Piano Generator - Autoencoder.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "0153a6e1",
   "metadata": {},
   "outputs": [],
   "source": [
    "PIANO_GENERATOR_MODEL_PATH = 'piano_model'\n",
    "PIANO_GENERATOR_MODEL_WEIGHTS_PATH = './piano models/piano_model_autoencoder_weights'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "7a25788a",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "import glob\n",
    "import pickle\n",
    "import numpy\n",
    "import pandas\n",
    "import numpy as np\n",
    "from itertools import chain\n",
    "from pydub import AudioSegment\n",
    "from music21 import converter, instrument, note, chord, stream\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "from keras.layers import Dropout\n",
    "from keras.layers import LSTM\n",
    "from keras.layers import Activation\n",
    "from keras.layers import BatchNormalization as BatchNorm\n",
    "from keras.utils import np_utils\n",
    "from keras.callbacks import ModelCheckpoint\n",
    "\n",
    "import keras\n",
    "from keras import layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "bdf41ccc",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_notes(path):\n",
    "    \"\"\" Get all the notes and chords from the midi files in the 'path' directory \"\"\"\n",
    "    notes = []\n",
    "\n",
    "    print(len(glob.glob(path + '/*.mid')))\n",
    "    for file in glob.glob(path + '/*.mid'):\n",
    "        midi = converter.parse(file)\n",
    "\n",
    "#         print('parsing %s' % file)\n",
    "\n",
    "        notes_to_parse = None\n",
    "\n",
    "        try: # file has instrument parts\n",
    "            s2 = instrument.partitionByInstrument(midi)\n",
    "            notes_to_parse = s2.parts[0].recurse() \n",
    "        except: # file has notes in a flat structure\n",
    "            notes_to_parse = midi.flat.notes\n",
    "\n",
    "        for element in notes_to_parse:\n",
    "            if isinstance(element, note.Note):\n",
    "                notes.append(str(element.pitch))\n",
    "            elif isinstance(element, chord.Chord):\n",
    "                notes.append('.'.join(str(n) for n in element.normalOrder))\n",
    "\n",
    "    with open('data/notes', 'wb') as filepath:\n",
    "        pickle.dump(notes, filepath)\n",
    "\n",
    "    return notes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "beb85420",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_notes_v2(path):\n",
    "    \"\"\" Get all the notes and chords from the midi files in the 'path' directory \"\"\"\n",
    "    notes = []\n",
    "\n",
    "    print(len(glob.glob(path + '/*.mid')))\n",
    "    for file in glob.glob(path + '/*.mid'):\n",
    "        midi = converter.parse(file)\n",
    "\n",
    "#         print('parsing %s' % file)\n",
    "\n",
    "        notes_to_parse = None\n",
    "\n",
    "        try: # file has instrument parts\n",
    "            s2 = instrument.partitionByInstrument(midi)\n",
    "            notes_to_parse = s2.parts[0].recurse() \n",
    "        except: # file has notes in a flat structure\n",
    "            notes_to_parse = midi.flat.notes\n",
    "        \n",
    "        file_notes = []\n",
    "        for element in notes_to_parse:\n",
    "            if isinstance(element, note.Note):\n",
    "                file_notes.append(str(element.pitch))\n",
    "            elif isinstance(element, chord.Chord):\n",
    "                file_notes.append('.'.join(str(n) for n in element.normalOrder))\n",
    "        \n",
    "        notes.append(file_notes)\n",
    "    with open('data/notes', 'wb') as filepath:\n",
    "        pickle.dump(notes, filepath)\n",
    "\n",
    "    return notes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "8fc92a25",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "93\n"
     ]
    }
   ],
   "source": [
    "notes = get_notes_v2('midi_songs')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "411c1c3c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "93"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(notes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "951aa577",
   "metadata": {},
   "outputs": [],
   "source": [
    "notes_len = [len(file_notes) for file_notes in notes]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "d486b370",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(20.43010752688172, 0)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mean(notes_len), min(notes_len)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "002e9c49",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0, 3, 4, 4, 5, 5, 5, 6, 6, 6]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sorted(notes_len)[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "932044a5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "79"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(np.array(notes_len) >= 8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "be005eda",
   "metadata": {},
   "outputs": [],
   "source": [
    "def prepare_sequences(notes, n_vocab):\n",
    "    \"\"\" Prepare the sequences used by the Neural Network \"\"\"\n",
    "    sequence_length = 32\n",
    "\n",
    "    # get all pitch names\n",
    "    pitchnames = sorted(set(item for item in notes))\n",
    "\n",
    "     # create a dictionary to map pitches to integers\n",
    "    note_to_int = dict((note, number) for number, note in enumerate(pitchnames))\n",
    "\n",
    "    network_input = []\n",
    "    network_output = []\n",
    "\n",
    "    # create input sequences and the corresponding outputs\n",
    "    for i in range(0, len(notes) - sequence_length, 1):\n",
    "        sequence_in = notes[i:i + sequence_length]\n",
    "        sequence_out = notes[i + sequence_length]\n",
    "        network_input.append([note_to_int[char] for char in sequence_in])\n",
    "        network_output.append(note_to_int[sequence_out])\n",
    "\n",
    "    n_patterns = len(network_input)\n",
    "\n",
    "    # reshape the input into a format compatible with LSTM layers\n",
    "    network_input = numpy.reshape(network_input, (n_patterns, sequence_length, 1))\n",
    "    # normalize input\n",
    "    network_input = network_input / float(n_vocab)\n",
    "\n",
    "    network_output = np_utils.to_categorical(network_output)\n",
    "\n",
    "    return (network_input, network_output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "e6168808",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "93\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "79"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(len(notes))\n",
    "len([note for note in notes if len(note) >= 8])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "36c38a69",
   "metadata": {},
   "outputs": [],
   "source": [
    "def prepare_sequences_v2(notes, n_vocab):\n",
    "    \"\"\" Prepare the sequences used by the Neural Network \"\"\"\n",
    "    sequence_length = 32\n",
    "\n",
    "    # get all pitch names\n",
    "    pitchnames = sorted(set(item for item in notes))\n",
    "\n",
    "     # create a dictionary to map pitches to integers\n",
    "    note_to_int = dict((note, number) for number, note in enumerate(pitchnames))\n",
    "\n",
    "    network_input = []\n",
    "    network_output = []\n",
    "\n",
    "    # create input sequences and the corresponding outputs\n",
    "    for i in range(0, len(notes) - sequence_length, 1):\n",
    "        sequence_in = notes[i:i + sequence_length]\n",
    "        sequence_out = notes[i + sequence_length]\n",
    "        network_input.append([note_to_int[char] for char in sequence_in])\n",
    "        network_output.append(note_to_int[sequence_out])\n",
    "\n",
    "    n_patterns = len(network_input)\n",
    "\n",
    "    # reshape the input into a format compatible with LSTM layers\n",
    "    network_input = numpy.reshape(network_input, (n_patterns, sequence_length, 1))\n",
    "    # normalize input\n",
    "    network_input = network_input / float(n_vocab)\n",
    "\n",
    "    network_output = np_utils.to_categorical(network_output)\n",
    "\n",
    "    return (network_input, network_output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "b690b9d8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[['2.3.7.10',\n",
       "  '10.2',\n",
       "  '7.10.0.3',\n",
       "  '10.2',\n",
       "  '5.9',\n",
       "  '10.2',\n",
       "  '3.7',\n",
       "  'D4',\n",
       "  '3.7',\n",
       "  '10.2',\n",
       "  '0.3.7',\n",
       "  '10',\n",
       "  '10.2',\n",
       "  '5.9',\n",
       "  '10.2'],\n",
       " ['6.9.11.2',\n",
       "  'D5',\n",
       "  '4.7.9.0',\n",
       "  '11.0.4.7',\n",
       "  'F#5',\n",
       "  'E5',\n",
       "  'D5',\n",
       "  'B4',\n",
       "  '6.9.11.2',\n",
       "  'D5',\n",
       "  '0.4.7',\n",
       "  '11.0.4.7',\n",
       "  '6.7.11.2'],\n",
       " ['8.11', '3.6', '8.11', 'E-4', 'B3', '8.11.3', 'F#4', '8.11.3'],\n",
       " ['4.6.9.11',\n",
       "  '1',\n",
       "  '11',\n",
       "  '9.11.1.4',\n",
       "  'C#4',\n",
       "  '9.11.1.4',\n",
       "  '8.11.1.4',\n",
       "  '1',\n",
       "  '7.10.0.3',\n",
       "  'C#4',\n",
       "  'B1',\n",
       "  '2.6.9',\n",
       "  '4',\n",
       "  'C#4',\n",
       "  '11',\n",
       "  'B1',\n",
       "  '2.6.9',\n",
       "  'C#4',\n",
       "  'D2',\n",
       "  '6.9.1',\n",
       "  'C#4',\n",
       "  'C#4',\n",
       "  '9',\n",
       "  '4.6.9.11',\n",
       "  '1',\n",
       "  '11',\n",
       "  '9.11.1.4',\n",
       "  'C#4',\n",
       "  '9.11.1.4',\n",
       "  '8.11.1.4',\n",
       "  '1',\n",
       "  '7.10.0.3',\n",
       "  'C#4',\n",
       "  'B1',\n",
       "  '2.6.9',\n",
       "  '4',\n",
       "  'C#4',\n",
       "  '6',\n",
       "  'B1',\n",
       "  '2.6.9',\n",
       "  'C#4',\n",
       "  'D2',\n",
       "  '6.9.1',\n",
       "  '1.4',\n",
       "  '9.1'],\n",
       " ['11.4',\n",
       "  '7',\n",
       "  'F#4',\n",
       "  'G3',\n",
       "  'D5',\n",
       "  'B5',\n",
       "  '11.0',\n",
       "  '9.2',\n",
       "  '7.11.2',\n",
       "  'B5',\n",
       "  'F#4',\n",
       "  '7.11.2',\n",
       "  '7.0',\n",
       "  '9.0',\n",
       "  '0.4.7',\n",
       "  '7',\n",
       "  '7.0',\n",
       "  'F#4',\n",
       "  '11.2',\n",
       "  'B5',\n",
       "  '7.0',\n",
       "  '0.2',\n",
       "  '7.9',\n",
       "  'C4',\n",
       "  'D5',\n",
       "  'B3',\n",
       "  'F#4',\n",
       "  '9.2',\n",
       "  'B5',\n",
       "  '7.0',\n",
       "  'G5',\n",
       "  '11.4',\n",
       "  '7',\n",
       "  'F#4',\n",
       "  'G3',\n",
       "  'D5',\n",
       "  '11.0',\n",
       "  '9.2',\n",
       "  '7.11.2',\n",
       "  'F#4',\n",
       "  '11.2',\n",
       "  '7.0',\n",
       "  'C5',\n",
       "  '0.4.7',\n",
       "  'G4',\n",
       "  '7.0',\n",
       "  'F#4',\n",
       "  '11.2',\n",
       "  '7.0',\n",
       "  '0.2',\n",
       "  '7.9',\n",
       "  'C4',\n",
       "  'B3',\n",
       "  'F#4',\n",
       "  '9.2',\n",
       "  '7.0'],\n",
       " ['9.11', 'D4', 'E4', 'F#4', '6.7', 'G4', 'F#4', 'E4', 'D4', 'A3', 'B3'],\n",
       " ['E3',\n",
       "  'E4',\n",
       "  'G7',\n",
       "  'G4',\n",
       "  'B4',\n",
       "  'D5',\n",
       "  'E5',\n",
       "  'G5',\n",
       "  'B5',\n",
       "  'D6',\n",
       "  'G6',\n",
       "  'B7',\n",
       "  'F#6',\n",
       "  'C8',\n",
       "  'B6',\n",
       "  'D8',\n",
       "  'E6',\n",
       "  '0.4',\n",
       "  'C4',\n",
       "  'E6',\n",
       "  'B6',\n",
       "  'E4',\n",
       "  'G4',\n",
       "  'B4',\n",
       "  'C5',\n",
       "  'E5',\n",
       "  'G5',\n",
       "  'B5',\n",
       "  'E6',\n",
       "  'D6',\n",
       "  'G6',\n",
       "  'B5',\n",
       "  'E3',\n",
       "  'E4',\n",
       "  'G7',\n",
       "  'G6',\n",
       "  'B5',\n",
       "  'G4',\n",
       "  'B4',\n",
       "  'D5',\n",
       "  'E5',\n",
       "  'G5',\n",
       "  'B5',\n",
       "  'D6',\n",
       "  'G6',\n",
       "  'B7',\n",
       "  'F#6',\n",
       "  'C8',\n",
       "  '0.2.4',\n",
       "  '2.8',\n",
       "  '9.0',\n",
       "  'C4',\n",
       "  'A8',\n",
       "  '2.8',\n",
       "  'E4',\n",
       "  'G4',\n",
       "  'B4',\n",
       "  'C5',\n",
       "  'E5',\n",
       "  'G5',\n",
       "  'B5',\n",
       "  'E6',\n",
       "  'D6',\n",
       "  '7.11'],\n",
       " ['2.6.7',\n",
       "  'B4',\n",
       "  'B4',\n",
       "  'D5',\n",
       "  '6.9.11.2',\n",
       "  'C#5',\n",
       "  '6.11',\n",
       "  '2.4.6',\n",
       "  'B4',\n",
       "  'B4',\n",
       "  'D5',\n",
       "  '2.6',\n",
       "  'A4',\n",
       "  'C#5',\n",
       "  'B4',\n",
       "  'F#5',\n",
       "  'B4',\n",
       "  'A4',\n",
       "  'F#5',\n",
       "  'E5',\n",
       "  '2.6.7',\n",
       "  'B4',\n",
       "  'B4',\n",
       "  'D5',\n",
       "  '6.9.11.2',\n",
       "  'C#5',\n",
       "  '6.11',\n",
       "  '2.4.6',\n",
       "  'B4',\n",
       "  'B4',\n",
       "  'D5',\n",
       "  '2.6',\n",
       "  'A4',\n",
       "  'C#5',\n",
       "  'B4',\n",
       "  'F#5',\n",
       "  'B4',\n",
       "  'A4',\n",
       "  'F#5',\n",
       "  'E5'],\n",
       " ['1.4.8',\n",
       "  'F#5',\n",
       "  'A5',\n",
       "  'G#5',\n",
       "  'F#5',\n",
       "  '11.3',\n",
       "  '6.11',\n",
       "  '8.11.3',\n",
       "  'E5',\n",
       "  'G#5',\n",
       "  'F#5',\n",
       "  'E5',\n",
       "  '1.4.6.9',\n",
       "  'F#4',\n",
       "  '3.6.8.11',\n",
       "  '8.9.1.4'],\n",
       " ['4.7.9.11.0',\n",
       "  'E4',\n",
       "  'G4',\n",
       "  'A5',\n",
       "  'E5',\n",
       "  'A4',\n",
       "  '4.5.9.0',\n",
       "  'A4',\n",
       "  '2.4.7.9',\n",
       "  '4.7.9.0',\n",
       "  'A4',\n",
       "  '4.7.9.11.0',\n",
       "  'E4',\n",
       "  'G4',\n",
       "  'A5',\n",
       "  'E5',\n",
       "  'A4',\n",
       "  '4.5.9.0',\n",
       "  'A4',\n",
       "  '2.4.7.9',\n",
       "  '4.7.9.0',\n",
       "  'A4'],\n",
       " ['G#4',\n",
       "  'E3',\n",
       "  'E-4',\n",
       "  'E3',\n",
       "  'G#4',\n",
       "  'E3',\n",
       "  'E-4',\n",
       "  'E3',\n",
       "  'G#4',\n",
       "  'E3',\n",
       "  'E-4',\n",
       "  'E3',\n",
       "  'G#4',\n",
       "  '11.4',\n",
       "  'E3',\n",
       "  'G#4',\n",
       "  '3.4',\n",
       "  'E3',\n",
       "  'G#4',\n",
       "  '3.4',\n",
       "  'E3',\n",
       "  'G#4',\n",
       "  'E3',\n",
       "  '3.4',\n",
       "  'G#4',\n",
       "  'E3',\n",
       "  '11.4',\n",
       "  'G#4',\n",
       "  'E3',\n",
       "  'E-4',\n",
       "  '4.8',\n",
       "  'E3',\n",
       "  '3.4',\n",
       "  'G#4',\n",
       "  'E3',\n",
       "  '3.4',\n",
       "  'G#4',\n",
       "  'E3',\n",
       "  '11.4',\n",
       "  'G#4',\n",
       "  'E3',\n",
       "  '3.4',\n",
       "  'G#4',\n",
       "  'E3',\n",
       "  '3.4',\n",
       "  'G#4',\n",
       "  'E3',\n",
       "  '3.4',\n",
       "  'G#4',\n",
       "  'E3',\n",
       "  'B4',\n",
       "  'E3'],\n",
       " ['4.7.11',\n",
       "  'E4',\n",
       "  'G4',\n",
       "  'E5',\n",
       "  '6.9.11.2',\n",
       "  'F#4',\n",
       "  'F#5',\n",
       "  'E5',\n",
       "  'G5',\n",
       "  'B5',\n",
       "  '4.7.11',\n",
       "  'E4',\n",
       "  'G4',\n",
       "  'E5',\n",
       "  '6.9.11.2',\n",
       "  'F#4',\n",
       "  'F#5',\n",
       "  'E5',\n",
       "  'G5',\n",
       "  'B5'],\n",
       " ['D2',\n",
       "  '2.6.9',\n",
       "  'C#4',\n",
       "  'D3',\n",
       "  'F#3',\n",
       "  '4.9',\n",
       "  '1.6',\n",
       "  'D3',\n",
       "  'A2',\n",
       "  '7.11.1',\n",
       "  'E4',\n",
       "  '7.11',\n",
       "  '1.7',\n",
       "  '6.9.10',\n",
       "  '6.7.10.1',\n",
       "  '6.11',\n",
       "  'A3',\n",
       "  '1.2.6',\n",
       "  'A3',\n",
       "  '1.2.6',\n",
       "  'A3',\n",
       "  'A2',\n",
       "  'G2',\n",
       "  '11.2',\n",
       "  'G3',\n",
       "  'G3',\n",
       "  '2.6.9',\n",
       "  '2.6.9',\n",
       "  'G3',\n",
       "  '2.6.9',\n",
       "  'G2',\n",
       "  'D2',\n",
       "  '2.6.9',\n",
       "  'C#4',\n",
       "  'D3',\n",
       "  'F#3',\n",
       "  '4.9',\n",
       "  '1.6',\n",
       "  'D3',\n",
       "  'A2',\n",
       "  '7.11.1',\n",
       "  'E4',\n",
       "  '7.11',\n",
       "  '1.7',\n",
       "  '6.9.10',\n",
       "  '6.7.10.1',\n",
       "  '6.11',\n",
       "  'A3',\n",
       "  '1.2.6',\n",
       "  'A3',\n",
       "  '1.2.6',\n",
       "  'A3',\n",
       "  'A2',\n",
       "  'G2',\n",
       "  '11.2',\n",
       "  'G3',\n",
       "  'G3',\n",
       "  '2.6.9',\n",
       "  '2.6.9',\n",
       "  'G3',\n",
       "  '2.6.9',\n",
       "  'G2'],\n",
       " ['8.11.3',\n",
       "  'F#4',\n",
       "  'B4',\n",
       "  'E5',\n",
       "  'E-5',\n",
       "  'G#4',\n",
       "  'F#4',\n",
       "  'C#4',\n",
       "  '8.9.11.1.4',\n",
       "  '1.4.6.8.9',\n",
       "  '8.11.3',\n",
       "  'F#4',\n",
       "  'B4',\n",
       "  'E5',\n",
       "  'E-5',\n",
       "  'G#4',\n",
       "  'F#4',\n",
       "  'C#4',\n",
       "  '8.9.11.1.4',\n",
       "  '1.4.6.8.9',\n",
       "  '3.6.8.11'],\n",
       " ['C#5',\n",
       "  'B-4',\n",
       "  'C#5',\n",
       "  '10.1',\n",
       "  '10.1',\n",
       "  'B-4',\n",
       "  'C#5',\n",
       "  '10.1',\n",
       "  'B-4',\n",
       "  'C#5',\n",
       "  '10.1',\n",
       "  'B-4',\n",
       "  'C#5',\n",
       "  'B-4',\n",
       "  'C#5',\n",
       "  'B-4',\n",
       "  'C#5',\n",
       "  '10.1',\n",
       "  'B-4',\n",
       "  '10.1',\n",
       "  'C#5',\n",
       "  'B-4',\n",
       "  'C#5',\n",
       "  '10.1',\n",
       "  'B-4',\n",
       "  'E-5',\n",
       "  'F5',\n",
       "  'G#5',\n",
       "  'B-5',\n",
       "  'G#5',\n",
       "  'F5',\n",
       "  '1.3.5',\n",
       "  'E-5',\n",
       "  'C#5',\n",
       "  '8.10',\n",
       "  'B-4',\n",
       "  'C#5',\n",
       "  'E-5',\n",
       "  'F5',\n",
       "  'G#5',\n",
       "  'B-5',\n",
       "  'C#6',\n",
       "  'E-6',\n",
       "  'F6',\n",
       "  'B-6',\n",
       "  'G#6',\n",
       "  'F6',\n",
       "  '3.5',\n",
       "  'E-6',\n",
       "  'C#6',\n",
       "  'B-5',\n",
       "  'G#5',\n",
       "  'F5',\n",
       "  '3.5',\n",
       "  'E-5',\n",
       "  'C#5',\n",
       "  '8.10',\n",
       "  'F4'],\n",
       " ['C3',\n",
       "  'E-5',\n",
       "  '10.0',\n",
       "  '3.7',\n",
       "  '2.3.7',\n",
       "  'E3',\n",
       "  '5.9',\n",
       "  '3.5',\n",
       "  '0.3.5',\n",
       "  'A2',\n",
       "  'D3',\n",
       "  'A4',\n",
       "  'C4',\n",
       "  '0.5',\n",
       "  'F#3',\n",
       "  '5.7',\n",
       "  '7.11',\n",
       "  '2.5.7',\n",
       "  'B2',\n",
       "  'C3',\n",
       "  '10.0',\n",
       "  '3.7',\n",
       "  '2.3.7',\n",
       "  'E3',\n",
       "  '5.9',\n",
       "  '3.5',\n",
       "  '0.3.5',\n",
       "  'A2',\n",
       "  '9.2',\n",
       "  'C4',\n",
       "  '0.5',\n",
       "  'F#3',\n",
       "  '5.7',\n",
       "  '7.11',\n",
       "  '2.5.7',\n",
       "  'G5',\n",
       "  '5.11'],\n",
       " ['4.5.7.9.0',\n",
       "  '9.0.2.5',\n",
       "  '11.2.4.7',\n",
       "  '4.5.9.0',\n",
       "  '4.5.7.9.0',\n",
       "  '9.0.2.5',\n",
       "  '11.2.4.7',\n",
       "  '4.5.9.0'],\n",
       " ['C2',\n",
       "  '0.2.3.7',\n",
       "  'B-4',\n",
       "  'F5',\n",
       "  'D5',\n",
       "  'F2',\n",
       "  '10.0.2.5',\n",
       "  'A4',\n",
       "  'F5',\n",
       "  'D5',\n",
       "  'G5',\n",
       "  'C5',\n",
       "  'C2',\n",
       "  '0.2.3.7',\n",
       "  'B-4',\n",
       "  'F5',\n",
       "  'D5',\n",
       "  'F2',\n",
       "  'B-1',\n",
       "  '10.0.2.5',\n",
       "  'A4',\n",
       "  'F5',\n",
       "  'D5',\n",
       "  'G2',\n",
       "  'G5',\n",
       "  'C5'],\n",
       " ['4.7.9.0',\n",
       "  '11.2.4.7',\n",
       "  'D6',\n",
       "  'B5',\n",
       "  'G5',\n",
       "  '4.7.9.0',\n",
       "  'A5',\n",
       "  'G5',\n",
       "  'B5',\n",
       "  'E5',\n",
       "  'E3',\n",
       "  '7.11.2',\n",
       "  'E5'],\n",
       " ['D4',\n",
       "  'B2',\n",
       "  'F#3',\n",
       "  'E4',\n",
       "  'C#4',\n",
       "  'F#4',\n",
       "  'D4',\n",
       "  'A4',\n",
       "  'E4',\n",
       "  'B4',\n",
       "  'C#5',\n",
       "  'F#4',\n",
       "  'A4',\n",
       "  '9.11',\n",
       "  'D4',\n",
       "  'B2',\n",
       "  'F#3',\n",
       "  'E4',\n",
       "  'C#4',\n",
       "  'C#5',\n",
       "  'A4',\n",
       "  'F#4',\n",
       "  '1.6',\n",
       "  'E3',\n",
       "  'B3',\n",
       "  'G4',\n",
       "  'F#4',\n",
       "  'B4',\n",
       "  'G4',\n",
       "  'D5',\n",
       "  '4.9',\n",
       "  'F#5',\n",
       "  'B4',\n",
       "  'A5',\n",
       "  'D5',\n",
       "  '11.1',\n",
       "  'E3',\n",
       "  'B3',\n",
       "  'G4',\n",
       "  'F#4',\n",
       "  'F#5',\n",
       "  'A5',\n",
       "  '4.9',\n",
       "  '11.1',\n",
       "  'D5',\n",
       "  'A5',\n",
       "  '11.2.6',\n",
       "  'A5',\n",
       "  'A5',\n",
       "  'B3',\n",
       "  'E3'],\n",
       " ['3.4.8.11',\n",
       "  '8.1',\n",
       "  'B4',\n",
       "  'A4',\n",
       "  '1.4.6.9',\n",
       "  '1.4.6.9',\n",
       "  '9.1.4',\n",
       "  '8.1',\n",
       "  '8.11.1.4',\n",
       "  '4.9',\n",
       "  'G#4',\n",
       "  'F#4',\n",
       "  '8.11.1.4',\n",
       "  '8.11.1.4',\n",
       "  '11.3.6',\n",
       "  '11.4'],\n",
       " ['E-4',\n",
       "  'B-4',\n",
       "  'D5',\n",
       "  'F5',\n",
       "  'C4',\n",
       "  'B-4',\n",
       "  'G4',\n",
       "  'B-4',\n",
       "  'D5',\n",
       "  'B-3',\n",
       "  'C4',\n",
       "  'G4',\n",
       "  'B-4',\n",
       "  'F4',\n",
       "  'A4',\n",
       "  'C5',\n",
       "  'B-3',\n",
       "  'F4',\n",
       "  'C5',\n",
       "  'C5',\n",
       "  'A4',\n",
       "  'F4',\n",
       "  'E-4',\n",
       "  'B-3',\n",
       "  'B-4',\n",
       "  'D5',\n",
       "  'F5',\n",
       "  'C4',\n",
       "  'B-4',\n",
       "  'G4',\n",
       "  'B-4',\n",
       "  'D5',\n",
       "  'B-3',\n",
       "  'C4',\n",
       "  'G4',\n",
       "  'B-4',\n",
       "  'F4',\n",
       "  'A4',\n",
       "  'C5',\n",
       "  'B-3',\n",
       "  'F4',\n",
       "  'C5',\n",
       "  'C5',\n",
       "  'A4',\n",
       "  'F4',\n",
       "  'B-3'],\n",
       " ['8.11', '1.6', '1.6', '1.6', '3.6', '3.6'],\n",
       " ['4.7.11',\n",
       "  'E5',\n",
       "  '4.7.9.0',\n",
       "  '0.4.7',\n",
       "  'E5',\n",
       "  '7.9.2',\n",
       "  'B5',\n",
       "  'A5',\n",
       "  'G5',\n",
       "  '4.7.11',\n",
       "  'E5',\n",
       "  '4.7.9.0',\n",
       "  '0.4.7',\n",
       "  'E5',\n",
       "  '7.9.2',\n",
       "  'B5',\n",
       "  'A5',\n",
       "  'G5'],\n",
       " ['0.3.7',\n",
       "  '2.5.9',\n",
       "  '2.5.9',\n",
       "  '0.3',\n",
       "  'G4',\n",
       "  'D3',\n",
       "  'D4',\n",
       "  'F4',\n",
       "  'A4',\n",
       "  'D3',\n",
       "  'D4',\n",
       "  'C5',\n",
       "  'A4',\n",
       "  'F4',\n",
       "  'D4',\n",
       "  '0.3.7',\n",
       "  '2.5.9',\n",
       "  '2.5.9',\n",
       "  '0.3.7',\n",
       "  '0.3.7',\n",
       "  'B-4',\n",
       "  'D3',\n",
       "  '2.5',\n",
       "  'A4',\n",
       "  'D3',\n",
       "  '2.5',\n",
       "  'F4',\n",
       "  'D4',\n",
       "  'C4'],\n",
       " ['4.7.9.11.0', 'E4', 'G4', '0.2.4.7', 'C4', 'A4', '6.9.11.2'],\n",
       " ['9.1.4',\n",
       "  'A5',\n",
       "  'G#5',\n",
       "  'F#5',\n",
       "  '8.11.3',\n",
       "  '3.6',\n",
       "  '9.1',\n",
       "  'G#5',\n",
       "  'E5',\n",
       "  'F#5',\n",
       "  'E5',\n",
       "  '11.3.6',\n",
       "  '6.8'],\n",
       " ['11.2.4.7',\n",
       "  'D5',\n",
       "  'E5',\n",
       "  'F#5',\n",
       "  'B4',\n",
       "  'D5',\n",
       "  'E5',\n",
       "  'F#5',\n",
       "  'B4',\n",
       "  'D5',\n",
       "  'E5',\n",
       "  'F#5',\n",
       "  'A5',\n",
       "  'F#5',\n",
       "  'E5',\n",
       "  'D5',\n",
       "  'F#5',\n",
       "  'D5',\n",
       "  '6.9.11.2',\n",
       "  'F#5'],\n",
       " ['5.8.10.1',\n",
       "  '3.4.6.8',\n",
       "  'B3',\n",
       "  'F#4',\n",
       "  'B3',\n",
       "  'F3',\n",
       "  '5.8.10.1',\n",
       "  '3.4.6.8',\n",
       "  'B3',\n",
       "  'F#4',\n",
       "  'B3',\n",
       "  'F3'],\n",
       " ['4.7.9.0', '2.5.7.9', '9.11.2.4', '7.9.11.2.4'],\n",
       " ['C#6',\n",
       "  'E-6',\n",
       "  'F6',\n",
       "  'B-6',\n",
       "  'G#6',\n",
       "  'F6',\n",
       "  'E-6',\n",
       "  'C#6',\n",
       "  'B-5',\n",
       "  'C#6',\n",
       "  'C#6',\n",
       "  'E-6',\n",
       "  'F6',\n",
       "  'B-6',\n",
       "  'G#6',\n",
       "  'F6',\n",
       "  'E-6',\n",
       "  'C#6',\n",
       "  'B-5',\n",
       "  'C#6'],\n",
       " ['9.11.0.2.4',\n",
       "  'E4',\n",
       "  'G4',\n",
       "  'E5',\n",
       "  '11.2.4.6.7',\n",
       "  'F#4',\n",
       "  'E5',\n",
       "  'B4',\n",
       "  'G5',\n",
       "  '9.11.0.2.4',\n",
       "  'E4',\n",
       "  'G4',\n",
       "  'E5',\n",
       "  '11.2.4.6.7',\n",
       "  'F#4',\n",
       "  'E5',\n",
       "  'G5'],\n",
       " ['7.9.11.0',\n",
       "  'B4',\n",
       "  '3.5.7.11',\n",
       "  'G3',\n",
       "  'A4',\n",
       "  '11.3.5',\n",
       "  'G4',\n",
       "  '3.5.7.11',\n",
       "  '11.0.4.7',\n",
       "  '2.7',\n",
       "  'B4',\n",
       "  '4.7.11',\n",
       "  'C4',\n",
       "  'F4',\n",
       "  'C4',\n",
       "  '8.11.0',\n",
       "  'F4',\n",
       "  'F4',\n",
       "  '5.7',\n",
       "  'E4'],\n",
       " ['6.7.11.2',\n",
       "  'B4',\n",
       "  'G4',\n",
       "  'D4',\n",
       "  '4.5.9.0',\n",
       "  'D5',\n",
       "  'G5',\n",
       "  'G4',\n",
       "  '6.7.11.2',\n",
       "  'B4',\n",
       "  'G4',\n",
       "  'D4',\n",
       "  '4.5.9.0',\n",
       "  'D5',\n",
       "  'G5',\n",
       "  'G4'],\n",
       " ['E3',\n",
       "  '9.11',\n",
       "  '2.7',\n",
       "  'B5',\n",
       "  'D6',\n",
       "  '1.2',\n",
       "  'A5',\n",
       "  '6.11',\n",
       "  'B5',\n",
       "  'F#6',\n",
       "  '2.4',\n",
       "  'A5',\n",
       "  'G5',\n",
       "  'B5',\n",
       "  'B5',\n",
       "  'D6',\n",
       "  '1.2.6',\n",
       "  'A5',\n",
       "  'B5',\n",
       "  'B5',\n",
       "  'F#6',\n",
       "  'E6'],\n",
       " ['2.5.7.10',\n",
       "  '4.7.9.0',\n",
       "  '9.10.2',\n",
       "  '4.7.9.0',\n",
       "  '4.5.9.0',\n",
       "  '2.5.7.10',\n",
       "  '4.7.9.0',\n",
       "  '9.10.2.5'],\n",
       " ['5.8.10.0.1',\n",
       "  'B-4',\n",
       "  'B-2',\n",
       "  '5.8.10.0.1',\n",
       "  'B-2',\n",
       "  'B-4',\n",
       "  '5.8.10.0.1',\n",
       "  'B-4',\n",
       "  'B-2',\n",
       "  '5.8.10.0.1',\n",
       "  'B-4',\n",
       "  'G#2',\n",
       "  'E-2',\n",
       "  '6.8.10.1',\n",
       "  'F4',\n",
       "  'F4',\n",
       "  'F#2',\n",
       "  '6.8.10.1',\n",
       "  'F4',\n",
       "  'F#2',\n",
       "  'F4',\n",
       "  '6.8.10.1.3',\n",
       "  'F4',\n",
       "  'F4',\n",
       "  'C#3',\n",
       "  '6.8.10.1',\n",
       "  'F4',\n",
       "  'F4',\n",
       "  'G#2'],\n",
       " ['F4',\n",
       "  'E-4',\n",
       "  'C#4',\n",
       "  'C#3',\n",
       "  '0.3.5.8',\n",
       "  'C#3',\n",
       "  'E3',\n",
       "  '6.10.1',\n",
       "  'F#3',\n",
       "  'E-4',\n",
       "  'C4',\n",
       "  'B-3',\n",
       "  'B-2',\n",
       "  '0.1.5.8',\n",
       "  'B-2',\n",
       "  'C#3',\n",
       "  '4.6.10',\n",
       "  'E3'],\n",
       " ['10.1.5',\n",
       "  'E-5',\n",
       "  '5.8.0',\n",
       "  '5.8.0',\n",
       "  'F5',\n",
       "  '10.1.5',\n",
       "  'F5',\n",
       "  'B-5',\n",
       "  '10.1.5',\n",
       "  'E-5',\n",
       "  '5.8.0',\n",
       "  '5.8.0',\n",
       "  'F5',\n",
       "  '10.1.5',\n",
       "  'F5',\n",
       "  'B-5'],\n",
       " ['5.9.0',\n",
       "  'E4',\n",
       "  'G6',\n",
       "  'F6',\n",
       "  'E6',\n",
       "  'A5',\n",
       "  'G5',\n",
       "  'D6',\n",
       "  'E5',\n",
       "  '11.0',\n",
       "  '4.7',\n",
       "  'E5',\n",
       "  'E4',\n",
       "  'G5',\n",
       "  'G3',\n",
       "  'C5',\n",
       "  'D5',\n",
       "  '5.9.0',\n",
       "  'E4',\n",
       "  'G6',\n",
       "  'F6',\n",
       "  'E6',\n",
       "  'A5',\n",
       "  'G5',\n",
       "  'D6',\n",
       "  'E5',\n",
       "  '11.0',\n",
       "  '4.7',\n",
       "  'E5',\n",
       "  'E4',\n",
       "  'G5',\n",
       "  'G3',\n",
       "  'C5',\n",
       "  'D5'],\n",
       " ['6.7.11.2', '9.0.2.5', '11.0.4.7', '7.10.0.3'],\n",
       " ['2.7',\n",
       "  '7.11',\n",
       "  'D4',\n",
       "  '2.7',\n",
       "  '7.11',\n",
       "  'D4',\n",
       "  'A2',\n",
       "  '4.7.11',\n",
       "  'E4',\n",
       "  '2.7',\n",
       "  'A2',\n",
       "  'E4',\n",
       "  '7.11',\n",
       "  'D4',\n",
       "  '6.11',\n",
       "  '2.6',\n",
       "  'A3',\n",
       "  '11.4',\n",
       "  '4.7',\n",
       "  'B3',\n",
       "  '11.4',\n",
       "  '4.7',\n",
       "  'B3',\n",
       "  'F#2',\n",
       "  '0.4.7',\n",
       "  'C4',\n",
       "  '11.4',\n",
       "  'F#2',\n",
       "  'C4',\n",
       "  '4.7',\n",
       "  'B3',\n",
       "  '2.7',\n",
       "  '11.2',\n",
       "  'F#3'],\n",
       " ['4.5.9.0', '9.0.2.5', '4.7.11', 'G5', 'C6', 'B4'],\n",
       " ['10.2',\n",
       "  '9.0',\n",
       "  'F4',\n",
       "  '10.2.5',\n",
       "  '10.2',\n",
       "  'F4',\n",
       "  '7.10',\n",
       "  'D5',\n",
       "  '4.5.9.0',\n",
       "  'A4',\n",
       "  '9.0.2.5',\n",
       "  '9.0.2.5',\n",
       "  'F4',\n",
       "  'G3',\n",
       "  'G3'],\n",
       " ['5.7.0',\n",
       "  'G#4',\n",
       "  'E-5',\n",
       "  '5.7.0',\n",
       "  'E-5',\n",
       "  'G#4',\n",
       "  'E-5',\n",
       "  '1.3.8',\n",
       "  'F4',\n",
       "  'C5',\n",
       "  'C6',\n",
       "  'F5',\n",
       "  'G#5',\n",
       "  '1.3.8',\n",
       "  'F4',\n",
       "  'C5',\n",
       "  'C6',\n",
       "  'F6',\n",
       "  'G#5',\n",
       "  '5.7.0',\n",
       "  'G#4',\n",
       "  'E-5',\n",
       "  '5.7.0',\n",
       "  'E-5',\n",
       "  'G#4',\n",
       "  'E-5',\n",
       "  '5.7.0',\n",
       "  'E-5',\n",
       "  'G#4',\n",
       "  'E-5',\n",
       "  '1.3.8',\n",
       "  'F4',\n",
       "  'C5',\n",
       "  'C6',\n",
       "  'F6',\n",
       "  'G#5'],\n",
       " ['5.9.0',\n",
       "  '0.4',\n",
       "  'C5',\n",
       "  'A4',\n",
       "  'G4',\n",
       "  '10.2',\n",
       "  'A4',\n",
       "  '10.2',\n",
       "  'F4',\n",
       "  'C#3',\n",
       "  '8.1',\n",
       "  '5.8.0',\n",
       "  'C#3',\n",
       "  'G#4',\n",
       "  'C#4'],\n",
       " ['0.1.3.5.8',\n",
       "  'G#5',\n",
       "  'G5',\n",
       "  'E-5',\n",
       "  'C5',\n",
       "  '0.1.3.5.8',\n",
       "  'C5',\n",
       "  'B-4',\n",
       "  'C5',\n",
       "  '3.7.10',\n",
       "  'E-5',\n",
       "  'C5',\n",
       "  '0.3.5.8',\n",
       "  'G#5',\n",
       "  '0.3.5.7.8',\n",
       "  'G5',\n",
       "  'E-5',\n",
       "  'C5',\n",
       "  '0.3.5.7.8',\n",
       "  'C5',\n",
       "  'B-4',\n",
       "  '7.8.10.0.3',\n",
       "  '7.8.10.0.3',\n",
       "  'B-5',\n",
       "  'C6'],\n",
       " ['3.6.8.11',\n",
       "  'B4',\n",
       "  '8.11.1.4',\n",
       "  '3.4.8.11',\n",
       "  'G#4',\n",
       "  '3.6.8.11',\n",
       "  'B4',\n",
       "  '8.11.1.4',\n",
       "  '3.4.8.11',\n",
       "  'E4',\n",
       "  'F#4',\n",
       "  'G#4',\n",
       "  'B4'],\n",
       " ['2.5.7.10', '4.7.9.0', '9.10.2.5', '4.7.9.0', '2.5.7.10', '4.7.9.0', '10.2'],\n",
       " ['8.9.1.4',\n",
       "  '8.9.1.4',\n",
       "  '11.1.3.6',\n",
       "  '8.11.1.4',\n",
       "  '8.11.1.4',\n",
       "  'C#3',\n",
       "  '8.11.1.4',\n",
       "  '8.11.1.4',\n",
       "  '3.6.8.11',\n",
       "  '3.6.8.11',\n",
       "  '3.6.8.11',\n",
       "  '3.6.8.11',\n",
       "  '3.6.8.11',\n",
       "  '4.8.9.11',\n",
       "  '3.6.8.11'],\n",
       " ['0.3.5.8', '10.1.3.6', '0.1.5.8', '0.3.5.8', '10.1.3.6', '0.1.5.8'],\n",
       " ['7.10.0.3',\n",
       "  'G5',\n",
       "  '2.5.7.10',\n",
       "  '2.3.7.10',\n",
       "  'D5',\n",
       "  'B-3',\n",
       "  'A4',\n",
       "  '7.10.0.3',\n",
       "  'G5',\n",
       "  '2.5.7.10',\n",
       "  '2.3.7.10',\n",
       "  'D5',\n",
       "  'B-3',\n",
       "  'A4',\n",
       "  '7.10.0.3'],\n",
       " ['0.3.5.7.8',\n",
       "  '7.10.0.2.3',\n",
       "  '4.7.9.11.0',\n",
       "  '8.0.3',\n",
       "  '7.8',\n",
       "  '7.8.10',\n",
       "  '0.3.5.7.8',\n",
       "  '7.10.0.2.3',\n",
       "  '4.7.9.11.0',\n",
       "  '7.8.10.0.3'],\n",
       " ['6.7.11.2',\n",
       "  '6.9.11.2',\n",
       "  '5.8.10.1',\n",
       "  '4.7.9.0',\n",
       "  '4.7.9.0',\n",
       "  'G3',\n",
       "  'D5',\n",
       "  'F#5',\n",
       "  'B5',\n",
       "  '9.11.2',\n",
       "  '5.8.10.1',\n",
       "  '4.7.9.0',\n",
       "  '4.7.9.0'],\n",
       " ['7.9.11.2.4',\n",
       "  'B5',\n",
       "  '11.2',\n",
       "  '7.11.2',\n",
       "  'F#6',\n",
       "  'G5',\n",
       "  '7.9.11.2.4',\n",
       "  'A5',\n",
       "  '0.2.4.7.9',\n",
       "  'B5',\n",
       "  'D6',\n",
       "  '7.9.11.2.4',\n",
       "  'B5',\n",
       "  '11.2',\n",
       "  '7.11.2',\n",
       "  'F#6',\n",
       "  'G5',\n",
       "  '7.9.11.2.4',\n",
       "  'A5',\n",
       "  '0.2.4.7.9',\n",
       "  'B5',\n",
       "  '9.0',\n",
       "  'D6',\n",
       "  'E5'],\n",
       " ['C6',\n",
       "  '5.9.0',\n",
       "  '9.0.4',\n",
       "  'D6',\n",
       "  'C6',\n",
       "  'A5',\n",
       "  'B5',\n",
       "  '0.4',\n",
       "  '11.2.4',\n",
       "  'A5',\n",
       "  'G4',\n",
       "  '11.2',\n",
       "  '11.2.4',\n",
       "  'G4',\n",
       "  '11.2',\n",
       "  '7.0',\n",
       "  'E5',\n",
       "  'E4',\n",
       "  'B4',\n",
       "  '4.5.9.0',\n",
       "  'E5',\n",
       "  'B4',\n",
       "  'A5',\n",
       "  'B5',\n",
       "  'C6',\n",
       "  'A5',\n",
       "  '11.2.4.7',\n",
       "  'F5',\n",
       "  'G5',\n",
       "  '11.2.4.7',\n",
       "  'F5',\n",
       "  '4.7',\n",
       "  '9.0.4',\n",
       "  '9.0',\n",
       "  'D5',\n",
       "  'E5',\n",
       "  'D5',\n",
       "  'C5'],\n",
       " ['9.10.2.5',\n",
       "  '9.10.2.5',\n",
       "  'C5',\n",
       "  'A4',\n",
       "  '0.4.7',\n",
       "  'A4',\n",
       "  '4.7.10.0',\n",
       "  '10.0.5',\n",
       "  '4.5.7.9.0',\n",
       "  '4.5.9.0',\n",
       "  'C5',\n",
       "  'A4',\n",
       "  '5.7.9.0.2',\n",
       "  'A4',\n",
       "  '9.0.2.5',\n",
       "  'F4',\n",
       "  'G4',\n",
       "  '2.5.9',\n",
       "  '9.10.2.5',\n",
       "  '9.10.2.5',\n",
       "  'C5',\n",
       "  'A4',\n",
       "  '4.7.10.0',\n",
       "  'A4',\n",
       "  '4.7.10.0',\n",
       "  '10.0.2.5',\n",
       "  '9.0.2.4.5',\n",
       "  '4.5.9.0',\n",
       "  'D5',\n",
       "  '0.2.3.5.7',\n",
       "  '0.3.5.7',\n",
       "  'C5',\n",
       "  '5.7.9.0'],\n",
       " ['3.6.10',\n",
       "  'C#4',\n",
       "  '5.8.0',\n",
       "  'E-4',\n",
       "  '10.1.5',\n",
       "  '10.1.5',\n",
       "  '10.1.5',\n",
       "  '10.1.5',\n",
       "  'G#5',\n",
       "  'G#4',\n",
       "  '5.8.0',\n",
       "  '5.8.0',\n",
       "  'E-5',\n",
       "  '5.8.0',\n",
       "  '5.8.0',\n",
       "  'E-4'],\n",
       " ['9.0.2.4.5',\n",
       "  '9.0.2.4.5',\n",
       "  'A4',\n",
       "  'G4',\n",
       "  '2.4.5.9',\n",
       "  'D4',\n",
       "  '4.7.9.0',\n",
       "  '4.7.9.0',\n",
       "  '2.5.7.9.10',\n",
       "  '9.0.2.5',\n",
       "  'C5'],\n",
       " ['7.10.0',\n",
       "  'G5',\n",
       "  '2.3.7.10',\n",
       "  '7.10.2',\n",
       "  'B-5',\n",
       "  'B-3',\n",
       "  '7.10.0.2.3',\n",
       "  '7.10.0',\n",
       "  'G5',\n",
       "  '2.3.7.10',\n",
       "  '7.10.2',\n",
       "  'B-5',\n",
       "  'B-3',\n",
       "  '7.10.0.2.3'],\n",
       " ['2.3.7.10',\n",
       "  'B-5',\n",
       "  'C6',\n",
       "  '5.8.0',\n",
       "  'D6',\n",
       "  '5.8.0',\n",
       "  'C6',\n",
       "  '2.5.7.10',\n",
       "  'F#5',\n",
       "  'D5',\n",
       "  'G5',\n",
       "  'E-5',\n",
       "  'F5',\n",
       "  '3',\n",
       "  '2.3.7.10',\n",
       "  'B-5',\n",
       "  'C6',\n",
       "  '5.8.0',\n",
       "  'D6',\n",
       "  '5.8.0',\n",
       "  'C6',\n",
       "  '2.5.7.10',\n",
       "  'F#5',\n",
       "  'D5',\n",
       "  'G5',\n",
       "  'E-5',\n",
       "  'F5',\n",
       "  'E-5'],\n",
       " ['B-1',\n",
       "  '10.2.5',\n",
       "  '5.9.0',\n",
       "  '5.9.0',\n",
       "  '5.9.0',\n",
       "  'C6',\n",
       "  '5.9.10.0',\n",
       "  'A5',\n",
       "  '7.9.0',\n",
       "  'G1',\n",
       "  '7.10.2',\n",
       "  '2.5.9',\n",
       "  '2.5.9',\n",
       "  '2.5.9',\n",
       "  'C6',\n",
       "  '2.5.7.9',\n",
       "  '7.9.2',\n",
       "  'D1',\n",
       "  '2.5.9',\n",
       "  '5.9.0',\n",
       "  '5.9.0',\n",
       "  '5.9.0',\n",
       "  'C6',\n",
       "  '4.9',\n",
       "  '5.9.0',\n",
       "  '7.0',\n",
       "  'F1',\n",
       "  '5.9.0',\n",
       "  '5.9.0',\n",
       "  '5.9.0',\n",
       "  '5.9.0',\n",
       "  'D6',\n",
       "  'C2',\n",
       "  '5.9.0',\n",
       "  '7.0'],\n",
       " ['10.1.3.6',\n",
       "  '6.10.11.1',\n",
       "  '6.10.11.1',\n",
       "  '10.11.3.6',\n",
       "  'F#3',\n",
       "  '1.5.8',\n",
       "  'G#5',\n",
       "  'B-5',\n",
       "  '10.1.3.6',\n",
       "  '6.10.11.1',\n",
       "  '6.10.11.1',\n",
       "  '10.11.3.6',\n",
       "  'F#3',\n",
       "  '1.5.8',\n",
       "  'G#5',\n",
       "  'B-5'],\n",
       " ['6.7.11.2',\n",
       "  'F#6',\n",
       "  'D6',\n",
       "  'B5',\n",
       "  'E5',\n",
       "  'G5',\n",
       "  '9.0.2.5',\n",
       "  'B5',\n",
       "  'A5',\n",
       "  '11.0.4.7',\n",
       "  'A5',\n",
       "  '7.10.0.3'],\n",
       " ['9.0', 'C6', 'D6', 'E6', 'G6', '2.4', 'E6', 'C6', 'D6', 'D6', 'C6', '7'],\n",
       " ['1.3.4.8',\n",
       "  'C#5',\n",
       "  'B4',\n",
       "  '10.1.3.6',\n",
       "  'B-4',\n",
       "  '3.6.10',\n",
       "  '3.4.8.10.11',\n",
       "  '10.1.3.6',\n",
       "  'G#4',\n",
       "  '8.11.1.4',\n",
       "  '1.3.4.8',\n",
       "  'C#5',\n",
       "  'B4',\n",
       "  '10.1.3.6',\n",
       "  'B-4',\n",
       "  '3.6.10',\n",
       "  '3.4.8.10.11',\n",
       "  '10.1.3.6',\n",
       "  'G#4',\n",
       "  '8.11.1.4'],\n",
       " ['9.10.2.5', '4.5.9.0', '2.3.7.10'],\n",
       " ['4.7.9.0', '11.0.4.7', '11.0.4.7', 'D3', '5.9.0', 'D3', '5.9.0'],\n",
       " ['11.0.4.7',\n",
       "  '11.2.4.7',\n",
       "  '11.0.4.7',\n",
       "  'E4',\n",
       "  '11.2.4.7',\n",
       "  'A4',\n",
       "  '11.0.4.7',\n",
       "  '11.2.4.7',\n",
       "  '11.0.4.7',\n",
       "  'E4',\n",
       "  '11.2.4.7',\n",
       "  'A4'],\n",
       " ['4.7.11',\n",
       "  'D4',\n",
       "  'E4',\n",
       "  'B3',\n",
       "  'D4',\n",
       "  '2.6.9',\n",
       "  '1.4',\n",
       "  '2.6.9',\n",
       "  '1.4',\n",
       "  '2.5.9',\n",
       "  '7.11',\n",
       "  'F3',\n",
       "  'F#3',\n",
       "  'G3',\n",
       "  'F#3',\n",
       "  '9.1.4',\n",
       "  '6.9',\n",
       "  '4.7.11',\n",
       "  'D4',\n",
       "  'E4',\n",
       "  'B3',\n",
       "  'D4',\n",
       "  '2.6.9',\n",
       "  '1.4',\n",
       "  '2.6.9',\n",
       "  '1.4',\n",
       "  '2.5.9',\n",
       "  '7.11',\n",
       "  'F3',\n",
       "  'F#3',\n",
       "  'G3',\n",
       "  'F#3',\n",
       "  '9.1.4',\n",
       "  '6.9'],\n",
       " ['2.5.7.10',\n",
       "  '7.10',\n",
       "  '2.5.7.10',\n",
       "  '2.5.7',\n",
       "  '7.10.2',\n",
       "  'F4',\n",
       "  'B-3',\n",
       "  '4.5.9.0',\n",
       "  'B-3',\n",
       "  'F4',\n",
       "  '4.5.9.0',\n",
       "  '5.9.0',\n",
       "  '0.4',\n",
       "  '4.5.9.0',\n",
       "  '4.5.9.0'],\n",
       " ['7.10.0.2.3',\n",
       "  'G4',\n",
       "  'B-4',\n",
       "  '2.3.5.7.10',\n",
       "  'E-5',\n",
       "  'G5',\n",
       "  '7.10.0.2.3',\n",
       "  'G4',\n",
       "  'B-4',\n",
       "  '2.3.5.7.10',\n",
       "  'E-5',\n",
       "  'G5'],\n",
       " ['6.10',\n",
       "  '1.6',\n",
       "  '10.1.3.6',\n",
       "  '4.8.11',\n",
       "  'G#5',\n",
       "  'F#5',\n",
       "  'E5',\n",
       "  'E-5',\n",
       "  'D5',\n",
       "  'C#5',\n",
       "  'B4',\n",
       "  'B-4',\n",
       "  '6.10',\n",
       "  '1.6',\n",
       "  '10.1.3.6',\n",
       "  '1.3.4.8',\n",
       "  '10.1.3.6',\n",
       "  'F#5',\n",
       "  'E5',\n",
       "  'E-5',\n",
       "  'D5',\n",
       "  'C#5',\n",
       "  'B4',\n",
       "  'B-4',\n",
       "  'G#4'],\n",
       " ['0.3.7',\n",
       "  'B-4',\n",
       "  'B-5',\n",
       "  'G5',\n",
       "  'E-5',\n",
       "  'C5',\n",
       "  'B-4',\n",
       "  '7.10',\n",
       "  '2.5',\n",
       "  'F5',\n",
       "  'D5',\n",
       "  'B-4',\n",
       "  'G4',\n",
       "  '0.3',\n",
       "  '7.10',\n",
       "  'B-5',\n",
       "  'G5',\n",
       "  'E-5',\n",
       "  'C5',\n",
       "  'B-4',\n",
       "  '7.10',\n",
       "  'F4',\n",
       "  'F5',\n",
       "  'D5',\n",
       "  'B-4',\n",
       "  'G4'],\n",
       " ['C#6',\n",
       "  'E-6',\n",
       "  'F6',\n",
       "  'B-6',\n",
       "  'G#6',\n",
       "  'F6',\n",
       "  'E-6',\n",
       "  'F6',\n",
       "  'C#6',\n",
       "  'E-6',\n",
       "  'F6',\n",
       "  'B-6',\n",
       "  'G#6',\n",
       "  'F6',\n",
       "  'E-6',\n",
       "  'F6'],\n",
       " ['4.5.9.0',\n",
       "  '4.5.9.0',\n",
       "  'C5',\n",
       "  'E5',\n",
       "  'D5',\n",
       "  'B4',\n",
       "  '11.0',\n",
       "  'G4',\n",
       "  '9.0.2.5',\n",
       "  'G4',\n",
       "  '9.0.2.5',\n",
       "  'G4',\n",
       "  'E4',\n",
       "  '2.7',\n",
       "  'A4',\n",
       "  'C5',\n",
       "  '4.5.9.0',\n",
       "  '4.5.9.0',\n",
       "  'C5',\n",
       "  'E5',\n",
       "  'D5',\n",
       "  'B4',\n",
       "  '7.0',\n",
       "  'E5',\n",
       "  '9.0.2.5',\n",
       "  'E5',\n",
       "  '9.0.2.5',\n",
       "  'E5',\n",
       "  'D5',\n",
       "  '2.7',\n",
       "  '2.5.9',\n",
       "  'A4',\n",
       "  'C5'],\n",
       " ['0.3.7',\n",
       "  'E-5',\n",
       "  'D5',\n",
       "  'G4',\n",
       "  '8.10.0.3.5',\n",
       "  'C4',\n",
       "  '0.3.5.8',\n",
       "  'F5',\n",
       "  'E-5',\n",
       "  'F4',\n",
       "  '10.0.2.5.7',\n",
       "  'G#4',\n",
       "  '0.3.7',\n",
       "  'E-5',\n",
       "  'D5',\n",
       "  'G4',\n",
       "  '8.10.0.3.5',\n",
       "  'G#5',\n",
       "  'G5',\n",
       "  'C4',\n",
       "  '0.3.5.8',\n",
       "  'F5',\n",
       "  'E-5',\n",
       "  'F4',\n",
       "  '10.0.2.5.7',\n",
       "  'G#5',\n",
       "  'G5',\n",
       "  'G#4',\n",
       "  'E-5'],\n",
       " ['B-2',\n",
       "  '9.10.2.5',\n",
       "  '5',\n",
       "  'B-2',\n",
       "  '4',\n",
       "  'F5',\n",
       "  'E5',\n",
       "  'D5',\n",
       "  'C5',\n",
       "  '9.0.2.5',\n",
       "  '2',\n",
       "  'D2',\n",
       "  '9',\n",
       "  'D5',\n",
       "  'E5',\n",
       "  'A2',\n",
       "  'D5',\n",
       "  'G4'],\n",
       " ['11.2.4.6.7', '1.2.4.6.9', '6.9.11.2', '1.2.4.6.9', '6.9.11.2'],\n",
       " ['7.9.10.0.3', 'F#4', '10.0.3', '6.8', 'F#4', 'F4', '5.9.10.0', '2.4.7.10'],\n",
       " ['2.5.7.10',\n",
       "  '2.5.7.10',\n",
       "  '2.5.7.10',\n",
       "  '2.5.7.10',\n",
       "  '4.5.9.0',\n",
       "  '4.5.9.0',\n",
       "  '4.5.9.0',\n",
       "  '4.5.9.0'],\n",
       " ['3.6.10',\n",
       "  '3.5.10',\n",
       "  '3.6.8.10.11',\n",
       "  '10.1.3.5.6',\n",
       "  '10.2.5',\n",
       "  '3.6.10',\n",
       "  '3.5.10',\n",
       "  '3.6.8.10.11',\n",
       "  '10.1.3.5.6',\n",
       "  '10.2.5'],\n",
       " ['10.0.1.5',\n",
       "  'B4',\n",
       "  'B-4',\n",
       "  'G#4',\n",
       "  '3.8',\n",
       "  '1.3.4.8',\n",
       "  'D5',\n",
       "  'C#5',\n",
       "  'B4',\n",
       "  '6.11',\n",
       "  '0.2.3.7',\n",
       "  'C#5',\n",
       "  'C5',\n",
       "  'B-4',\n",
       "  '5.10',\n",
       "  '11.1.2.6',\n",
       "  'C5',\n",
       "  'B4',\n",
       "  'A4',\n",
       "  '4.9'],\n",
       " ['1.2.6.9',\n",
       "  'A3',\n",
       "  'B3',\n",
       "  '11.2.6',\n",
       "  'F#3',\n",
       "  'A3',\n",
       "  'G3',\n",
       "  '11.2.4',\n",
       "  'G3',\n",
       "  '4.6.7',\n",
       "  '4.9',\n",
       "  'B3',\n",
       "  'C#4',\n",
       "  'G3',\n",
       "  'B3',\n",
       "  'C#4',\n",
       "  'D3',\n",
       "  '6.9.1',\n",
       "  'D4',\n",
       "  '6.11',\n",
       "  'A3',\n",
       "  'B3',\n",
       "  'D4',\n",
       "  'A3',\n",
       "  'B3',\n",
       "  'D4',\n",
       "  '2.4.6.7',\n",
       "  'E3',\n",
       "  'A2',\n",
       "  '11.1.4.7',\n",
       "  'A2',\n",
       "  '11.1.4.7',\n",
       "  'C#3'],\n",
       " ['9.0.4',\n",
       "  'C5',\n",
       "  'D5',\n",
       "  'C5',\n",
       "  'B4',\n",
       "  'A4',\n",
       "  '2.5.7',\n",
       "  'A4',\n",
       "  'B4',\n",
       "  'G4',\n",
       "  '2.4.5.9',\n",
       "  'G4',\n",
       "  'A4',\n",
       "  'G4',\n",
       "  'F4',\n",
       "  'E4',\n",
       "  '9.0.2.5',\n",
       "  'E4',\n",
       "  '11.2.4.7'],\n",
       " [],\n",
       " ['6.9.11.1.2',\n",
       "  '6.9.11.2',\n",
       "  '11.2.4.7',\n",
       "  '1.2.4.6.9',\n",
       "  '6.9.11.1.2',\n",
       "  '6.9.11.2',\n",
       "  '11.2.4.7',\n",
       "  '1.2.4.6.9',\n",
       "  '6.9.11.1.2'],\n",
       " ['3.4.8.11', 'B4', 'G#4', 'E-4', '3.4.8.11'],\n",
       " ['10.0',\n",
       "  '0.3.7',\n",
       "  '0.2.3.7',\n",
       "  '0.2.3.7',\n",
       "  'E-2',\n",
       "  '3.5',\n",
       "  'A3',\n",
       "  '0.3.5',\n",
       "  '0.3.5',\n",
       "  'E-2',\n",
       "  '0.2',\n",
       "  'F3',\n",
       "  'A3',\n",
       "  '0.2',\n",
       "  '0.2.5',\n",
       "  'F#2',\n",
       "  '5.7',\n",
       "  'B3',\n",
       "  '2.7',\n",
       "  '5.7',\n",
       "  'F3',\n",
       "  '3.7',\n",
       "  '3.5.7'],\n",
       " ['1.4.6.9', '1.2.4.6.9', '6.9.11.2', '6.9.11.1.2', '1.2.4.6.9'],\n",
       " ['10.0.2.5.7',\n",
       "  '9.10.2.5',\n",
       "  'C5',\n",
       "  'D5',\n",
       "  'C5',\n",
       "  'D3',\n",
       "  '7.10.0.3',\n",
       "  'C5',\n",
       "  'B-4',\n",
       "  '7.10.0.3',\n",
       "  'D3',\n",
       "  'F2',\n",
       "  '10.0.2.5.7',\n",
       "  'B-4',\n",
       "  'G2',\n",
       "  'A2',\n",
       "  '9.10.2.5',\n",
       "  'C5',\n",
       "  'D5',\n",
       "  'C5',\n",
       "  'D3',\n",
       "  '7.10.0.3',\n",
       "  'C5',\n",
       "  'B-4',\n",
       "  '7.10.0.3',\n",
       "  'D3',\n",
       "  'F2'],\n",
       " ['3.6.8.10.11',\n",
       "  '3.6.8.10.11',\n",
       "  '10.1.3.5.6',\n",
       "  '10.1.3.6',\n",
       "  '6.8.10.1.2',\n",
       "  '5.8.10.0.1'],\n",
       " ['6.9.11.2',\n",
       "  'B4',\n",
       "  'D5',\n",
       "  'E5',\n",
       "  '6.7.11.2',\n",
       "  '8.11.1.4',\n",
       "  'B4',\n",
       "  'G#4',\n",
       "  'E4',\n",
       "  '11.0.4.7']]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "notes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "41f4f64c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n"
     ]
    }
   ],
   "source": [
    "for i in range(0, 1, 1):\n",
    "    print(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ffebebdd",
   "metadata": {},
   "outputs": [],
   "source": [
    "def prepare_sequences_v2(notes, n_vocab):\n",
    "    \"\"\" Prepare the sequences used by the Neural Network \"\"\"\n",
    "    sequence_length = 8\n",
    "\n",
    "    notes = [note for note in notes if len(note) >= sequence_length]\n",
    "    \n",
    "    # get all pitch names\n",
    "    pitchnames = sorted(set(item for item in list(chain.from_iterable(notes))))\n",
    "\n",
    "     # create a dictionary to map pitches to integers\n",
    "    note_to_int = dict((note, number) for number, note in enumerate(pitchnames))\n",
    "\n",
    "    network_input = []\n",
    "    network_output = []\n",
    "\n",
    "    # create input sequences and the corresponding outputs\n",
    "    for file_note in notes:\n",
    "        for i in range(0, len(file_note) - sequence_length + 1, 1):\n",
    "            sequence_in = file_note[i:i + sequence_length]\n",
    "            network_input.append([note_to_int[char] for char in sequence_in])\n",
    "\n",
    "    n_patterns = len(network_input)\n",
    "\n",
    "    # reshape the input into a format compatible with LSTM layers\n",
    "    network_input = numpy.reshape(network_input, (n_patterns, sequence_length, 1))\n",
    "    # normalize input\n",
    "    network_input = network_input / float(n_vocab)\n",
    "\n",
    "    return (network_input, network_output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "6e4c807c",
   "metadata": {},
   "outputs": [],
   "source": [
    "n_vocab = len(set(list(chain.from_iterable(notes))))\n",
    "network_input, network_output = prepare_sequences_v2(notes, n_vocab)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "fa4ea7bf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1276, 8, 1)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "network_input.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "69d26b83",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "93"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(notes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "82badd8f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1276, 8)"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(network_input), len(network_input[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "82599f50",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.25590551],\n",
       "       [0.17322835],\n",
       "       [0.58267717],\n",
       "       [0.17322835],\n",
       "       [0.48818898],\n",
       "       [0.17322835],\n",
       "       [0.37795276],\n",
       "       [0.85433071]])"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "network_input[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "cca20719",
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_autoencoder_network():\n",
    "    \"\"\" create the structure of the neural network \"\"\"\n",
    "\n",
    "    # This is the size of our encoded representations\n",
    "    encoding_dim = 4\n",
    "\n",
    "    # This is our input image\n",
    "    input_img = keras.Input(shape=(8,))\n",
    "    # \"encoded\" is the encoded representation of the input\n",
    "    encoded = layers.Dense(encoding_dim, activation='relu')(input_img)\n",
    "    # \"decoded\" is the lossy reconstruction of the input\n",
    "    decoded = layers.Dense(8, activation='sigmoid')(encoded)\n",
    "\n",
    "    # This model maps an input to its reconstruction\n",
    "    autoencoder = keras.Model(input_img, decoded)\n",
    "    \n",
    "    return autoencoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "4e8f81e3",
   "metadata": {},
   "outputs": [],
   "source": [
    "encoder = create_autoencoder_network()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "4a5dbc99",
   "metadata": {},
   "outputs": [],
   "source": [
    "encoder.compile(optimizer='adam', loss='binary_crossentropy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "cd8257ae",
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "40/40 [==============================] - 0s 914us/step - loss: 0.6976\n",
      "Epoch 2/100\n",
      "40/40 [==============================] - 0s 634us/step - loss: 0.6821\n",
      "Epoch 3/100\n",
      "40/40 [==============================] - 0s 544us/step - loss: 0.6713\n",
      "Epoch 4/100\n",
      "40/40 [==============================] - 0s 547us/step - loss: 0.6615\n",
      "Epoch 5/100\n",
      "40/40 [==============================] - 0s 528us/step - loss: 0.6521\n",
      "Epoch 6/100\n",
      "40/40 [==============================] - 0s 522us/step - loss: 0.6429\n",
      "Epoch 7/100\n",
      "40/40 [==============================] - 0s 524us/step - loss: 0.6334\n",
      "Epoch 8/100\n",
      "40/40 [==============================] - 0s 553us/step - loss: 0.6240\n",
      "Epoch 9/100\n",
      "40/40 [==============================] - 0s 549us/step - loss: 0.6155\n",
      "Epoch 10/100\n",
      "40/40 [==============================] - 0s 577us/step - loss: 0.6086\n",
      "Epoch 11/100\n",
      "40/40 [==============================] - 0s 542us/step - loss: 0.6034\n",
      "Epoch 12/100\n",
      "40/40 [==============================] - 0s 535us/step - loss: 0.5994\n",
      "Epoch 13/100\n",
      "40/40 [==============================] - 0s 547us/step - loss: 0.5963\n",
      "Epoch 14/100\n",
      "40/40 [==============================] - 0s 527us/step - loss: 0.5939\n",
      "Epoch 15/100\n",
      "40/40 [==============================] - 0s 534us/step - loss: 0.5918\n",
      "Epoch 16/100\n",
      "40/40 [==============================] - 0s 542us/step - loss: 0.5900\n",
      "Epoch 17/100\n",
      "40/40 [==============================] - 0s 542us/step - loss: 0.5883\n",
      "Epoch 18/100\n",
      "40/40 [==============================] - 0s 533us/step - loss: 0.5867\n",
      "Epoch 19/100\n",
      "40/40 [==============================] - 0s 540us/step - loss: 0.5851\n",
      "Epoch 20/100\n",
      "40/40 [==============================] - 0s 533us/step - loss: 0.5834\n",
      "Epoch 21/100\n",
      "40/40 [==============================] - 0s 536us/step - loss: 0.5817\n",
      "Epoch 22/100\n",
      "40/40 [==============================] - 0s 535us/step - loss: 0.5800\n",
      "Epoch 23/100\n",
      "40/40 [==============================] - 0s 527us/step - loss: 0.5781\n",
      "Epoch 24/100\n",
      "40/40 [==============================] - 0s 536us/step - loss: 0.5761\n",
      "Epoch 25/100\n",
      "40/40 [==============================] - 0s 550us/step - loss: 0.5739\n",
      "Epoch 26/100\n",
      "40/40 [==============================] - 0s 546us/step - loss: 0.5718\n",
      "Epoch 27/100\n",
      "40/40 [==============================] - 0s 556us/step - loss: 0.5696\n",
      "Epoch 28/100\n",
      "40/40 [==============================] - 0s 560us/step - loss: 0.5675\n",
      "Epoch 29/100\n",
      "40/40 [==============================] - 0s 592us/step - loss: 0.5654\n",
      "Epoch 30/100\n",
      "40/40 [==============================] - 0s 521us/step - loss: 0.5635\n",
      "Epoch 31/100\n",
      "40/40 [==============================] - 0s 564us/step - loss: 0.5616\n",
      "Epoch 32/100\n",
      "40/40 [==============================] - 0s 520us/step - loss: 0.5598\n",
      "Epoch 33/100\n",
      "40/40 [==============================] - 0s 508us/step - loss: 0.5581\n",
      "Epoch 34/100\n",
      "40/40 [==============================] - 0s 573us/step - loss: 0.5565\n",
      "Epoch 35/100\n",
      "40/40 [==============================] - 0s 538us/step - loss: 0.5550\n",
      "Epoch 36/100\n",
      "40/40 [==============================] - 0s 524us/step - loss: 0.5536\n",
      "Epoch 37/100\n",
      "40/40 [==============================] - 0s 570us/step - loss: 0.5523\n",
      "Epoch 38/100\n",
      "40/40 [==============================] - 0s 559us/step - loss: 0.5511\n",
      "Epoch 39/100\n",
      "40/40 [==============================] - 0s 528us/step - loss: 0.5500\n",
      "Epoch 40/100\n",
      "40/40 [==============================] - 0s 543us/step - loss: 0.5489\n",
      "Epoch 41/100\n",
      "40/40 [==============================] - 0s 540us/step - loss: 0.5479\n",
      "Epoch 42/100\n",
      "40/40 [==============================] - 0s 546us/step - loss: 0.5470\n",
      "Epoch 43/100\n",
      "40/40 [==============================] - 0s 533us/step - loss: 0.5462\n",
      "Epoch 44/100\n",
      "40/40 [==============================] - 0s 539us/step - loss: 0.5454\n",
      "Epoch 45/100\n",
      "40/40 [==============================] - 0s 533us/step - loss: 0.5447\n",
      "Epoch 46/100\n",
      "40/40 [==============================] - 0s 538us/step - loss: 0.5440\n",
      "Epoch 47/100\n",
      "40/40 [==============================] - 0s 541us/step - loss: 0.5434\n",
      "Epoch 48/100\n",
      "40/40 [==============================] - 0s 552us/step - loss: 0.5428\n",
      "Epoch 49/100\n",
      "40/40 [==============================] - 0s 551us/step - loss: 0.5422\n",
      "Epoch 50/100\n",
      "40/40 [==============================] - 0s 544us/step - loss: 0.5417\n",
      "Epoch 51/100\n",
      "40/40 [==============================] - 0s 547us/step - loss: 0.5412\n",
      "Epoch 52/100\n",
      "40/40 [==============================] - 0s 531us/step - loss: 0.5407\n",
      "Epoch 53/100\n",
      "40/40 [==============================] - 0s 528us/step - loss: 0.5403\n",
      "Epoch 54/100\n",
      "40/40 [==============================] - 0s 546us/step - loss: 0.5399\n",
      "Epoch 55/100\n",
      "40/40 [==============================] - 0s 541us/step - loss: 0.5394\n",
      "Epoch 56/100\n",
      "40/40 [==============================] - 0s 679us/step - loss: 0.5390\n",
      "Epoch 57/100\n",
      "40/40 [==============================] - 0s 533us/step - loss: 0.5387\n",
      "Epoch 58/100\n",
      "40/40 [==============================] - 0s 578us/step - loss: 0.5383\n",
      "Epoch 59/100\n",
      "40/40 [==============================] - 0s 525us/step - loss: 0.5379\n",
      "Epoch 60/100\n",
      "40/40 [==============================] - 0s 535us/step - loss: 0.5376\n",
      "Epoch 61/100\n",
      "40/40 [==============================] - 0s 533us/step - loss: 0.5373\n",
      "Epoch 62/100\n",
      "40/40 [==============================] - 0s 502us/step - loss: 0.5370\n",
      "Epoch 63/100\n",
      "40/40 [==============================] - 0s 548us/step - loss: 0.5366\n",
      "Epoch 64/100\n",
      "40/40 [==============================] - 0s 525us/step - loss: 0.5363\n",
      "Epoch 65/100\n",
      "40/40 [==============================] - 0s 560us/step - loss: 0.5360\n",
      "Epoch 66/100\n",
      "40/40 [==============================] - 0s 539us/step - loss: 0.5357\n",
      "Epoch 67/100\n",
      "40/40 [==============================] - 0s 521us/step - loss: 0.5355\n",
      "Epoch 68/100\n",
      "40/40 [==============================] - 0s 555us/step - loss: 0.5352\n",
      "Epoch 69/100\n",
      "40/40 [==============================] - 0s 583us/step - loss: 0.5350\n",
      "Epoch 70/100\n",
      "40/40 [==============================] - 0s 570us/step - loss: 0.5347\n",
      "Epoch 71/100\n",
      "40/40 [==============================] - 0s 528us/step - loss: 0.5345\n",
      "Epoch 72/100\n",
      "40/40 [==============================] - 0s 568us/step - loss: 0.5342\n",
      "Epoch 73/100\n",
      "40/40 [==============================] - 0s 566us/step - loss: 0.5340\n",
      "Epoch 74/100\n",
      "40/40 [==============================] - 0s 640us/step - loss: 0.5338\n",
      "Epoch 75/100\n",
      "40/40 [==============================] - 0s 556us/step - loss: 0.5336\n",
      "Epoch 76/100\n",
      "40/40 [==============================] - 0s 522us/step - loss: 0.5334\n",
      "Epoch 77/100\n",
      "40/40 [==============================] - 0s 584us/step - loss: 0.5332\n",
      "Epoch 78/100\n",
      "40/40 [==============================] - 0s 548us/step - loss: 0.5330\n",
      "Epoch 79/100\n",
      "40/40 [==============================] - 0s 536us/step - loss: 0.5328\n",
      "Epoch 80/100\n",
      "40/40 [==============================] - 0s 541us/step - loss: 0.5326\n",
      "Epoch 81/100\n",
      "40/40 [==============================] - 0s 573us/step - loss: 0.5325\n",
      "Epoch 82/100\n",
      "40/40 [==============================] - 0s 555us/step - loss: 0.5323\n",
      "Epoch 83/100\n",
      "40/40 [==============================] - 0s 561us/step - loss: 0.5322\n",
      "Epoch 84/100\n",
      "40/40 [==============================] - 0s 559us/step - loss: 0.5320\n",
      "Epoch 85/100\n",
      "40/40 [==============================] - 0s 576us/step - loss: 0.5319\n",
      "Epoch 86/100\n",
      "40/40 [==============================] - 0s 546us/step - loss: 0.5317\n",
      "Epoch 87/100\n",
      "40/40 [==============================] - 0s 520us/step - loss: 0.5316\n",
      "Epoch 88/100\n",
      "40/40 [==============================] - 0s 574us/step - loss: 0.5315\n",
      "Epoch 89/100\n",
      "40/40 [==============================] - 0s 527us/step - loss: 0.5313\n",
      "Epoch 90/100\n",
      "40/40 [==============================] - 0s 552us/step - loss: 0.5312\n",
      "Epoch 91/100\n",
      "40/40 [==============================] - 0s 574us/step - loss: 0.5311\n",
      "Epoch 92/100\n",
      "40/40 [==============================] - 0s 566us/step - loss: 0.5310\n",
      "Epoch 93/100\n",
      "40/40 [==============================] - 0s 553us/step - loss: 0.5308\n",
      "Epoch 94/100\n",
      "40/40 [==============================] - 0s 585us/step - loss: 0.5307\n",
      "Epoch 95/100\n",
      "40/40 [==============================] - 0s 563us/step - loss: 0.5306\n",
      "Epoch 96/100\n",
      "40/40 [==============================] - 0s 561us/step - loss: 0.5305\n",
      "Epoch 97/100\n",
      "40/40 [==============================] - 0s 574us/step - loss: 0.5304\n",
      "Epoch 98/100\n",
      "40/40 [==============================] - 0s 539us/step - loss: 0.5303\n",
      "Epoch 99/100\n",
      "40/40 [==============================] - 0s 525us/step - loss: 0.5303\n",
      "Epoch 100/100\n",
      "40/40 [==============================] - 0s 574us/step - loss: 0.5301\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x7f8ae28778b0>"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "encoder.fit(network_input, network_input,\n",
    "            epochs=100,\n",
    "            batch_size=32,\n",
    "            shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "03314c3e",
   "metadata": {},
   "outputs": [],
   "source": [
    "note"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a55e8133",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "22a984f3",
   "metadata": {},
   "outputs": [],
   "source": [
    "def train(model, network_input, network_output):\n",
    "    \"\"\" train the neural network \"\"\"\n",
    "    filepath = \"updated-lofi-hip-hop-weights-improvement-1{epoch:02d}-{loss:.4f}.hdf5\"\n",
    "    checkpoint = ModelCheckpoint(\n",
    "        filepath,\n",
    "        monitor='loss',\n",
    "        verbose=0,\n",
    "        save_best_only=True,\n",
    "        mode='min'\n",
    "    )\n",
    "    callbacks_list = [checkpoint]\n",
    "\n",
    "    model.fit(network_input, network_output, epochs=100, batch_size=64, callbacks=callbacks_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "3d49ffa4",
   "metadata": {},
   "outputs": [],
   "source": [
    "def train_network():\n",
    "    \"\"\" Train a Neural Network to generate music \"\"\"\n",
    "    notes = get_notes()\n",
    "\n",
    "    # get amount of pitch names\n",
    "    n_vocab = len(set(notes))\n",
    "\n",
    "    network_input, network_output = prepare_sequences(notes, n_vocab)\n",
    "    model = create_network(network_input, n_vocab)\n",
    "\n",
    "    train(model, network_input, network_output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "d7f99430",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = create_network(network_input, n_vocab)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "3827a5e6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/30\n",
      "30/30 [==============================] - 32s 958ms/step - loss: 5.9099\n",
      "Epoch 2/30\n",
      "30/30 [==============================] - 28s 921ms/step - loss: 5.6505\n",
      "Epoch 3/30\n",
      "30/30 [==============================] - 31s 1s/step - loss: 5.5291\n",
      "Epoch 4/30\n",
      "30/30 [==============================] - 38s 1s/step - loss: 5.5055\n",
      "Epoch 5/30\n",
      "30/30 [==============================] - 30s 1s/step - loss: 5.3513\n",
      "Epoch 6/30\n",
      "30/30 [==============================] - 31s 1s/step - loss: 5.3021\n",
      "Epoch 7/30\n",
      "30/30 [==============================] - 32s 1s/step - loss: 5.2602\n",
      "Epoch 8/30\n",
      "30/30 [==============================] - 31s 1s/step - loss: 5.1564\n",
      "Epoch 9/30\n",
      "30/30 [==============================] - 31s 1s/step - loss: 5.1468\n",
      "Epoch 10/30\n",
      "30/30 [==============================] - 31s 1s/step - loss: 5.0318\n",
      "Epoch 11/30\n",
      "30/30 [==============================] - 31s 1s/step - loss: 5.0592\n",
      "Epoch 12/30\n",
      "30/30 [==============================] - 32s 1s/step - loss: 5.0199\n",
      "Epoch 13/30\n",
      "30/30 [==============================] - 32s 1s/step - loss: 4.9358\n",
      "Epoch 14/30\n",
      "30/30 [==============================] - 31s 1s/step - loss: 4.8521\n",
      "Epoch 15/30\n",
      "30/30 [==============================] - 31s 1s/step - loss: 4.7969\n",
      "Epoch 16/30\n",
      "30/30 [==============================] - 32s 1s/step - loss: 4.7655\n",
      "Epoch 17/30\n",
      "30/30 [==============================] - 32s 1s/step - loss: 4.7060\n",
      "Epoch 18/30\n",
      "30/30 [==============================] - 32s 1s/step - loss: 4.7035\n",
      "Epoch 19/30\n",
      "30/30 [==============================] - 31s 1s/step - loss: 4.6614\n",
      "Epoch 20/30\n",
      "30/30 [==============================] - 32s 1s/step - loss: 4.5976\n",
      "Epoch 21/30\n",
      "30/30 [==============================] - 33s 1s/step - loss: 4.5844\n",
      "Epoch 22/30\n",
      "30/30 [==============================] - 32s 1s/step - loss: 4.5743\n",
      "Epoch 23/30\n",
      "30/30 [==============================] - 31s 1s/step - loss: 4.5312\n",
      "Epoch 24/30\n",
      "30/30 [==============================] - 31s 1s/step - loss: 4.4956\n",
      "Epoch 25/30\n",
      "30/30 [==============================] - 32s 1s/step - loss: 4.4066\n",
      "Epoch 26/30\n",
      "30/30 [==============================] - 31s 1s/step - loss: 4.3910\n",
      "Epoch 27/30\n",
      "30/30 [==============================] - 31s 1s/step - loss: 4.3592\n",
      "Epoch 28/30\n",
      "30/30 [==============================] - 32s 1s/step - loss: 4.3169\n",
      "Epoch 29/30\n",
      "30/30 [==============================] - 33s 1s/step - loss: 4.3104\n",
      "Epoch 30/30\n",
      "30/30 [==============================] - 34s 1s/step - loss: 4.2807\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x7fdfa3e594f0>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(network_input, network_output, epochs=30, batch_size=64, callbacks=callbacks_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "0a0636b0",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:Found untraced functions such as lstm_cell_2_layer_call_and_return_conditional_losses, lstm_cell_2_layer_call_fn, lstm_cell_2_layer_call_fn, lstm_cell_2_layer_call_and_return_conditional_losses, lstm_cell_2_layer_call_and_return_conditional_losses while saving (showing 5 of 5). These functions will not be directly callable after loading.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: piano_model/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: piano_model/assets\n"
     ]
    }
   ],
   "source": [
    "model.save(PIANO_GENERATOR_MODEL_PATH)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "c5c68965",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.save_weights(PIANO_GENERATOR_MODEL_WEIGHTS_PATH + '_02')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4a99e674",
   "metadata": {},
   "source": [
    "## generate piano roll midi file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "4fabe4c2",
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_learned_network(network_input, n_vocab, weights_path):\n",
    "    \"\"\" create the structure of the neural network \"\"\"\n",
    "    model = Sequential()\n",
    "    model.add(LSTM(\n",
    "        512,\n",
    "        input_shape=(network_input.shape[1], network_input.shape[2]),\n",
    "        recurrent_dropout=0.3,\n",
    "        return_sequences=True\n",
    "    ))\n",
    "    model.add(LSTM(512, return_sequences=True, recurrent_dropout=0.3,))\n",
    "    model.add(LSTM(512))\n",
    "    model.add(BatchNorm())\n",
    "    model.add(Dropout(0.3))\n",
    "    model.add(Dense(256))\n",
    "    model.add(Activation('relu'))\n",
    "    model.add(BatchNorm())\n",
    "    model.add(Dropout(0.3))\n",
    "    model.add(Dense(n_vocab))\n",
    "    model.add(Activation('softmax'))\n",
    "    model.compile(loss='categorical_crossentropy', optimizer='adam')\n",
    "\n",
    "    # Load the weights to each node\n",
    "    model.load_weights(weights_path)\n",
    "\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "5c76586b",
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate_notes(model, network_input, pitchnames, n_vocab):\n",
    "    \"\"\" Generate notes from the neural network based on a sequence of notes \"\"\"\n",
    "    # pick a random sequence from the input as a starting point for the prediction\n",
    "    start = numpy.random.randint(0, len(network_input)-1)\n",
    "\n",
    "    int_to_note = dict((number, note) for number, note in enumerate(pitchnames))\n",
    "\n",
    "    pattern = network_input[start]\n",
    "    prediction_output = []\n",
    "\n",
    "    # generate 500 notes\n",
    "    for note_index in range(200):\n",
    "        prediction_input = numpy.reshape(pattern, (1, len(pattern), 1))\n",
    "        prediction_input = prediction_input / float(n_vocab)\n",
    "\n",
    "        prediction = model.predict(prediction_input, verbose=0)\n",
    "\n",
    "        index = numpy.argmax(prediction)\n",
    "        result = int_to_note[index]\n",
    "        prediction_output.append(result)\n",
    "\n",
    "        pattern.append(index)\n",
    "        pattern = pattern[1:len(pattern)]\n",
    "\n",
    "    return prediction_output"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "dd0d5b83",
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_midi(prediction_output, \n",
    "                output_path):\n",
    "    \"\"\" convert the output from the prediction to notes and create a midi file\n",
    "        from the notes \"\"\"\n",
    "    offset = 0\n",
    "    output_notes = []\n",
    "\n",
    "    # create note and chord objects based on the values generated by the model\n",
    "    for pattern in prediction_output:\n",
    "        # pattern is a chord\n",
    "        if ('.' in pattern) or pattern.isdigit():\n",
    "            notes_in_chord = pattern.split('.')\n",
    "            notes = []\n",
    "            for current_note in notes_in_chord:\n",
    "                new_note = note.Note(int(current_note))\n",
    "                new_note.storedInstrument = instrument.Piano()\n",
    "                notes.append(new_note)\n",
    "            new_chord = chord.Chord(notes)\n",
    "            new_chord.offset = offset\n",
    "            output_notes.append(new_chord)\n",
    "        # pattern is a note\n",
    "        else:\n",
    "            new_note = note.Note(pattern)\n",
    "            new_note.offset = offset\n",
    "            new_note.storedInstrument = instrument.Piano()\n",
    "            output_notes.append(new_note)\n",
    "\n",
    "        # increase offset each iteration so that notes do not stack\n",
    "#         offset += 0.5\n",
    "        offset += 1\n",
    "\n",
    "    midi_stream = stream.Stream(output_notes)\n",
    "\n",
    "    midi_stream.write('midi', fp=output_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "08a3136b",
   "metadata": {},
   "outputs": [],
   "source": [
    "def prepare_sequences(notes, pitchnames, n_vocab):\n",
    "    \"\"\" Prepare the sequences used by the Neural Network \"\"\"\n",
    "    # map between notes and integers and back\n",
    "    note_to_int = dict((note, number) for number, note in enumerate(pitchnames))\n",
    "\n",
    "    sequence_length = 32\n",
    "    network_input = []\n",
    "    output = []\n",
    "    for i in range(0, len(notes) - sequence_length, 1):\n",
    "        sequence_in = notes[i:i + sequence_length]\n",
    "        sequence_out = notes[i + sequence_length]\n",
    "        network_input.append([note_to_int[char] for char in sequence_in])\n",
    "        output.append(note_to_int[sequence_out])\n",
    "\n",
    "    n_patterns = len(network_input)\n",
    "\n",
    "    # reshape the input into a format compatible with LSTM layers\n",
    "    normalized_input = numpy.reshape(network_input, (n_patterns, sequence_length, 1))\n",
    "    # normalize input\n",
    "    normalized_input = normalized_input / float(n_vocab)\n",
    "\n",
    "    return (network_input, normalized_input)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "b7207594",
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate(weights_path, output_path):\n",
    "    \"\"\" Generate a piano midi file \"\"\"\n",
    "    #load the notes used to train the model\n",
    "    with open('data/notes', 'rb') as filepath:\n",
    "        notes = pickle.load(filepath)\n",
    "\n",
    "    # Get all pitch names\n",
    "    pitchnames = sorted(set(item for item in notes))\n",
    "    # Get all pitch names\n",
    "    n_vocab = len(set(notes))\n",
    "\n",
    "    network_input, normalized_input = prepare_sequences(notes, pitchnames, n_vocab)\n",
    "    model = load_learned_network(normalized_input, n_vocab, weights_path)\n",
    "    prediction_output = generate_notes(model, network_input, pitchnames, n_vocab)\n",
    "    create_midi(prediction_output, output_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "20eb5098",
   "metadata": {},
   "outputs": [],
   "source": [
    "generate(weights_path=PIANO_GENERATOR_MODEL_WEIGHTS_PATH+'_01', output_path='lofi_output7.mid')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "e7a9ddbb",
   "metadata": {},
   "outputs": [],
   "source": [
    "_ = os.system('fluidsynth -ni Touhou.sf2 lofi_output7.mid -F piano_generated_07.wav -r 44100')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "06e463b9",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "606b0fe1",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c0c8bd76",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9220228d",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}