{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /Users/charnock/Library/Python/3.7/lib/python/site-packages/tensorflow/python/framework/op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Colocations handled automatically by placer.\n"
     ]
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import tensorflow as tf\n",
    "\n",
    "arbitrary_shape = (11, 17, 3)\n",
    "\n",
    "data = tf.placeholder(dtype=tf.float32, \n",
    "                      shape=arbitrary_shape,\n",
    "                      name=\"data\")\n",
    "\n",
    "init_scaling = tf.constant_initializer(1.)\n",
    "init_bias = tf.constant_initializer(0.)\n",
    "\n",
    "a = tf.get_variable(\"a\",\n",
    "                    shape=(),\n",
    "                    dtype=tf.float32,\n",
    "                    initializer=init_scaling)\n",
    "b = tf.get_variable(\"b\",\n",
    "                    shape=(),\n",
    "                    dtype=tf.float32,\n",
    "                    initializer=init_bias)\n",
    "\n",
    "scaled_biased_data = tf.add(\n",
    "    tf.multiply(a, data, name=\"scaled_data\"), \n",
    "    b, name=\"scaled_biased_data\")\n",
    "\n",
    "reduced_dynamic_data = tf.asinh(scaled_biased_data, \n",
    "                                name=\"reduced_dynamic_data\")\n",
    "\n",
    "c = tf.get_variable(\"c\",\n",
    "                    shape=(),\n",
    "                    dtype=tf.float32,\n",
    "                    initializer=init_scaling)\n",
    "d = tf.get_variable(\"d\",\n",
    "                    shape=(),\n",
    "                    dtype=tf.float32,\n",
    "                    initializer=init_bias)\n",
    "\n",
    "even_further_scaled_biased_reduced_dynamic_data = tf.add(\n",
    "    tf.multiply(c, reduced_dynamic_data, name=\"scaled_reduced_dynamic_data\"), \n",
    "                d, name=\"scaled_biased_reduced_dynamic_data\")\n",
    "\n",
    "initialiser = tf.global_variables_initializer()\n",
    "sess = tf.Session()\n",
    "sess.run(initialiser)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "sess = tf.Session()\n",
    "sess.run(initialiser)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "input_data = np.random.uniform(-1e6, 1e6, arbitrary_shape)\n",
    "output_data = sess.run(\"scaled_biased_reduced_dynamic_data:0\", feed_dict={\"data:0\": input_data})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x132bdd780>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x432 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(2, 3, figsize=(15, 6))\n",
    "ax[0, 0].imshow(input_data[:, :, 0])\n",
    "ax[0, 1].imshow(input_data[:, :, 1])\n",
    "ax[0, 2].imshow(input_data[:, :, 2])\n",
    "ax[1, 0].imshow(output_data[:, :, 0])\n",
    "ax[1, 1].imshow(output_data[:, :, 1])\n",
    "ax[1, 2].imshow(output_data[:, :, 2])"
   ]
  },
  {
   "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": "3.7.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}