function_approx_nn_tf.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Imports\n",
    "import numpy as np\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load training data\n",
    "x = -50 + np.random.random((25000,1))*100\n",
    "y = x**2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x216e2e65388>]"
      ]
     },
     "execution_count": 3,
     "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": [
    "import matplotlib.pyplot as plt\n",
    "plt.plot(x, y, 'ro')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 25000 samples\n",
      "Epoch 1/15\n",
      "25000/25000 [==============================] - 3s 117us/sample - loss: 571212.1455\n",
      "Epoch 2/15\n",
      "25000/25000 [==============================] - 1s 32us/sample - loss: 78864.1392\n",
      "Epoch 3/15\n",
      "25000/25000 [==============================] - 1s 26us/sample - loss: 72184.4599\n",
      "Epoch 4/15\n",
      "25000/25000 [==============================] - 1s 26us/sample - loss: 60329.7515\n",
      "Epoch 5/15\n",
      "25000/25000 [==============================] - 1s 26us/sample - loss: 36910.6294\n",
      "Epoch 6/15\n",
      "25000/25000 [==============================] - 1s 27us/sample - loss: 14917.5852\n",
      "Epoch 7/15\n",
      "25000/25000 [==============================] - 1s 27us/sample - loss: 5051.5360\n",
      "Epoch 8/15\n",
      "25000/25000 [==============================] - 1s 26us/sample - loss: 2024.9181\n",
      "Epoch 9/15\n",
      "25000/25000 [==============================] - 1s 26us/sample - loss: 959.2767\n",
      "Epoch 10/15\n",
      "25000/25000 [==============================] - 1s 27us/sample - loss: 555.3603\n",
      "Epoch 11/15\n",
      "25000/25000 [==============================] - 1s 27us/sample - loss: 348.9903\n",
      "Epoch 12/15\n",
      "25000/25000 [==============================] - 1s 30us/sample - loss: 238.7864\n",
      "Epoch 13/15\n",
      "25000/25000 [==============================] - 1s 27us/sample - loss: 155.8856\n",
      "Epoch 14/15\n",
      "25000/25000 [==============================] - 1s 26us/sample - loss: 120.1804\n",
      "Epoch 15/15\n",
      "25000/25000 [==============================] - 1s 27us/sample - loss: 84.8273\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x216e416ffc8>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Define model\n",
    "model = Sequential()\n",
    "model.add(Dense(40, input_dim=1, activation='relu'))\n",
    "model.add(Dense(20, activation='relu'))\n",
    "model.add(Dense(10, activation='relu'))\n",
    "model.add(Dense(1))\n",
    "model.compile(loss='mean_squared_error', optimizer='adam')\n",
    "model.fit(x, y, epochs=15, batch_size=50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[  101.73765 ]\n",
      " [   20.854744]\n",
      " [10774.016   ]\n",
      " [  173.8088  ]]\n"
     ]
    }
   ],
   "source": [
    "predictions = model.predict([10, 5, 200, 13])\n",
    "print(predictions) # Approximately 100, 25, 40000, 169"
   ]
  },
  {
   "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": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}