3_Deep_Neural_Network.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Todos\n",
    "\n",
    "- modularity and the layers of neural network\n",
    "- Activation function\n",
    "- Relu and Sigmoid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow import keras\n",
    "from tensorflow.keras import layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = keras.Sequential([\n",
    "    layers.Dense(units = 4, activation='relu', input_shape=[2]),\n",
    "    layers.Dense(units = 3, activation='relu'),\n",
    "    layers.Dense(units= 1)\n",
    "])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- relu: rectifier function\n",
    "- activation function is something that is applied to each of a layers putputs (or its activations)\n",
    "- relu max(0, x) (jati positive, teti activated)\n",
    "- to convert model to adapt outside linear scope\n",
    "- dense layers by themselves can never move out of lines/planes/hyperplanes\n",
    "- usually output layer does not need activation (but wait until we cover binary classification)"
   ]
  },
  {
   "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}