42366265 0.59823727 0.59808993 0.5827159  0.4673909\n",
      " 0.52463612 0.52082989 0.5531203  0.30902086 0.60209609 0.30708218\n",
      " 0.50596653 0.36636632 0.54225549 0.26022189 0.56665279 0.73672429\n",
      " 0.67852524 0.53837463 0.58392745 0.6439232  0.65492752 0.65000021\n",
      " 0.44987398 0.39728682 0.60979508 0.43437104 0.45717693 0.51989008\n",
      " 0.36255352 0.39272118 0.68109247 0.56882923 0.39722543 0.51849394\n",
      " 0.5351474  0.60256006 0.6744249  0.51006902 0.3954339  0.44207191\n",
      " 0.67080643 0.6361218  0.38761745 0.32670473 0.4335695  0.71254661\n",
      " 0.45292876 0.34038006 0.75444166 0.21232934 0.28001037 0.56030434\n",
      " 0.57086622 0.60853999 0.61098265 0.38223038 0.45868969 0.44649367\n",
      " 0.64873895 0.50496514 0.65323293 0.4017033  0.49354934 0.67867925\n",
      " 0.43342492 0.32951133 0.77047853 0.49235637 0.53965974 0.66464263\n",
      " 0.61334171 0.39062911 0.63105777 0.56252436 0.65389195 0.39960687\n",
      " 0.51117428 0.5976866  0.64036206 0.45719881 0.45010404 0.68458958\n",
      " 0.56374854 0.54199892 0.35305169 0.56583937 0.54779028 0.70260398\n",
      " 0.48759188 0.47248156 0.34944716 0.44884188 0.64366818 0.5122322\n",
      " 0.47135502 0.39204934 0.36773023 0.70063057 0.6486144  0.49872615\n",
      " 0.45561732 0.51817813 0.39569991 0.45492754]\n"
     ]
    }
   ],
   "source": [
    "print(f'The valence values for the songs are:\\n {y_pred_valence}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Arousal')"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(y_pred_valence,y_pred_arousal)\n",
    "plt.xlabel('Valence')\n",
    "plt.ylabel('Arousal')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Chart_1',\n",
       " 'Chart_2',\n",
       " 'Chart_3',\n",
       " 'Chart_4',\n",
       " 'Chart_5',\n",
       " 'Chart_6',\n",
       " 'Chart_7',\n",
       " 'Chart_8',\n",
       " 'Chart_9',\n",
       " 'Chart_10',\n",
       " 'Chart_11',\n",
       " 'Chart_12',\n",
       " 'Chart_13',\n",
       " 'Chart_14',\n",
       " 'Chart_15',\n",
       " 'Chart_16',\n",
       " 'Chart_17',\n",
       " 'Chart_19',\n",
       " 'Chart_20',\n",
       " 'Chart_21',\n",
       " 'Chart_22',\n",
       " 'Chart_23',\n",
       " 'Chart_24',\n",
       " 'Chart_25',\n",
       " 'Chart_26',\n",
       " 'Chart_28',\n",
       " 'Chart_29',\n",
       " 'Chart_30',\n",
       " 'Chart_31',\n",
       " 'Chart_32',\n",
       " 'Chart_33',\n",
       " 'Chart_34',\n",
       " 'Chart_35',\n",
       " 'Chart_36',\n",
       " 'Chart_37',\n",
       " 'Chart_38',\n",
       " 'Chart_39',\n",
       " 'Chart_40',\n",
       " 'Chart_41',\n",
       " 'Chart_43',\n",
       " 'Chart_44',\n",
       " 'Chart_45',\n",
       " 'Chart_46',\n",
       " 'Chart_47',\n",
       " 'Chart_48',\n",
       " 'Chart_51',\n",
       " 'Chart_54',\n",
       " 'Chart_55',\n",
       " 'Chart_57',\n",
       " 'Chart_58',\n",
       " 'Chart_59',\n",
       " 'Chart_60',\n",
       " 'Chart_61',\n",
       " 'Chart_62',\n",
       " 'Chart_64',\n",
       " 'Chart_65',\n",
       " 'Chart_66',\n",
       " 'Chart_67',\n",
       " 'Chart_68',\n",
       " 'Chart_71',\n",
       " 'Chart_72',\n",
       " 'Chart_73',\n",
       " 'Chart_74',\n",
       " 'Chart_76',\n",
       " 'Chart_77',\n",
       " 'Chart_78',\n",
       " 'Chart_79',\n",
       " 'Chart_80',\n",
       " 'Chart_81',\n",
       " 'Chart_83',\n",
       " 'Chart_85',\n",
       " 'Chart_86',\n",
       " 'Chart_87',\n",
       " 'Chart_90',\n",
       " 'Chart_91',\n",
       " 'Chart_92',\n",
       " 'Chart_93',\n",
       " 'Chart_94',\n",
       " 'Chart_95',\n",
       " 'Chart_96',\n",
       " 'Chart_97',\n",
       " 'Chart_98',\n",
       " 'Chart_102',\n",
       " 'Chart_103',\n",
       " 'Chart_105',\n",
       " 'Chart_106',\n",
       " 'Chart_107',\n",
       " 'Chart_108',\n",
       " 'Chart_109',\n",
       " 'Chart_110',\n",
       " 'Chart_111',\n",
       " 'Chart_113',\n",
       " 'Chart_114',\n",
       " 'Chart_115',\n",
       " 'Chart_116',\n",
       " 'Chart_117',\n",
       " 'Chart_118',\n",
       " 'Chart_120',\n",
       " 'Chart_121',\n",
       " 'Chart_122',\n",
       " 'Chart_123',\n",
       " 'Chart_124',\n",
       " 'Chart_125',\n",
       " 'Chart_128',\n",
       " 'Chart_129',\n",
       " 'Chart_130',\n",
       " 'Chart_131',\n",
       " 'Chart_132',\n",
       " 'Chart_133',\n",
       " 'Chart_134',\n",
       " 'Chart_135',\n",
       " 'Chart_139',\n",
       " 'Chart_140',\n",
       " 'Chart_141',\n",
       " 'Chart_142',\n",
       " 'Chart_143',\n",
       " 'Chart_144',\n",
       " 'Chart_145',\n",
       " 'Chart_146',\n",
       " 'Chart_147',\n",
       " 'Chart_150',\n",
       " 'Chart_151',\n",
       " 'Chart_153',\n",
       " 'Chart_154',\n",
       " 'Chart_155',\n",
       " 'Chart_156',\n",
       " 'Chart_157',\n",
       " 'Chart_158',\n",
       " 'Chart_160',\n",
       " 'Chart_162',\n",
       " 'Chart_163',\n",
       " 'Chart_164',\n",
       " 'Chart_165',\n",
       " 'Chart_166',\n",
       " 'Chart_167',\n",
       " 'Chart_168',\n",
       " 'Chart_170',\n",
       " 'Chart_171',\n",
       " 'Chart_172',\n",
       " 'Chart_173',\n",
       " 'Chart_174',\n",
       " 'Chart_175',\n",
       " 'Chart_176',\n",
       " 'Chart_177',\n",
       " 'Chart_178',\n",
       " 'Chart_179',\n",
       " 'Chart_180',\n",
       " 'Chart_181',\n",
       " 'Chart_182',\n",
       " 'Chart_184',\n",
       " 'Chart_185',\n",
       " 'Chart_186',\n",
       " 'Chart_187',\n",
       " 'Chart_188',\n",
       " 'Chart_189',\n",
       " 'Chart_192',\n",
       " 'Chart_193',\n",
       " 'Chart_194',\n",
       " 'Chart_195',\n",
       " 'Chart_196',\n",
       " 'Chart_198',\n",
       " 'Chart_199',\n",
       " 'Chart_200',\n",
       " 'Chart_201',\n",
       " 'Chart_202',\n",
       " 'Chart_204',\n",
       " 'Chart_205',\n",
       " 'Chart_206',\n",
       " 'Chart_207',\n",
       " 'Chart_209',\n",
       " 'Chart_210',\n",
       " 'Chart_211',\n",
       " 'Chart_212',\n",
       " 'Chart_213',\n",
       " 'Chart_214',\n",
       " 'Chart_216',\n",
       " 'Chart_217',\n",
       " 'Chart_218',\n",
       " 'Chart_219',\n",
       " 'Chart_220',\n",
       " 'Chart_221',\n",
       " 'Chart_222',\n",
       " 'Chart_223',\n",
       " 'Chart_224',\n",
       " 'Chart_225',\n",
       " 'Chart_226',\n",
       " 'Chart_227',\n",
       " 'Chart_228',\n",
       " 'Chart_229',\n",
       " 'Chart_232',\n",
       " 'Chart_233',\n",
       " 'Chart_235',\n",
       " 'Chart_236',\n",
       " 'Chart_237',\n",
       " 'Chart_239',\n",
       " 'Chart_240',\n",
       " 'Chart_241',\n",
       " 'Chart_242',\n",
       " 'Chart_243',\n",
       " 'Chart_245',\n",
       " 'Chart_246',\n",
       " 'Chart_249',\n",
       " 'Chart_250',\n",
       " 'Chart_252',\n",
       " 'Chart_253',\n",
       " 'Chart_254',\n",
       " 'Chart_255',\n",
       " 'Chart_256',\n",
       " 'Chart_257',\n",
       " 'Chart_258',\n",
       " 'Chart_260',\n",
       " 'Chart_261',\n",
       " 'Chart_262',\n",
       " 'Chart_263',\n",
       " 'Chart_265',\n",
       " 'Chart_266',\n",
       " 'Chart_267',\n",
       " 'Chart_268',\n",
       " 'Chart_270',\n",
       " 'Chart_271',\n",
       " 'Chart_272',\n",
       " 'Chart_273',\n",
       " 'Chart_275',\n",
       " 'Chart_276',\n",
       " 'Chart_277',\n",
       " 'Chart_280',\n",
       " 'Chart_281',\n",
       " 'Chart_283',\n",
       " 'Chart_284',\n",
       " 'Chart_286',\n",
       " 'Chart_287',\n",
       " 'Chart_288',\n",
       " 'Chart_289',\n",
       " 'Chart_290',\n",
       " 'Chart_291',\n",
       " 'Chart_292',\n",
       " 'Chart_293',\n",
       " 'Chart_294',\n",
       " 'Chart_295',\n",
       " 'Chart_296',\n",
       " 'Chart_298',\n",
       " 'Chart_299',\n",
       " 'Chart_300',\n",
       " 'Chart_301',\n",
       " 'Chart_303',\n",
       " 'Chart_304',\n",
       " 'Chart_305',\n",
       " 'Chart_306',\n",
       " 'Chart_309',\n",
       " 'Chart_310',\n",
       " 'Chart_311',\n",
       " 'Chart_312',\n",
       " 'Chart_314',\n",
       " 'Chart_315',\n",
       " 'Chart_316',\n",
       " 'Chart_318',\n",
       " 'Chart_320',\n",
       " 'Chart_322',\n",
       " 'Chart_323',\n",
       " 'Chart_324',\n",
       " 'Chart_325',\n",
       " 'Chart_326',\n",
       " 'Chart_327',\n",
       " 'Chart_328',\n",
       " 'Chart_329',\n",
       " 'Chart_331',\n",
       " 'Chart_332',\n",
       " 'Chart_333',\n",
       " 'Chart_334',\n",
       " 'Chart_335',\n",
       " 'Chart_336',\n",
       " 'Chart_337',\n",
       " 'Chart_338',\n",
       " 'Chart_339',\n",
       " 'Chart_340',\n",
       " 'Chart_341',\n",
       " 'Chart_342',\n",
       " 'Chart_343',\n",
       " 'Chart_344',\n",
       " 'Chart_345',\n",
       " 'Chart_346',\n",
       " 'Chart_348',\n",
       " 'Chart_349',\n",
       " 'Chart_350',\n",
       " 'Chart_351',\n",
       " 'Chart_352',\n",
       " 'Chart_353',\n",
       " 'Chart_354',\n",
       " 'Chart_355',\n",
       " 'Chart_356',\n",
       " 'Chart_357',\n",
       " 'Chart_358',\n",
       " 'Chart_359',\n",
       " 'Chart_360',\n",
       " 'Chart_361',\n",
       " 'Chart_362',\n",
       " 'Chart_363',\n",
       " 'Chart_364',\n",
       " 'Chart_366',\n",
       " 'Chart_367',\n",
       " 'Chart_368',\n",
       " 'Chart_369',\n",
       " 'Chart_370',\n",
       " 'Chart_371',\n",
       " 'Chart_374',\n",
       " 'Chart_376',\n",
       " 'Chart_377',\n",
       " 'Chart_378',\n",
       " 'Chart_379',\n",
       " 'Chart_380',\n",
       " 'Chart_381',\n",
       " 'Chart_382',\n",
       " 'Chart_383',\n",
       " 'Chart_384',\n",
       " 'Chart_385',\n",
       " 'Chart_386',\n",
       " 'Chart_387',\n",
       " 'Chart_388',\n",
       " 'Chart_389',\n",
       " 'Chart_390',\n",
       " 'Chart_391',\n",
       " 'Chart_392',\n",
       " 'Chart_393',\n",
       " 'Chart_394',\n",
       " 'Chart_395',\n",
       " 'Chart_396',\n",
       " 'Chart_397',\n",
       " 'Chart_398',\n",
       " 'Chart_399',\n",
       " 'Chart_400',\n",
       " 'Chart_401',\n",
       " 'Chart_402',\n",
       " 'Chart_403',\n",
       " 'Chart_404',\n",
       " 'Chart_405',\n",
       " 'Chart_406',\n",
       " 'Chart_408',\n",
       " 'Chart_409',\n",
       " 'Chart_410',\n",
       " 'Chart_411',\n",
       " 'Chart_412',\n",
       " 'Chart_413',\n",
       " 'Chart_416',\n",
       " 'Chart_419',\n",
       " 'Chart_420',\n",
       " 'Chart_421',\n",
       " 'Chart_422',\n",
       " 'Chart_423',\n",
       " 'Chart_425',\n",
       " 'Chart_426',\n",
       " 'Chart_427',\n",
       " 'Chart_428',\n",
       " 'Chart_429',\n",
       " 'Chart_430',\n",
       " 'Chart_433',\n",
       " 'Chart_435',\n",
       " 'Chart_436',\n",
       " 'Chart_437',\n",
       " 'Chart_439',\n",
       " 'Chart_440',\n",
       " 'Chart_441',\n",
       " 'Chart_442',\n",
       " 'Chart_443',\n",
       " 'Chart_444',\n",
       " 'Chart_445',\n",
       " 'Chart_447',\n",
       " 'Chart_448',\n",
       " 'Chart_449',\n",
       " 'Chart_450',\n",
       " 'Chart_451',\n",
       " 'Chart_452',\n",
       " 'Chart_453',\n",
       " 'Chart_455',\n",
       " 'Chart_456',\n",
       " 'Chart_457',\n",
       " 'Chart_458',\n",
       " 'Chart_459',\n",
       " 'Chart_460',\n",
       " 'Chart_461',\n",
       " 'Chart_462',\n",
       " 'Chart_463',\n",
       " 'Chart_465',\n",
       " 'Chart_467',\n",
       " 'Chart_468',\n",
       " 'Chart_470',\n",
       " 'Chart_471',\n",
       " 'Chart_472',\n",
       " 'Chart_473',\n",
       " 'Chart_474',\n",
       " 'Chart_475',\n",
       " 'Chart_476',\n",
       " 'Chart_478',\n",
       " 'Chart_479',\n",
       " 'Chart_481',\n",
       " 'Chart_482',\n",
       " 'Chart_484',\n",
       " 'Chart_486',\n",
       " 'Chart_487',\n",
       " 'Chart_488',\n",
       " 'Chart_489',\n",
       " 'Chart_490',\n",
       " 'Chart_491',\n",
       " 'Chart_492',\n",
       " 'Chart_493',\n",
       " 'Chart_495',\n",
       " 'Chart_496',\n",
       " 'Chart_497',\n",
       " 'Chart_499',\n",
       " 'Chart_501',\n",
       " 'Chart_502',\n",
       " 'Chart_503',\n",
       " 'Chart_504',\n",
       " 'Chart_505',\n",
       " 'Chart_507',\n",
       " 'Chart_508',\n",
       " 'Chart_509',\n",
       " 'Chart_510',\n",
       " 'Chart_511',\n",
       " 'Chart_512',\n",
       " 'Chart_513',\n",
       " 'Chart_516',\n",
       " 'Chart_517',\n",
       " 'Chart_518',\n",
       " 'Chart_519',\n",
       " 'Chart_521',\n",
       " 'Chart_522',\n",
       " 'Chart_523',\n",
       " 'Chart_524',\n",
       " 'Chart_526',\n",
       " 'Chart_527',\n",
       " 'Chart_528',\n",
       " 'Chart_529',\n",
       " 'Chart_530',\n",
       " 'Chart_531',\n",
       " 'Chart_532',\n",
       " 'Chart_534',\n",
       " 'Chart_536',\n",
       " 'Chart_537',\n",
       " 'Chart_538',\n",
       " 'Chart_539',\n",
       " 'Chart_540',\n",
       " 'Chart_543',\n",
       " 'Chart_544',\n",
       " 'Chart_545',\n",
       " 'Chart_546',\n",
       " 'Chart_547',\n",
       " 'Chart_548',\n",
       " 'Chart_549',\n",
       " 'Chart_550',\n",
       " 'Chart_551',\n",
       " 'Chart_552',\n",
       " 'Chart_554',\n",
       " 'Chart_555',\n",
       " 'Chart_556',\n",
       " 'Chart_557',\n",
       " 'Chart_559',\n",
       " 'Chart_561',\n",
       " 'Chart_564',\n",
       " 'Chart_565',\n",
       " 'Chart_566',\n",
       " 'Chart_568',\n",
       " 'Chart_569',\n",
       " 'Chart_570',\n",
       " 'Chart_572',\n",
       " 'Chart_573',\n",
       " 'Chart_574',\n",
       " 'Chart_575',\n",
       " 'Chart_576',\n",
       " 'Chart_577',\n",
       " 'Chart_578',\n",
       " 'Chart_580',\n",
       " 'Chart_581',\n",
       " 'Chart_582',\n",
       " 'Chart_584',\n",
       " 'Chart_585',\n",
       " 'Chart_586',\n",
       " 'Chart_590',\n",
       " 'Chart_591',\n",
       " 'Chart_592',\n",
       " 'Chart_593',\n",
       " 'Chart_596',\n",
       " 'Chart_598',\n",
       " 'Chart_600',\n",
       " 'Chart_602',\n",
       " 'Chart_604',\n",
       " 'Chart_605',\n",
       " 'Chart_606',\n",
       " 'Chart_607',\n",
       " 'Chart_608',\n",
       " 'Chart_609',\n",
       " 'Chart_612',\n",
       " 'Chart_613',\n",
       " 'Chart_614',\n",
       " 'Chart_615',\n",
       " 'Chart_616',\n",
       " 'Chart_617',\n",
       " 'Chart_619',\n",
       " 'Chart_620',\n",
       " 'Chart_622',\n",
       " 'Chart_624',\n",
       " 'Chart_625',\n",
       " 'Chart_626',\n",
       " 'Chart_627',\n",
       " 'Chart_628',\n",
       " 'Chart_629',\n",
       " 'Chart_631',\n",
       " 'Chart_632',\n",
       " 'Chart_633',\n",
       " 'Chart_634',\n",
       " 'Chart_635',\n",
       " 'Chart_636',\n",
       " 'Chart_640',\n",
       " 'Chart_642',\n",
       " 'Chart_643',\n",
       " 'Chart_644',\n",
       " 'Chart_645',\n",
       " 'Chart_647',\n",
       " 'Chart_648',\n",
       " 'Chart_649',\n",
       " 'Chart_650',\n",
       " 'Chart_653',\n",
       " 'Chart_654',\n",
       " 'Chart_655',\n",
       " 'Chart_656',\n",
       " 'Chart_657',\n",
       " 'Chart_659',\n",
       " 'Chart_660',\n",
       " 'Chart_661',\n",
       " 'Chart_662',\n",
       " 'Chart_663',\n",
       " 'Chart_664',\n",
       " 'Chart_666',\n",
       " 'Chart_667',\n",
       " 'Chart_668',\n",
       " 'Chart_669',\n",
       " 'Chart_671',\n",
       " 'Chart_673',\n",
       " 'Chart_674',\n",
       " 'Chart_675',\n",
       " 'Chart_676',\n",
       " 'Chart_677',\n",
       " 'Chart_680',\n",
       " 'Chart_681',\n",
       " 'Chart_682',\n",
       " 'Chart_683',\n",
       " 'Chart_684',\n",
       " 'Chart_685',\n",
       " 'Chart_686',\n",
       " 'Chart_687',\n",
       " 'Chart_688',\n",
       " 'Chart_690',\n",
       " 'Chart_691',\n",
       " 'Chart_692',\n",
       " 'Chart_693',\n",
       " 'Chart_694',\n",
       " 'Chart_695',\n",
       " 'Chart_696',\n",
       " 'Chart_697',\n",
       " 'Chart_698',\n",
       " 'Chart_699',\n",
       " 'Chart_701',\n",
       " 'Chart_702',\n",
       " 'Chart_703',\n",
       " 'Chart_704',\n",
       " 'Chart_708',\n",
       " 'Chart_711',\n",
       " 'Chart_712',\n",
       " 'Chart_715',\n",
       " 'Chart_718',\n",
       " 'Chart_719',\n",
       " 'Chart_720',\n",
       " 'Chart_721',\n",
       " 'Chart_722',\n",
       " 'Chart_724',\n",
       " 'Chart_725',\n",
       " 'Chart_726',\n",
       " 'Chart_727',\n",
       " 'Chart_729',\n",
       " 'Chart_730',\n",
       " 'Chart_731',\n",
       " 'Chart_732',\n",
       " 'Chart_734',\n",
       " 'Chart_735',\n",
       " 'Chart_736',\n",
       " 'Chart_737',\n",
       " 'Chart_738',\n",
       " 'Chart_739',\n",
       " 'Chart_740',\n",
       " 'Chart_741',\n",
       " 'Chart_743',\n",
       " 'Chart_745',\n",
       " 'Chart_747',\n",
       " 'Chart_748',\n",
       " 'Chart_749',\n",
       " 'Chart_750',\n",
       " 'Chart_751',\n",
       " 'Chart_752',\n",
       " 'Chart_753',\n",
       " 'Chart_755',\n",
       " 'Chart_757',\n",
       " 'Chart_758',\n",
       " 'Chart_760',\n",
       " 'Chart_761',\n",
       " 'Chart_762',\n",
       " 'Chart_763',\n",
       " 'Chart_765',\n",
       " 'Chart_766',\n",
       " 'Chart_767',\n",
       " 'Chart_769',\n",
       " 'Chart_770',\n",
       " 'Chart_771',\n",
       " 'Chart_772',\n",
       " 'Chart_775',\n",
       " 'Chart_777',\n",
       " 'Chart_779',\n",
       " 'Chart_783',\n",
       " 'Chart_784',\n",
       " 'Chart_785',\n",
       " 'Chart_786',\n",
       " 'Chart_787',\n",
       " 'Chart_788',\n",
       " 'Chart_789',\n",
       " 'Chart_791',\n",
       " 'Chart_792',\n",
       " 'Chart_793',\n",
       " 'Chart_794',\n",
       " 'Chart_795',\n",
       " 'Chart_796',\n",
       " 'Chart_799',\n",
       " 'Chart_800',\n",
       " 'Chart_801',\n",
       " 'Chart_804',\n",
       " 'Chart_805',\n",
       " 'Chart_806',\n",
       " 'Chart_807',\n",
       " 'Chart_808',\n",
       " 'Chart_809',\n",
       " 'Chart_810',\n",
       " 'Chart_811',\n",
       " 'Chart_814',\n",
       " 'Chart_815',\n",
       " 'Chart_816',\n",
       " 'Chart_818',\n",
       " 'Chart_821',\n",
       " 'Chart_822',\n",
       " 'Chart_823',\n",
       " 'Chart_824',\n",
       " 'Chart_825',\n",
       " 'Chart_826',\n",
       " 'Chart_827',\n",
       " 'Chart_828',\n",
       " 'Chart_829',\n",
       " 'Chart_832',\n",
       " 'Chart_833',\n",
       " 'Chart_834',\n",
       " 'Chart_836',\n",
       " 'Chart_838',\n",
       " 'Chart_839',\n",
       " 'Chart_840',\n",
       " 'Chart_843',\n",
       " 'Chart_844',\n",
       " 'Chart_845',\n",
       " 'Chart_846',\n",
       " 'Chart_848',\n",
       " 'Chart_849',\n",
       " 'Chart_850',\n",
       " 'Chart_851',\n",
       " 'Chart_852',\n",
       " 'Chart_853',\n",
       " 'Chart_854',\n",
       " 'Chart_855',\n",
       " 'Chart_856',\n",
       " 'Chart_857',\n",
       " 'Chart_858',\n",
       " 'Chart_860',\n",
       " 'Chart_862',\n",
       " 'Chart_863',\n",
       " 'Chart_864',\n",
       " 'Chart_866',\n",
       " 'Chart_867',\n",
       " 'Chart_869',\n",
       " 'Chart_870',\n",
       " 'Chart_871',\n",
       " 'Chart_872',\n",
       " 'Chart_873',\n",
       " 'Chart_874',\n",
       " 'Chart_875',\n",
       " 'Chart_876',\n",
       " 'Chart_877',\n",
       " 'Chart_878',\n",
       " 'Chart_879',\n",
       " 'Chart_880',\n",
       " 'Chart_881',\n",
       " 'Chart_882',\n",
       " 'Chart_883',\n",
       " 'Chart_884',\n",
       " 'Chart_885',\n",
       " 'Chart_886',\n",
       " 'Chart_887',\n",
       " 'Chart_888',\n",
       " 'Chart_889',\n",
       " 'Chart_890',\n",
       " 'Chart_891',\n",
       " 'Chart_892',\n",
       " 'Chart_893',\n",
       " 'Chart_894',\n",
       " 'Chart_895',\n",
       " 'Chart_896',\n",
       " 'Chart_897',\n",
       " 'Chart_900',\n",
       " 'Chart_901',\n",
       " 'Chart_902',\n",
       " 'Chart_903',\n",
       " 'Chart_904',\n",
       " 'Chart_905',\n",
       " 'Chart_906',\n",
       " 'Chart_908',\n",
       " 'Chart_909',\n",
       " 'Chart_910',\n",
       " 'Chart_911',\n",
       " 'Chart_912',\n",
       " 'Chart_913',\n",
       " 'Chart_915',\n",
       " 'Chart_916',\n",
       " 'Chart_918',\n",
       " 'Chart_921',\n",
       " 'Chart_926',\n",
       " 'Chart_928',\n",
       " 'Chart_929',\n",
       " 'Chart_930',\n",
       " 'Chart_931',\n",
       " 'Chart_934',\n",
       " 'Chart_935',\n",
       " 'Chart_936',\n",
       " 'Chart_937',\n",
       " 'Chart_938',\n",
       " 'Chart_939',\n",
       " 'Chart_942',\n",
       " 'Chart_943',\n",
       " 'Chart_944',\n",
       " 'Chart_946',\n",
       " 'Chart_948',\n",
       " 'Chart_949',\n",
       " 'Chart_950',\n",
       " 'Chart_952',\n",
       " 'Chart_953',\n",
       " 'Chart_954',\n",
       " 'Chart_955']"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filenames"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.DataFrame(zip(filenames, y_pred_arousal, y_pred_valence), columns=['songID', 'arousal','valence'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.to_csv('FMA_av_prediction')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}