data_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pretty_midi\n",
    "import pypianoroll\n",
    "import mido\n",
    "\n",
    "from tiv import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [],
   "source": [
    "bar8_cp_np = np.load('data/bar8_cp_np.npy')\n",
    "bar8_cp_wq_np = np.load('data/bar8_cp_with_quality_np.npy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "complexity = bar8_cp_wq_np.sum(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "diff = (bar8_cp_np - bar8_cp_wq_np).sum(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "45143.0"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "diff.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "cd = diff.reshape(-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([168225.,      0.,      0.,  39322.,      0.,      0.,   1394.,\n",
       "             0.,      0.,   1011.]),\n",
       " array([0. , 0.3, 0.6, 0.9, 1.2, 1.5, 1.8, 2.1, 2.4, 2.7, 3. ]),\n",
       " <a list of 10 Patch objects>)"
      ]
     },
     "execution_count": 62,
     "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.hist(cd)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [],
   "source": [
    "x0 = bar8_cp_np[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "tiv_x0 = TIV.from_pcp(x0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "tiv_x0.plot_tiv()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([5.49038106e-01, 2.22044605e-16, 5.75000000e+00, 7.50000000e+00,\n",
       "       9.90368418e+00, 0.00000000e+00])"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tiv_x0.mags()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0.52359878,  0.        ,  1.57079633, -1.04719755, -0.52359878,\n",
       "        0.        ])"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tiv_x0.phases()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_tiv_polar(chroma):\n",
    "    '''\n",
    "    input: 12 dims chroma\n",
    "    return: 12 dims TIV in polar coordinate, first 6 dims--mags, last 6 dims-- phases\n",
    "    '''\n",
    "    tiv = TIV.from_pcp(chroma)\n",
    "    tiv_polar = np.concatenate((tiv.mags(), tiv.phases()))\n",
    "    \n",
    "    return tiv_polar\n",
    "\n",
    "def get_tiv_polar_list(chroma_list):\n",
    "    tiv_list = chroma_list.copy()\n",
    "    for i in range(chroma_list.shape[0]):\n",
    "        tiv_list[i][:] = get_tiv_polar(chroma_list[i][:])\n",
    "    return tiv_list\n",
    "\n",
    "def chroma2tiv_np(chroma_chunk):\n",
    "    tiv_chunk = chroma_chunk.copy()\n",
    "    for i in range(chroma_chunk.shape[0]):\n",
    "        tiv_chunk[i] = get_tiv_polar_list(chroma_chunk[i])\n",
    "    return tiv_chunk"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [],
   "source": [
    "bar8_cp_tiv_np = chroma2tiv_np(bar8_cp_np)\n",
    "bar8_cp_wq_tiv_np = chroma2tiv_np(bar8_cp_wq_np)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.save('bar8_cp_tiv_np.npy', bar8_cp_tiv_np)\n",
    "np.save('bar8_cp_wq_tiv_np.npy', bar8_cp_wq_tiv_np)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 5.49038106e-01,  2.22044605e-16,  5.75000000e+00,  7.50000000e+00,\n",
       "        9.90368418e+00,  0.00000000e+00,  5.23598776e-01,  0.00000000e+00,\n",
       "        1.57079633e+00, -1.04719755e+00, -5.23598776e-01,  0.00000000e+00])"
      ]
     },
     "execution_count": 115,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "get_tiv_polar(x0[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1., 0., 0., 1., 0., 0., 0., 1., 0., 0., 1., 0.])"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x0[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}