Probabilistic_PCA_inferpy.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "scrolled": true
   },
   "source": [
    "#### Probabilistic PCA\n",
    "In this notebook we try to do dimensionality reduction using probabilistic pca \\\n",
    "Estimating number of components jointly with lower dimension representation of data is a challenging task, here using hierarchical bayesian approach and to infer number of components we use __[dirichlet process](https://en.wikipedia.org/wiki/Dirichlet_process)__ on weights. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import inferpy as inf\n",
    "import tensorflow as tf\n",
    "import numpy as np\n",
    "from matplotlib import pyplot as plt\n",
    "import h5py as h5\n",
    "import numpy as np\n",
    "from matplotlib import animation\n",
    "from IPython.display import HTML"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x864 with 25 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(60000, 784)\n",
      "data contains non-positive elements\n",
      "you sould use PCA based analysis\n"
     ]
    }
   ],
   "source": [
    "# input info \n",
    "\n",
    "# dimension of the data\n",
    "data_dim = 28 *28\n",
    "\n",
    "# dimension of hidden layer\n",
    "hidden_dim = 30\n",
    "\n",
    "# number of epochs for training\n",
    "nr_epochs = 60000\n",
    "\n",
    "# number of input image in each batch during training\n",
    "nr_batch = 200\n",
    "\n",
    "# digits in mnist\n",
    "DIG = [0, 1, 3, 2, 4, 5, 6, 8, 9, 7]\n",
    "\n",
    "# number of mnist samples\n",
    "mnist_nr = 60000\n",
    "\n",
    "# data selection custom or MNIST\n",
    "MNIST = 1\n",
    "Cust = not MNIST\n",
    "\n",
    "\n",
    "# reading, reshaping input data\n",
    "\n",
    "if MNIST==1:\n",
    "    from inferpy.data import mnist\n",
    "    N = mnist_nr\n",
    "    # load the data\n",
    "    (x_train, y_train), _ = mnist.load_data(num_instances=N, digits=DIG)\n",
    "    mnist.plot_digits(x_train, grid=[5,5])\n",
    "    print(x_train.shape)\n",
    "\n",
    "if Cust:\n",
    "    # loading custom data\n",
    "    \n",
    "    File = h5.File('data address in local machine', 'r')\n",
    "    list(File.keys())\n",
    "    data = np.array(File['key in custom data'])\n",
    "    print(data.shape)\n",
    "    x_train = data.reshape(-1,data_dim)\n",
    "    print(x_train.shape)\n",
    "    N = x_train.shape[0]\n",
    "    \n",
    "    \n",
    "# Testing data\n",
    "if x_train.min() <= 0:\n",
    "    print('data contains non-positive elements')\n",
    "    print('you sould use PCA based analysis')\n",
    "else:\n",
    "    print('data is all positive')\n",
    "    print('you can use NMF analysis')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Defining model & inference model\n",
    "\n",
    "For likelihood (data) model we take multivariate normal with low rank model from inferpy package  \n",
    "Hidden variable (z) and transfer weights (w) are modeled with normal distribution   \n",
    "For weights we use dirichlet process with hierarchical priors beta and gamma distributions respectively"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "model function is defined\n"
     ]
    }
   ],
   "source": [
    "# Define a model\n",
    "#tf.reset_default_graph() # reset the graph before defining model\n",
    "\n",
    "# helper function to do dirichlet process prior on weights\n",
    "def Stick_Breaking(betas):\n",
    "    portion_remaining = tf.concat([[1], tf.cumprod(1 - betas)[:-1]], axis = 0)\n",
    "    return betas*portion_remaining\n",
    "\n",
    "\n",
    "# define model \n",
    "@inf.probmodel\n",
    "def PCA(k, d):\n",
    "    alpha = inf.Gamma(25,tf.ones([k]), name = 'alpha')\n",
    "    beta = inf.Beta(1., alpha, name = 'beta')\n",
    "    weight_ = inf.Deterministic(Stick_Breaking(beta), name = 'weight_')\n",
    "    w = inf.Normal(loc = tf.zeros([k,d]), scale = 1, name = 'w')\n",
    "    with inf.datamodel():\n",
    "        z = inf.Normal(loc = tf.ones(k), scale = 1, name = 'z')\n",
    "        x = inf.MultivariateNormalDiagPlusLowRank(\n",
    "            loc = tf.matmul(z, tf.expand_dims(weight_, axis = 1) * w), name = 'x')\n",
    "        \n",
    "\n",
    "@inf.probmodel        \n",
    "def qmodel(k, d):\n",
    "    \n",
    "    Param1 = tf.math.softplus(inf.Parameter(tf.ones([k]),\n",
    "                                              name = 'Param1'))\n",
    "    #Param2 = tf.math.softplus(inf.Parameter(tf.ones([k]),\n",
    "     #                                         name = 'Param2'))\n",
    "    \n",
    "    qalpha = inf.Gamma(50, Param1, name = 'alpha')\n",
    "    qbeta = inf.Beta(1., qalpha, name = 'beta')\n",
    "    qweight_ = inf.Deterministic(Stick_Breaking(qbeta), name = 'weight_')\n",
    "    \n",
    "    qw_loc = inf.Parameter(tf.ones([k, d]), name = 'qw_loc')\n",
    "    qw_scale = tf.math.softplus(inf.Parameter(tf.ones([k, d]),\n",
    "                                              name = 'qw_scale'))\n",
    "    qw = inf.Normal(loc = qw_loc, scale = qw_scale, name = 'w')\n",
    "    \n",
    "    with inf.datamodel():\n",
    "        qz_loc = inf.Parameter(tf.ones(k), name = 'qz_loc')\n",
    "        qz_scale = tf.math.softplus(inf.Parameter(tf.ones(k),\n",
    "                                                  name = 'qz_scale'))\n",
    "        qz = inf.Normal(loc = qz_loc, scale = qz_scale, name = 'z')\n",
    "        \n",
    "\n",
    "    \n",
    "print('model function is defined')   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_ = PCA(hidden_dim, data_dim)\n",
    "qmodel_ = qmodel(hidden_dim, data_dim)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model inference\n",
    "To do model inference on posterior we use variational inference technique which is provided in inferpy package\n",
    "Adam optimization method is used as optimizer with learning rate of <0.001"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " 0 epochs\t 102906437632.0....................\n",
      " 200 epochs\t 102890504192.0....................\n",
      " 400 epochs\t 102875586560.0....................\n",
      " 600 epochs\t 102854737920.0....................\n",
      " 800 epochs\t 102809501696.0....................\n",
      " 1000 epochs\t 102770950144.0....................\n",
      " 1200 epochs\t 102732922880.0....................\n",
      " 1400 epochs\t 102664355840.0....................\n",
      " 1600 epochs\t 102581903360.0....................\n",
      " 1800 epochs\t 102497361920.0....................\n",
      " 2000 epochs\t 102476120064.0....................\n",
      " 2200 epochs\t 102352691200.0....................\n",
      " 2400 epochs\t 102355582976.0....................\n",
      " 2600 epochs\t 102017679360.0....................\n",
      " 2800 epochs\t 102092480512.0....................\n",
      " 3000 epochs\t 101901549568.0....................\n",
      " 3200 epochs\t 101884551168.0....................\n",
      " 3400 epochs\t 101644607488.0....................\n",
      " 3600 epochs\t 101555585024.0....................\n",
      " 3800 epochs\t 101446418432.0....................\n",
      " 4000 epochs\t 101275369472.0....................\n",
      " 4200 epochs\t 101092261888.0....................\n",
      " 4400 epochs\t 101056937984.0....................\n",
      " 4600 epochs\t 100798922752.0....................\n",
      " 4800 epochs\t 100705771520.0....................\n",
      " 5000 epochs\t 100589600768.0....................\n",
      " 5200 epochs\t 100741644288.0....................\n",
      " 5400 epochs\t 100445331456.0....................\n",
      " 5600 epochs\t 100033912832.0....................\n",
      " 5800 epochs\t 100090503168.0....................\n",
      " 6000 epochs\t 99783442432.0....................\n",
      " 6200 epochs\t 99597598720.0....................\n",
      " 6400 epochs\t 99501449216.0....................\n",
      " 6600 epochs\t 99396567040.0....................\n",
      " 6800 epochs\t 99172597760.0....................\n",
      " 7000 epochs\t 99297550336.0....................\n",
      " 7200 epochs\t 98920472576.0....................\n",
      " 7400 epochs\t 98710396928.0....................\n",
      " 7600 epochs\t 99409625088.0....................\n",
      " 7800 epochs\t 98824273920.0....................\n",
      " 8000 epochs\t 98491940864.0....................\n",
      " 8200 epochs\t 98362499072.0....................\n",
      " 8400 epochs\t 98204131328.0....................\n",
      " 8600 epochs\t 97791926272.0....................\n",
      " 8800 epochs\t 98336292864.0....................\n",
      " 9000 epochs\t 97478983680.0....................\n",
      " 9200 epochs\t 97524441088.0....................\n",
      " 9400 epochs\t 97725276160.0....................\n",
      " 9600 epochs\t 97317314560.0....................\n",
      " 9800 epochs\t 97025925120.0....................\n",
      " 10000 epochs\t 97029234688.0....................\n",
      " 10200 epochs\t 97153523712.0....................\n",
      " 10400 epochs\t 96660316160.0....................\n",
      " 10600 epochs\t 96632111104.0....................\n",
      " 10800 epochs\t 96666730496.0....................\n",
      " 11000 epochs\t 96590577664.0....................\n",
      " 11200 epochs\t 96293658624.0....................\n",
      " 11400 epochs\t 96113868800.0....................\n",
      " 11600 epochs\t 96159367168.0....................\n",
      " 11800 epochs\t 96158785536.0....................\n",
      " 12000 epochs\t 96043573248.0....................\n",
      " 12200 epochs\t 95716507648.0....................\n",
      " 12400 epochs\t 95753633792.0....................\n",
      " 12600 epochs\t 95439577088.0....................\n",
      " 12800 epochs\t 95402893312.0....................\n",
      " 13000 epochs\t 95235219456.0....................\n",
      " 13200 epochs\t 95280562176.0....................\n",
      " 13400 epochs\t 95163203584.0....................\n",
      " 13600 epochs\t 95065505792.0....................\n",
      " 13800 epochs\t 94916452352.0....................\n",
      " 14000 epochs\t 94886789120.0....................\n",
      " 14200 epochs\t 94834794496.0....................\n",
      " 14400 epochs\t 95000141824.0....................\n",
      " 14600 epochs\t 94598602752.0....................\n",
      " 14800 epochs\t 94436220928.0....................\n",
      " 15000 epochs\t 94386847744.0....................\n",
      " 15200 epochs\t 94420197376.0....................\n",
      " 15400 epochs\t 94149795840.0....................\n",
      " 15600 epochs\t 94224465920.0....................\n",
      " 15800 epochs\t 94167949312.0....................\n",
      " 16000 epochs\t 94376886272.0....................\n",
      " 16200 epochs\t 93828661248.0....................\n",
      " 16400 epochs\t 93668818944.0....................\n",
      " 16600 epochs\t 93838598144.0....................\n",
      " 16800 epochs\t 93699252224.0....................\n",
      " 17000 epochs\t 93462577152.0....................\n",
      " 17200 epochs\t 93416620032.0....................\n",
      " 17400 epochs\t 93407141888.0....................\n",
      " 17600 epochs\t 93328416768.0....................\n",
      " 17800 epochs\t 93160570880.0....................\n",
      " 18000 epochs\t 93007495168.0....................\n",
      " 18200 epochs\t 93299122176.0....................\n",
      " 18400 epochs\t 92859015168.0....................\n",
      " 18600 epochs\t 92830310400.0....................\n",
      " 18800 epochs\t 92850380800.0....................\n",
      " 19000 epochs\t 92490260480.0....................\n",
      " 19200 epochs\t 92229713920.0....................\n",
      " 19400 epochs\t 92175515648.0....................\n",
      " 19600 epochs\t 92243329024.0....................\n",
      " 19800 epochs\t 91903623168.0....................\n",
      " 20000 epochs\t 91829616640.0....................\n",
      " 20200 epochs\t 91998363648.0....................\n",
      " 20400 epochs\t 91608735744.0....................\n",
      " 20600 epochs\t 92019163136.0....................\n",
      " 20800 epochs\t 91489370112.0....................\n",
      " 21000 epochs\t 91770798080.0....................\n",
      " 21200 epochs\t 91665530880.0....................\n",
      " 21400 epochs\t 91152982016.0....................\n",
      " 21600 epochs\t 91197054976.0....................\n",
      " 21800 epochs\t 91113316352.0....................\n",
      " 22000 epochs\t 90736345088.0....................\n",
      " 22200 epochs\t 91066761216.0....................\n",
      " 22400 epochs\t 91249590272.0....................\n",
      " 22600 epochs\t 91069792256.0....................\n",
      " 22800 epochs\t 90294116352.0....................\n",
      " 23000 epochs\t 90387947520.0....................\n",
      " 23200 epochs\t 90175537152.0....................\n",
      " 23400 epochs\t 90096680960.0....................\n",
      " 23600 epochs\t 91122548736.0....................\n",
      " 23800 epochs\t 89989873664.0....................\n",
      " 24000 epochs\t 90075840512.0....................\n",
      " 24200 epochs\t 89440198656.0....................\n",
      " 24400 epochs\t 89344794624.0....................\n",
      " 24600 epochs\t 88797184000.0....................\n",
      " 24800 epochs\t 88985739264.0....................\n",
      " 25000 epochs\t 88539488256.0....................\n",
      " 25200 epochs\t 88464039936.0....................\n",
      " 25400 epochs\t 88781717504.0....................\n",
      " 25600 epochs\t 89928957952.0....................\n",
      " 25800 epochs\t 87814848512.0....................\n",
      " 26000 epochs\t 88145633280.0....................\n",
      " 26200 epochs\t 88014241792.0....................\n",
      " 26400 epochs\t 88172707840.0....................\n",
      " 26600 epochs\t 88758501376.0....................\n",
      " 26800 epochs\t 87717281792.0....................\n",
      " 27000 epochs\t 88213839872.0....................\n",
      " 27200 epochs\t 87326531584.0....................\n",
      " 27400 epochs\t 87160520704.0....................\n",
      " 27600 epochs\t 86589808640.0....................\n",
      " 27800 epochs\t 87894376448.0....................\n",
      " 28000 epochs\t 87429816320.0....................\n",
      " 28200 epochs\t 86069813248.0....................\n",
      " 28400 epochs\t 87998357504.0....................\n",
      " 28600 epochs\t 86545866752.0....................\n",
      " 28800 epochs\t 85453996032.0....................\n",
      " 29000 epochs\t 86112821248.0....................\n",
      " 29200 epochs\t 86266798080.0....................\n",
      " 29400 epochs\t 85861343232.0....................\n",
      " 29600 epochs\t 87073816576.0....................\n",
      " 29800 epochs\t 85273305088.0....................\n",
      " 30000 epochs\t 84611883008.0....................\n",
      " 30200 epochs\t 85594398720.0....................\n",
      " 30400 epochs\t 88114249728.0....................\n",
      " 30600 epochs\t 84871331840.0....................\n",
      " 30800 epochs\t 84343980032.0....................\n",
      " 31000 epochs\t 88334139392.0....................\n",
      " 31200 epochs\t 83869368320.0....................\n",
      " 31400 epochs\t 84681506816.0....................\n",
      " 31600 epochs\t 84936572928.0....................\n",
      " 31800 epochs\t 84687175680.0....................\n",
      " 32000 epochs\t 84102184960.0....................\n",
      " 32200 epochs\t 83969646592.0....................\n",
      " 32400 epochs\t 83315384320.0....................\n",
      " 32600 epochs\t 83605757952.0....................\n",
      " 32800 epochs\t 83648929792.0....................\n",
      " 33000 epochs\t 84228677632.0....................\n",
      " 33200 epochs\t 83847528448.0....................\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " 33400 epochs\t 82302951424.0....................\n",
      " 33600 epochs\t 84306673664.0....................\n",
      " 33800 epochs\t 82715181056.0....................\n",
      " 34000 epochs\t 82008768512.0....................\n",
      " 34200 epochs\t 87283277824.0....................\n",
      " 34400 epochs\t 85313789952.0....................\n",
      " 34600 epochs\t 82644049920.0....................\n",
      " 34800 epochs\t 82367709184.0....................\n",
      " 35000 epochs\t 82552979456.0....................\n",
      " 35200 epochs\t 80597884928.0....................\n",
      " 35400 epochs\t 82198757376.0....................\n",
      " 35600 epochs\t 80051437568.0....................\n",
      " 35800 epochs\t 80899661824.0....................\n",
      " 36000 epochs\t 80050610176.0....................\n",
      " 36200 epochs\t 81787305984.0....................\n",
      " 36400 epochs\t 81261805568.0....................\n",
      " 36600 epochs\t 79163375616.0....................\n",
      " 36800 epochs\t 81091764224.0....................\n",
      " 37000 epochs\t 79173386240.0....................\n",
      " 37200 epochs\t 81807564800.0....................\n",
      " 37400 epochs\t 80919863296.0....................\n",
      " 37600 epochs\t 81827217408.0....................\n",
      " 37800 epochs\t 80050946048.0....................\n",
      " 38000 epochs\t 79374974976.0....................\n",
      " 38200 epochs\t 82774204416.0....................\n",
      " 38400 epochs\t 79727149056.0....................\n",
      " 38600 epochs\t 79658459136.0....................\n",
      " 38800 epochs\t 80280346624.0....................\n",
      " 39000 epochs\t 78853234688.0....................\n",
      " 39200 epochs\t 80062144512.0....................\n",
      " 39400 epochs\t 79108202496.0....................\n",
      " 39600 epochs\t 77714022400.0....................\n",
      " 39800 epochs\t 77965025280.0....................\n",
      " 40000 epochs\t 82021302272.0....................\n",
      " 40200 epochs\t 78738382848.0....................\n",
      " 40400 epochs\t 78640242688.0....................\n",
      " 40600 epochs\t 79219056640.0....................\n",
      " 40800 epochs\t 78734966784.0....................\n",
      " 41000 epochs\t 78637563904.0....................\n",
      " 41200 epochs\t 79647973376.0....................\n",
      " 41400 epochs\t 81445560320.0....................\n",
      " 41600 epochs\t 77084270592.0....................\n",
      " 41800 epochs\t 76743319552.0....................\n",
      " 42000 epochs\t 76818636800.0....................\n",
      " 42200 epochs\t 77705281536.0....................\n",
      " 42400 epochs\t 81533558784.0....................\n",
      " 42600 epochs\t 78518829056.0....................\n",
      " 42800 epochs\t 76646801408.0....................\n",
      " 43000 epochs\t 81306230784.0....................\n",
      " 43200 epochs\t 80655319040.0....................\n",
      " 43400 epochs\t 76556754944.0....................\n",
      " 43600 epochs\t 80494911488.0....................\n",
      " 43800 epochs\t 77226115072.0....................\n",
      " 44000 epochs\t 77495083008.0....................\n",
      " 44200 epochs\t 79207251968.0....................\n",
      " 44400 epochs\t 76353232896.0....................\n",
      " 44600 epochs\t 76021407744.0....................\n",
      " 44800 epochs\t 79752101888.0....................\n",
      " 45000 epochs\t 78461673472.0....................\n",
      " 45200 epochs\t 78060797952.0....................\n",
      " 45400 epochs\t 76041543680.0....................\n",
      " 45600 epochs\t 75994570752.0....................\n",
      " 45800 epochs\t 80915161088.0....................\n",
      " 46000 epochs\t 78096220160.0....................\n",
      " 46200 epochs\t 80740769792.0....................\n",
      " 46400 epochs\t 79256584192.0....................\n",
      " 46600 epochs\t 76562628608.0....................\n",
      " 46800 epochs\t 76038553600.0....................\n",
      " 47000 epochs\t 77423386624.0....................\n",
      " 47200 epochs\t 75769700352.0....................\n",
      " 47400 epochs\t 75090321408.0....................\n",
      " 47600 epochs\t 74698850304.0....................\n",
      " 47800 epochs\t 79541329920.0....................\n",
      " 48000 epochs\t 78529142784.0....................\n",
      " 48200 epochs\t 73803268096.0....................\n",
      " 48400 epochs\t 73831161856.0....................\n",
      " 48600 epochs\t 82215788544.0....................\n",
      " 48800 epochs\t 72968003584.0....................\n",
      " 49000 epochs\t 74216603648.0....................\n",
      " 49200 epochs\t 75424546816.0....................\n",
      " 49400 epochs\t 74276601856.0....................\n",
      " 49600 epochs\t 75088830464.0....................\n",
      " 49800 epochs\t 75074404352.0....................\n",
      " 50000 epochs\t 76013010944.0....................\n",
      " 50200 epochs\t 74256007168.0....................\n",
      " 50400 epochs\t 73919094784.0....................\n",
      " 50600 epochs\t 77791002624.0....................\n",
      " 50800 epochs\t 81230086144.0....................\n",
      " 51000 epochs\t 76264865792.0....................\n",
      " 51200 epochs\t 76916531200.0....................\n",
      " 51400 epochs\t 77143400448.0....................\n",
      " 51600 epochs\t 73095151616.0....................\n",
      " 51800 epochs\t 74891354112.0....................\n",
      " 52000 epochs\t 74090274816.0....................\n",
      " 52200 epochs\t 76255510528.0....................\n",
      " 52400 epochs\t 73836609536.0....................\n",
      " 52600 epochs\t 77303414784.0....................\n",
      " 52800 epochs\t 76572950528.0....................\n",
      " 53000 epochs\t 80192086016.0....................\n",
      " 53200 epochs\t 75733155840.0....................\n",
      " 53400 epochs\t 74515005440.0....................\n",
      " 53600 epochs\t 76145287168.0....................\n",
      " 53800 epochs\t 74390003712.0....................\n",
      " 54000 epochs\t 73032474624.0....................\n",
      " 54200 epochs\t 74121175040.0....................\n",
      " 54400 epochs\t 73675833344.0....................\n",
      " 54600 epochs\t 73396805632.0....................\n",
      " 54800 epochs\t 73080061952.0....................\n",
      " 55000 epochs\t 73484091392.0....................\n",
      " 55200 epochs\t 73696116736.0....................\n",
      " 55400 epochs\t 72742486016.0....................\n",
      " 55600 epochs\t 73054543872.0....................\n",
      " 55800 epochs\t 75235909632.0....................\n",
      " 56000 epochs\t 81216126976.0....................\n",
      " 56200 epochs\t 73430827008.0....................\n",
      " 56400 epochs\t 71968309248.0....................\n",
      " 56600 epochs\t 73881968640.0....................\n",
      " 56800 epochs\t 71634321408.0....................\n",
      " 57000 epochs\t 80463372288.0....................\n",
      " 57200 epochs\t 76263948288.0....................\n",
      " 57400 epochs\t 73002803200.0....................\n",
      " 57600 epochs\t 70262800384.0....................\n",
      " 57800 epochs\t 76309979136.0....................\n",
      " 58000 epochs\t 74614669312.0....................\n",
      " 58200 epochs\t 78026711040.0....................\n",
      " 58400 epochs\t 71561248768.0....................\n",
      " 58600 epochs\t 72824651776.0....................\n",
      " 58800 epochs\t 72399224832.0....................\n",
      " 59000 epochs\t 71020306432.0....................\n",
      " 59200 epochs\t 72211644416.0....................\n",
      " 59400 epochs\t 70021013504.0....................\n",
      " 59600 epochs\t 73732513792.0....................\n",
      " 59800 epochs\t 72557608960.0...................."
     ]
    }
   ],
   "source": [
    "# Inference\n",
    "VI = inf.inference.VI(qmodel_, optimizer=tf.train.AdamOptimizer(learning_rate = 0.0005), \n",
    "                      epochs=nr_epochs)\n",
    "model_.fit({'x': x_train- np.mean(x_train,axis=0)}, VI)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### infering posterior using MCMC method\n",
    "for very high dimensional data it seems convergence is very slow "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'1.3.0'"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "MC = inf.inference.MCMC(num_leapfrog_steps=5, num_burnin_steps=60000,num_results=300)\n",
    "inf.__version__\n",
    "model_.fit({'x': x_train-np.mean(x_train, axis = 0)}, MC)\n",
    "z_samples_mean = np.mean(model_.posterior('z').parameters()['samples'], axis = 0)\n",
    "w_samples_mean = np.mean(model_.posterior('w').parameters()['samples'], axis = 0)\n",
    "print('Z_sample_mean', z_samples_mean.shape, 'W_sample_mean', w_samples_mean.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Posterior parameters give parameter value for each random variable in the posterior distribution (example; z --> traces or w --> filters) \n",
    "Posterior sampling we get distribution for each element of each vaiable. For example get distribution (based on posterior distibution) for each pixel at filters or time point at races. Advantgae of this is getting uncertaintity over element. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'alpha': {'name': 'alpha',\n",
       "  'allow_nan_stats': True,\n",
       "  'validate_args': False,\n",
       "  'rate': array([ 4.0652833,  3.0517867,  5.0304503,  5.2161617,  5.957747 ,\n",
       "          5.94376  ,  4.972177 ,  5.279306 ,  5.6035843,  7.9734774,\n",
       "          4.877214 ,  5.920435 ,  5.3828897,  8.751523 , 10.722291 ,\n",
       "          9.249409 , 10.357039 ,  8.950016 ,  9.991635 ,  7.588208 ,\n",
       "          9.513534 ,  7.774371 , 15.742898 , 16.811611 , 15.4892845,\n",
       "         16.706179 , 14.956008 , 14.896372 , 18.011213 , 17.957197 ],\n",
       "        dtype=float32),\n",
       "  'concentration': 50},\n",
       " 'beta': {'name': 'beta',\n",
       "  'allow_nan_stats': True,\n",
       "  'validate_args': False,\n",
       "  'concentration0': array([13.176184 , 16.912535 , 11.143823 ,  8.489727 ,  9.124147 ,\n",
       "          7.8671575,  9.263139 , 10.555648 ,  7.421107 ,  6.784775 ,\n",
       "          8.888588 ,  7.73944  , 10.166553 ,  5.987043 ,  4.8087735,\n",
       "          6.5431986,  3.8316302,  4.757522 ,  6.0878   ,  6.901407 ,\n",
       "          5.468239 ,  6.135337 ,  3.913514 ,  3.380848 ,  2.9734952,\n",
       "          2.4629908,  3.3779035,  2.5956452,  2.9369683,  3.244911 ],\n",
       "        dtype=float32),\n",
       "  'concentration1': 1.0},\n",
       " 'weight_': {'name': 'weight_',\n",
       "  'allow_nan_stats': True,\n",
       "  'validate_args': False,\n",
       "  'rtol': None,\n",
       "  'atol': None,\n",
       "  'loc': array([5.2477259e-02, 1.7539334e-02, 2.5497508e-01, 2.0353833e-01,\n",
       "         7.1912192e-02, 1.1310530e-01, 9.6324803e-03, 3.7908845e-02,\n",
       "         3.7479130e-04, 6.2299609e-02, 4.9631237e-03, 6.7165350e-03,\n",
       "         3.6967276e-03, 3.3944797e-02, 7.0421766e-03, 2.7240707e-02,\n",
       "         3.1258799e-02, 1.3626139e-02, 8.7212734e-03, 4.1453444e-04,\n",
       "         1.3084940e-03, 1.0181811e-02, 1.4251358e-02, 5.8072747e-04,\n",
       "         5.3099580e-03, 3.0157687e-03, 9.4734470e-04, 1.7660843e-04,\n",
       "         8.7337242e-04, 1.4377358e-04], dtype=float32)},\n",
       " 'w': {'name': 'w',\n",
       "  'allow_nan_stats': True,\n",
       "  'validate_args': False,\n",
       "  'scale': array([[0.00587024, 0.00594336, 0.00596078, ..., 0.00590449, 0.00588672,\n",
       "          0.00572364],\n",
       "         [0.00948719, 0.0092589 , 0.00948046, ..., 0.00955709, 0.00926646,\n",
       "          0.00877063],\n",
       "         [0.00623436, 0.00596967, 0.00591789, ..., 0.00601446, 0.0059955 ,\n",
       "          0.00589015],\n",
       "         ...,\n",
       "         [0.2119673 , 0.2056537 , 0.21738684, ..., 0.20528853, 0.21937756,\n",
       "          0.20363961],\n",
       "         [0.19174004, 0.22503556, 0.2082078 , ..., 0.20156448, 0.23100214,\n",
       "          0.21928383],\n",
       "         [0.26602423, 0.2551662 , 0.28952944, ..., 0.2884858 , 0.25936612,\n",
       "          0.24477535]], dtype=float32),\n",
       "  'loc': array([[ 1.4656510e-03,  1.8102597e-04,  1.2849967e-03, ...,\n",
       "           8.2390776e-05, -1.8075047e-03,  9.4555202e-04],\n",
       "         [-1.5281842e-03,  2.0773339e-03, -1.3097655e-03, ...,\n",
       "           2.6300689e-03, -1.7509621e-03, -7.2688237e-04],\n",
       "         [-1.4525680e-03,  1.1627943e-03,  2.4938521e-03, ...,\n",
       "           2.2024105e-04,  8.0020889e-04, -1.8436743e-03],\n",
       "         ...,\n",
       "         [ 5.3700581e-03,  1.5284276e-03,  1.8501141e-03, ...,\n",
       "          -8.2489830e-03,  6.2149386e-03, -3.8149670e-02],\n",
       "         [ 1.1507865e-02,  3.5354965e-03,  1.8383719e-02, ...,\n",
       "           2.6953595e-03, -6.6077095e-03, -1.3133867e-02],\n",
       "         [ 1.9159002e-02,  2.1335317e-02,  5.3511951e-02, ...,\n",
       "          -1.1295422e-02,  1.7526328e-03,  2.0115804e-02]], dtype=float32)},\n",
       " 'z': {'name': 'z',\n",
       "  'allow_nan_stats': True,\n",
       "  'validate_args': False,\n",
       "  'scale': array([[0.14912851, 0.21019675, 0.22381061, ..., 0.97459364, 0.98522854,\n",
       "          0.99771583],\n",
       "         [0.34138137, 0.39643738, 0.3020305 , ..., 1.0447264 , 1.1528132 ,\n",
       "          1.1124377 ],\n",
       "         [0.17400053, 0.25383142, 0.35394847, ..., 1.0224406 , 1.0455645 ,\n",
       "          1.157092  ],\n",
       "         ...,\n",
       "         [0.22831234, 0.2689662 , 0.2361399 , ..., 1.0516961 , 1.0070574 ,\n",
       "          1.2227203 ],\n",
       "         [0.2755839 , 0.23451097, 0.2711853 , ..., 1.0588934 , 0.89488655,\n",
       "          1.1169059 ],\n",
       "         [0.16135973, 0.19426608, 0.17246914, ..., 1.0002701 , 0.9664875 ,\n",
       "          1.0721898 ]], dtype=float32),\n",
       "  'loc': array([[ -3.1009386 ,   5.9981556 ,  -3.050662  , ...,  12.731844  ,\n",
       "           12.439371  ,  -4.809799  ],\n",
       "         [ 17.24734   ,  18.895966  ,  10.249751  , ...,  -2.2421358 ,\n",
       "           11.42282   ,  -2.3363914 ],\n",
       "         [  3.3872662 ,  -5.654718  ,  13.390474  , ...,  -5.3187895 ,\n",
       "           -5.097742  ,   7.003034  ],\n",
       "         ...,\n",
       "         [ -8.84354   ,   9.277351  ,  -0.56242055, ...,   5.7098794 ,\n",
       "            0.32360604, -12.579071  ],\n",
       "         [ 12.553365  ,   0.14744496,  10.252143  , ..., -12.027735  ,\n",
       "            0.67489094,  -5.4574647 ],\n",
       "         [  4.1977897 ,   0.5447848 ,  -2.7322946 , ...,  -3.2933767 ,\n",
       "            3.722693  ,  -5.3775945 ]], dtype=float32)}}"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_.posterior().parameters()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.05286612 0.08767564 0.07164619 0.07759365 0.05331769 0.06248718\n",
      " 0.05093279 0.04083276 0.06937656 0.04793475 0.01575632 0.03124124\n",
      " 0.02135162 0.05411498 0.00331729 0.01557737 0.02874134 0.0429587\n",
      " 0.02119667 0.03702882 0.04681718 0.00580368 0.02059906 0.02023978\n",
      " 0.00490057 0.00054741 0.00460413 0.00305828 0.00296881 0.00132643]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Sorted components')"
      ]
     },
     "execution_count": 13,
     "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.plot(np.sort(model_.posterior('weight_').parameters()['loc']))\n",
    "Weights_ = model_.posterior('weight_').parameters()['loc']\n",
    "print(Weights_)\n",
    "plt.ylabel('Value of weights')\n",
    "plt.xlabel('Sorted components')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# posterior  parameters\n",
    "z_loc = model_.posterior('z').parameters()['loc']\n",
    "\n",
    "w_loc = model_.posterior('w').parameters()['loc']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_params(filters, traces, Weights):\n",
    "    plt.figure(figsize=(30,20))\n",
    "    for i in range(filters.shape[0]):\n",
    "        filters[i] = Weights[i] * filters[i]\n",
    "    \n",
    "    min_ = filters.min()\n",
    "    max_ = filters.max()\n",
    "    for i in range(30):\n",
    "        plt.subplot(10,8,i*2+1)\n",
    "        plt.imshow(filters[i], vmin=min_, vmax=max_)\n",
    "        plt.axis('off')\n",
    "        plt.title(i)\n",
    "        plt.subplot(10,8,i*2+2)\n",
    "        plt.plot(traces[i])\n",
    "        plt.title(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2160x1440 with 60 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_params(w_loc.reshape(-1,28,28), z_loc.T, Weights_)\n",
    "# It seems this needs more iteration"
   ]
  }
 ],
 "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": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}