playground.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/tachi/.pyenv/versions/anaconda3-5.2.0/lib/python3.6/site-packages/tensorflow/python/client/session.py:1702: UserWarning: An interactive session is already active. This can cause out-of-memory errors in some cases. You must explicitly call `InteractiveSession.close()` to release resources held by the other session(s).\n",
      "  warnings.warn('An interactive session is already active. This can '\n"
     ]
    }
   ],
   "source": [
    "# Import libraries\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import tensorflow as tf\n",
    "#import tensorflow.contrib.eager as tfe # For eager mode\n",
    "#tfe.enable_eager_execution()\n",
    "\n",
    "%matplotlib inline\n",
    "tf.set_random_seed(0)\n",
    "sess = tf.InteractiveSession()\n",
    "tfd = tf.contrib.distributions\n",
    "tfb = tfd.bijectors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Setting\n",
    "DTYPE = tf.float32\n",
    "NP_DTYPE = np.float32\n",
    "MODEL = 'NVP' # Which Normalizing Flow to use. 'NVP' or 'MAF' or 'IAF'\n",
    "USE_BATCHNORM = True"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Generate samples from a complex distribution\n",
    "batch_size=512\n",
    "x2_dist = tfd.Normal(loc=0., scale=4.)\n",
    "x2_samples = x2_dist.sample(batch_size)\n",
    "x1 = tfd.Normal(loc=.25 * tf.square(x2_samples),\n",
    "                scale=tf.ones(batch_size, dtype=tf.float32))\n",
    "x1_samples = x1.sample()\n",
    "x_samples = tf.stack([x1_samples, x2_samples], axis=1)\n",
    "\n",
    "# Generate samples from a source distribution\n",
    "base_dist = tfd.MultivariateNormalDiag(loc=tf.zeros([2], dtype=tf.float32))\n",
    "base_samples = base_dist.sample(batch_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Check the samples\n",
    "base_list = sess.run(base_dist.sample(batch_size))\n",
    "x_list = sess.run(x_samples)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(512, 2)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.shape(base_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(512, 2)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.shape(x_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-1.27435   , -0.91860074],\n",
       "       [ 1.0175929 , -0.3732663 ],\n",
       "       [ 0.31666058,  1.585016  ],\n",
       "       ...,\n",
       "       [-0.90286815,  0.04322631],\n",
       "       [ 1.0403734 , -0.44794634],\n",
       "       [-1.1801741 ,  2.3731496 ]], dtype=float32)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "base_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 3.0461538 ,  4.007266  ],\n",
       "       [ 3.918519  ,  4.575738  ],\n",
       "       [-1.0984588 , -0.29615825],\n",
       "       ...,\n",
       "       [ 1.58495   ,  3.0814867 ],\n",
       "       [ 6.6487174 , -5.3859544 ],\n",
       "       [10.750812  , -6.372147  ]], dtype=float32)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x1c3e25c1d0>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the samples\n",
    "plt.figure(figsize=(5, 5))\n",
    "plt.scatter(base_list[:,0], base_list[:,1], color='Blue')\n",
    "#plt.tick_params(bottom=False, left=False, right=False,top=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x1c3e3144a8>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(5, 5))\n",
    "plt.scatter(x_list[:,0], x_list[:,1], color='Red')\n",
    "#plt.tick_params(bottom=False, left=False, right=False, top=False)"
   ]
  },
  {
   "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}