06. Deep Residual Networks.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Deep Residual Networks\n",
    "\n",
    "Increasing the number of layers in CNNs has been observed to increase training errors.\n",
    "\n",
    "In theory, a deeper model should have at least the same error rate as the shallower counterpart, since the additional layers can perform _identity mapping_ (pass the input to the output without any transformations). In practice, optimization algorithms fail to find the optimal solution when going deeper.\n",
    "\n",
    "The solution proposed in [Deep Residual Learning for Image Recognition](https://arxiv.org/pdf/1512.03385.pdf) builds upon an assumption that the optimal function being modelled by a block of layers is closer to an identity mapping ($F(x) = x$) of the input than to something that produces completely unrelated output."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"data:image/svg+xml;base64,<?xml version="1.0" encoding="UTF-8" standalone="no"?><svg   xmlns:dc="http://purl.org/dc/elements/1.1/"   xmlns:cc="http://creativecommons.org/ns#"   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"   xmlns:svg="http://www.w3.org/2000/svg"   xmlns="http://www.w3.org/2000/svg"   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"   width="283.34451"   height="352.82236"   viewBox="0 0 74.968233 93.350917"   version="1.1"   id="svg8"   inkscape:version="0.92.3 (2405546, 2018-03-11)"   sodipodi:docname="resblock.svg">  <defs     id="defs2" />  <sodipodi:namedview     id="base"     pagecolor="#ffffff"     bordercolor="#666666"     borderopacity="1.0"     inkscape:pageopacity="0.0"     inkscape:pageshadow="2"     inkscape:zoom="5.6"     inkscape:cx="210.7607"     inkscape:cy="73.753515"     inkscape:document-units="mm"     inkscape:current-layer="layer1"     showgrid="false"     inkscape:window-width="1920"     inkscape:window-height="1014"     inkscape:window-x="0"     inkscape:window-y="32"     inkscape:window-maximized="1"     showguides="false"     fit-margin-top="0"     fit-margin-left="0"     fit-margin-right="0"     fit-margin-bottom="0"     units="px"     scale-x="0.5" />  <metadata     id="metadata5">    <rdf:RDF>      <cc:Work         rdf:about="">        <dc:format>image/svg+xml</dc:format>        <dc:type           rdf:resource="http://purl.org/dc/dcmitype/StillImage" />        <dc:title />      </cc:Work>    </rdf:RDF>  </metadata>  <g     inkscape:label="Layer 1"     inkscape:groupmode="layer"     id="layer1"     transform="translate(-29.747128,-32.475815)">    <flowRoot       xml:space="preserve"       id="flowRoot4795"       style="font-style:normal;font-weight:normal;font-size:40px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none"       transform="matrix(0.52380664,0,0,0.52380664,14.165384,15.464767)"><flowRegion         id="flowRegion4797"><rect           id="rect4799"           width="170.71428"           height="42.142857"           x="120"           y="142.51968" /></flowRegion><flowPara         id="flowPara4801" /></flowRoot>    <circle       style="fill:none;stroke:#000000;stroke-width:0.13880876;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"       id="path4803"       cx="43.794052"       cy="44.620998"       r="4.4546947" />    <text       xml:space="preserve"       style="font-style:normal;font-weight:normal;font-size:3.36914182px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.08422855"       x="41.017639"       y="45.861004"       id="text4807"><tspan         sodipodi:role="line"         id="tspan4805"         x="41.017639"         y="45.861004"         style="font-size:3.32617211px;stroke-width:0.08422855">dot</tspan></text>    <text       xml:space="preserve"       style="font-style:normal;font-weight:normal;font-size:3.36914206px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.08422855"       x="42.816338"       y="34.294815"       id="text4807-3"><tspan         sodipodi:role="line"         id="tspan4805-6"         x="42.816338"         y="34.294815"         style="font-size:3.32617211px;stroke-width:0.08422855">x</tspan></text>    <text       xml:space="preserve"       style="font-style:normal;font-weight:normal;font-size:3.36914206px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.08422855"       x="29.607454"       y="45.193584"       id="text4807-7"><tspan         sodipodi:role="line"         id="tspan4805-5"         x="29.607454"         y="45.193584"         style="stroke-width:0.08422855"><tspan           style="font-size:3.32617211px;stroke-width:0.08422855"           id="tspan4852">w</tspan><tspan           style="font-size:2.18994236px;baseline-shift:sub;stroke-width:0.08422855"           id="tspan4835">1</tspan></tspan></text>    <circle       r="4.4546947"       cy="59.192799"       cx="43.794052"       id="circle4843"       style="fill:none;stroke:#000000;stroke-width:0.13880876;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />    <text       id="text4847"       y="60.448826"       x="40.605927"       style="font-style:normal;font-weight:normal;font-size:3.36914182px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.08422855"       xml:space="preserve"><tspan         style="font-size:3.32617211px;stroke-width:0.08422855"         y="60.448826"         x="40.605927"         id="tspan4845"         sodipodi:role="line">relu</tspan></text>    <g       id="g4892"       transform="matrix(0.2682318,0,0,0.21903975,37.964401,29.744701)"       style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none">      <g         id="g4896"         transform="translate(0,-1.2654371)"         style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none">        <path           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"           d="m 21.733631,93.643839 v 17.958841 l -2.834822,-7.18155"           id="path4873"           inkscape:connector-curvature="0"           sodipodi:nodetypes="ccc" />        <path           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"           d="m 21.733631,111.60268 2.834822,-7.18155"           id="path4873-3"           inkscape:connector-curvature="0"           sodipodi:nodetypes="cc" />      </g>    </g>    <g       style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none"       transform="matrix(0.2682318,0,0,0.21903975,37.964401,15.331287)"       id="g4904">      <g         style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none"         transform="translate(0,-1.2654371)"         id="g4902">        <path           sodipodi:nodetypes="ccc"           inkscape:connector-curvature="0"           id="path4898"           d="m 21.733631,93.643839 v 17.958841 l -2.834822,-7.18155"           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />        <path           sodipodi:nodetypes="cc"           inkscape:connector-curvature="0"           id="path4900"           d="m 21.733631,111.60268 2.834822,-7.18155"           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />      </g>    </g>    <g       style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none"       transform="matrix(0,-0.2682318,0.21903975,0,14.301532,50.450649)"       id="g4912">      <g         style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none"         transform="translate(0,-1.2654371)"         id="g4910">        <path           sodipodi:nodetypes="ccc"           inkscape:connector-curvature="0"           id="path4906"           d="m 21.733631,93.643839 v 17.958841 l -2.834822,-7.18155"           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />        <path           sodipodi:nodetypes="cc"           inkscape:connector-curvature="0"           id="path4908"           d="m 21.733631,111.60268 2.834822,-7.18155"           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />      </g>    </g>    <circle       r="4.4546947"       cy="73.507217"       cx="43.794052"       id="circle4933"       style="fill:none;stroke:#000000;stroke-width:0.13880876;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />    <text       id="text4937"       y="74.747223"       x="41.017639"       style="font-style:normal;font-weight:normal;font-size:3.36914182px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.08422855"       xml:space="preserve"><tspan         style="font-size:3.32617211px;stroke-width:0.08422855"         y="74.747223"         x="41.017639"         id="tspan4935"         sodipodi:role="line">dot</tspan></text>    <text       id="text4945"       y="74.079803"       x="29.607454"       style="font-style:normal;font-weight:normal;font-size:3.36914206px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.08422855"       xml:space="preserve"><tspan         style="stroke-width:0.08422855"         y="74.079803"         x="29.607454"         id="tspan4943"         sodipodi:role="line"><tspan           id="tspan4939"           style="font-size:3.32617211px;stroke-width:0.08422855">w</tspan><tspan           id="tspan4941"           style="font-size:2.18994236px;baseline-shift:sub;stroke-width:0.08422855">2</tspan></tspan></text>    <circle       style="fill:none;stroke:#000000;stroke-width:0.13880876;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"       id="circle4947"       cx="43.794052"       cy="88.07901"       r="4.4546947" />    <text       xml:space="preserve"       style="font-style:normal;font-weight:normal;font-size:3.36914182px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.08422855"       x="40.605927"       y="89.335045"       id="text4951"><tspan         sodipodi:role="line"         id="tspan4949"         x="40.605927"         y="89.335045"         style="font-size:3.32617211px;stroke-width:0.08422855">relu</tspan></text>    <g       style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none"       transform="matrix(0.2682318,0,0,0.21903975,37.964401,58.630923)"       id="g4959">      <g         style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none"         transform="translate(0,-1.2654371)"         id="g4957">        <path           sodipodi:nodetypes="ccc"           inkscape:connector-curvature="0"           id="path4953"           d="m 21.733631,93.643839 v 17.958841 l -2.834822,-7.18155"           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />        <path           sodipodi:nodetypes="cc"           inkscape:connector-curvature="0"           id="path4955"           d="m 21.733631,111.60268 2.834822,-7.18155"           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />      </g>    </g>    <g       id="g4967"       transform="matrix(0.2682318,0,0,0.21903975,37.964401,44.217509)"       style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none">      <g         id="g4965"         transform="translate(0,-1.2654371)"         style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none">        <path           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"           d="m 21.733631,93.643839 v 17.958841 l -2.834822,-7.18155"           id="path4961"           inkscape:connector-curvature="0"           sodipodi:nodetypes="ccc" />        <path           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"           d="m 21.733631,111.60268 2.834822,-7.18155"           id="path4963"           inkscape:connector-curvature="0"           sodipodi:nodetypes="cc" />      </g>    </g>    <g       id="g4975"       transform="matrix(0,-0.2682318,0.21903975,0,14.301532,79.336874)"       style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none">      <g         id="g4973"         transform="translate(0,-1.2654371)"         style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none">        <path           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"           d="m 21.733631,93.643839 v 17.958841 l -2.834822,-7.18155"           id="path4969"           inkscape:connector-curvature="0"           sodipodi:nodetypes="ccc" />        <path           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"           d="m 21.733631,111.60268 2.834822,-7.18155"           id="path4971"           inkscape:connector-curvature="0"           sodipodi:nodetypes="cc" />      </g>    </g>    <g       id="g71"       transform="matrix(0.2682318,0,0,0.21903975,37.964401,73.321558)"       style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none">      <g         id="g69"         transform="translate(0,-1.2654371)"         style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none">        <path           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"           d="m 21.733631,93.643839 v 17.958841 l -2.834822,-7.18155"           id="path65"           inkscape:connector-curvature="0"           sodipodi:nodetypes="ccc" />        <path           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"           d="m 21.733631,111.60268 2.834822,-7.18155"           id="path67"           inkscape:connector-curvature="0"           sodipodi:nodetypes="cc" />      </g>    </g>    <text       id="text75"       y="101.47139"       x="40.240505"       style="font-style:normal;font-weight:normal;font-size:3.36914206px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.08422855"       xml:space="preserve"><tspan         style="font-size:3.32617211px;stroke-width:0.08422855"         y="101.47139"         x="40.240505"         sodipodi:role="line"         id="tspan71">H(x)</tspan></text>    <circle       r="4.4546947"       cy="44.620998"       cx="86.479889"       id="circle76"       style="fill:none;stroke:#000000;stroke-width:0.13880876;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />    <text       id="text80"       y="45.861004"       x="83.703491"       style="font-style:normal;font-weight:normal;font-size:3.36914182px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.08422855"       xml:space="preserve"><tspan         style="font-size:3.32617211px;stroke-width:0.08422855"         y="45.861004"         x="83.703491"         id="tspan78"         sodipodi:role="line">dot</tspan></text>    <text       id="text84"       y="34.294815"       x="85.502197"       style="font-style:normal;font-weight:normal;font-size:3.36914206px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.08422855"       xml:space="preserve"><tspan         style="font-size:3.32617211px;stroke-width:0.08422855"         y="34.294815"         x="85.502197"         id="tspan82"         sodipodi:role="line">x</tspan></text>    <text       id="text92"       y="45.193584"       x="72.293335"       style="font-style:normal;font-weight:normal;font-size:3.36914206px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.08422855"       xml:space="preserve"><tspan         style="stroke-width:0.08422855"         y="45.193584"         x="72.293335"         id="tspan90"         sodipodi:role="line"><tspan           id="tspan86"           style="font-size:3.32617211px;stroke-width:0.08422855">w</tspan><tspan           id="tspan88"           style="font-size:2.18994236px;baseline-shift:sub;stroke-width:0.08422855">1</tspan></tspan></text>    <circle       style="fill:none;stroke:#000000;stroke-width:0.13880876;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"       id="circle94"       cx="86.479889"       cy="59.192799"       r="4.4546947" />    <text       xml:space="preserve"       style="font-style:normal;font-weight:normal;font-size:3.36914182px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.08422855"       x="83.291779"       y="60.448826"       id="text98"><tspan         sodipodi:role="line"         id="tspan96"         x="83.291779"         y="60.448826"         style="font-size:3.32617211px;stroke-width:0.08422855">relu</tspan></text>    <g       style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none"       transform="matrix(0.2682318,0,0,0.21903975,80.650296,29.744701)"       id="g106">      <g         style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none"         transform="translate(0,-1.2654371)"         id="g104">        <path           sodipodi:nodetypes="ccc"           inkscape:connector-curvature="0"           id="path100"           d="m 21.733631,93.643839 v 17.958841 l -2.834822,-7.18155"           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />        <path           sodipodi:nodetypes="cc"           inkscape:connector-curvature="0"           id="path102"           d="m 21.733631,111.60268 2.834822,-7.18155"           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />      </g>    </g>    <g       id="g114"       transform="matrix(0.2682318,0,0,0.21903975,80.650296,15.331287)"       style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none">      <g         id="g112"         transform="translate(0,-1.2654371)"         style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none">        <path           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"           d="m 21.733631,93.643839 v 17.958841 l -2.834822,-7.18155"           id="path108"           inkscape:connector-curvature="0"           sodipodi:nodetypes="ccc" />        <path           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"           d="m 21.733631,111.60268 2.834822,-7.18155"           id="path110"           inkscape:connector-curvature="0"           sodipodi:nodetypes="cc" />      </g>    </g>    <g       id="g122"       transform="matrix(0,-0.2682318,0.21903975,0,56.987412,50.450649)"       style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none">      <g         id="g120"         transform="translate(0,-1.2654371)"         style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none">        <path           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"           d="m 21.733631,93.643839 v 17.958841 l -2.834822,-7.18155"           id="path116"           inkscape:connector-curvature="0"           sodipodi:nodetypes="ccc" />        <path           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"           d="m 21.733631,111.60268 2.834822,-7.18155"           id="path118"           inkscape:connector-curvature="0"           sodipodi:nodetypes="cc" />      </g>    </g>    <circle       style="fill:none;stroke:#000000;stroke-width:0.13880876;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"       id="circle124"       cx="86.479889"       cy="73.507217"       r="4.4546947" />    <text       xml:space="preserve"       style="font-style:normal;font-weight:normal;font-size:3.36914182px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.08422855"       x="83.609131"       y="74.747223"       id="text128"><tspan         sodipodi:role="line"         id="tspan126"         x="83.609131"         y="74.747223"         style="font-size:3.32617211px;stroke-width:0.08422855">dot</tspan></text>    <text       xml:space="preserve"       style="font-style:normal;font-weight:normal;font-size:3.36914206px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.08422855"       x="72.293335"       y="74.079803"       id="text136"><tspan         sodipodi:role="line"         id="tspan134"         x="72.293335"         y="74.079803"         style="stroke-width:0.08422855"><tspan           style="font-size:3.32617211px;stroke-width:0.08422855"           id="tspan130">w</tspan><tspan           style="font-size:2.18994236px;baseline-shift:sub;stroke-width:0.08422855"           id="tspan132">2</tspan></tspan></text>    <circle       r="4.4546947"       cy="111.91642"       cx="86.479889"       id="circle138"       style="fill:none;stroke:#000000;stroke-width:0.13880876;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />    <text       id="text142"       y="113.17245"       x="83.291779"       style="font-style:normal;font-weight:normal;font-size:3.36914182px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.08422855"       xml:space="preserve"><tspan         style="font-size:3.32617211px;stroke-width:0.08422855"         y="113.17245"         x="83.291779"         id="tspan140"         sodipodi:role="line">relu</tspan></text>    <g       id="g150"       transform="matrix(0.2682318,0,0,0.21903975,80.650296,58.630923)"       style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none">      <g         id="g148"         transform="translate(0,-1.2654371)"         style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none">        <path           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"           d="m 21.733631,93.643839 v 17.958841 l -2.834822,-7.18155"           id="path144"           inkscape:connector-curvature="0"           sodipodi:nodetypes="ccc" />        <path           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"           d="m 21.733631,111.60268 2.834822,-7.18155"           id="path146"           inkscape:connector-curvature="0"           sodipodi:nodetypes="cc" />      </g>    </g>    <g       style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none"       transform="matrix(0.2682318,0,0,0.21903975,80.650296,44.217509)"       id="g158">      <g         style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none"         transform="translate(0,-1.2654371)"         id="g156">        <path           sodipodi:nodetypes="ccc"           inkscape:connector-curvature="0"           id="path152"           d="m 21.733631,93.643839 v 17.958841 l -2.834822,-7.18155"           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />        <path           sodipodi:nodetypes="cc"           inkscape:connector-curvature="0"           id="path154"           d="m 21.733631,111.60268 2.834822,-7.18155"           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />      </g>    </g>    <g       style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none"       transform="matrix(0,-0.2682318,0.21903975,0,56.987412,79.336874)"       id="g166">      <g         style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none"         transform="translate(0,-1.2654371)"         id="g164">        <path           sodipodi:nodetypes="ccc"           inkscape:connector-curvature="0"           id="path160"           d="m 21.733631,93.643839 v 17.958841 l -2.834822,-7.18155"           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />        <path           sodipodi:nodetypes="cc"           inkscape:connector-curvature="0"           id="path162"           d="m 21.733631,111.60268 2.834822,-7.18155"           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />      </g>    </g>    <g       style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none"       transform="matrix(0.2682318,0,0,0.21903975,80.650296,97.159212)"       id="g174">      <g         style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none"         transform="translate(0,-1.2654371)"         id="g172">        <path           sodipodi:nodetypes="ccc"           inkscape:connector-curvature="0"           id="path168"           d="m 21.733631,93.643839 v 17.958841 l -2.834822,-7.18155"           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />        <path           sodipodi:nodetypes="cc"           inkscape:connector-curvature="0"           id="path170"           d="m 21.733631,111.60268 2.834822,-7.18155"           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />      </g>    </g>    <text       xml:space="preserve"       style="font-style:normal;font-weight:normal;font-size:3.36914206px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.08422855"       x="83.401512"       y="86.562973"       id="text178"><tspan         id="tspan176"         sodipodi:role="line"         x="83.401512"         y="86.562973"         style="font-size:3.32617211px;stroke-width:0.08422855">F(x)</tspan></text>    <circle       r="4.4546947"       cy="97.34465"       cx="86.479889"       id="circle180"       style="fill:none;stroke:#000000;stroke-width:0.13880876;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />    <text       id="text184"       y="98.387321"       x="82.969742"       style="font-style:normal;font-weight:normal;font-size:3.36914182px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.08422855"       xml:space="preserve"><tspan         style="font-size:3.32617211px;stroke-width:0.08422855"         y="98.387321"         x="82.969742"         id="tspan182"         sodipodi:role="line">sum</tspan></text>    <g       style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none"       transform="matrix(0.2682318,0,0,0.21903975,80.650296,82.468557)"       id="g192">      <g         style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none"         transform="translate(0,-1.2654371)"         id="g190">        <path           sodipodi:nodetypes="ccc"           inkscape:connector-curvature="0"           id="path186"           d="m 21.733631,93.643839 v 17.958841 l -2.834822,-7.18155"           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />        <path           sodipodi:nodetypes="cc"           inkscape:connector-curvature="0"           id="path188"           d="m 21.733631,111.60268 2.834822,-7.18155"           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />      </g>    </g>    <g       id="g233"       style="stroke-width:0.25549346;stroke-miterlimit:4;stroke-dasharray:none"       transform="matrix(0.52366914,0,0,0.56365995,14.187089,14.107981)">      <path         sodipodi:nodetypes="cc"         inkscape:connector-curvature="0"         id="path194"         d="m 142.87502,34.117558 c 36.6184,11.088421 40.00279,89.849732 7.74851,112.258982"         style="fill:none;stroke:#000000;stroke-width:0.25549346;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />      <path         style="fill:none;stroke:#000000;stroke-width:0.25549346;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"         d="m 150.62353,146.37654 3.46616,-0.29671"         id="path188-3"         inkscape:connector-curvature="0"         sodipodi:nodetypes="cc" />      <path         style="fill:none;stroke:#000000;stroke-width:0.25549346;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1"         d="m 150.62353,146.37654 1.38523,-2.98176"         id="path188-3-6"         inkscape:connector-curvature="0"         sodipodi:nodetypes="cc" />    </g>    <text       xml:space="preserve"       style="font-style:normal;font-weight:normal;font-size:3.36914206px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.08422855"       x="73.706299"       y="125.38822"       id="text228"><tspan         id="tspan226"         sodipodi:role="line"         x="73.706299"         y="125.38822"         style="font-size:3.32617211px;stroke-width:0.08422855">H(x) = F(x) + x</tspan></text>    <g       style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none"       transform="matrix(0.2682318,0,0,0.21903975,80.631922,67.777902)"       id="g1046">      <g         style="stroke-width:0.57266456;stroke-miterlimit:4;stroke-dasharray:none"         transform="translate(0,-1.2654371)"         id="g1044">        <path           sodipodi:nodetypes="ccc"           inkscape:connector-curvature="0"           id="path1040"           d="m 21.733631,93.643839 v 17.958841 l -2.834822,-7.18155"           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />        <path           sodipodi:nodetypes="cc"           inkscape:connector-curvature="0"           id="path1042"           d="m 21.733631,111.60268 2.834822,-7.18155"           style="fill:none;stroke:#000000;stroke-width:0.57266456;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" />      </g>    </g>  </g></svg>\" width=\"360\">\n",
    "\n",
    "_(Left: layers that learn an unreferenced function $H(x)$, right: layers that learn a residual function $F(x)$ with reference to the input)_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Residual networks have _shortcut connections_ (those that skip one or more layers) that perform identity mapping, with their outputs added element-wise to the outputs of skipped layers. These connections introduce no additional parameters to be learned.\n",
    "\n",
    "Residual functions are learned by two or more layers. If a function is learned by a single layer, $F(x) + x$ becomes $w^T x + x$, for which the authors remark to have observed no advantages compared to a plain layer. Experiments in the paper involve functions that have two or three layers. Layers can be convolutional, in which case \"element-wise addition is performed on two feature maps, channel by channel\".\n",
    "\n",
    "## ResNet\n",
    "\n",
    "Several residual networks were evaluated in the paper:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img width=\"800\" src=\"data:image/png;base64,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\">"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The input is a $224 \\times 224$ image. Downsampling is performed by convolution layers with a stride of 2 -- the first layer in the first block of `conv3_x`, the first layer in the first block of `conv4_x`, the first layer in the first block of `conv5_x`.\n",
    "\n",
    "Deeper networks (50+ layers) use blocks with three instead of two layers, arranged in what the authors call _bottleneck design_. The 1x1 layers are responsible for reducing and then increasing (restoring) dimensions. This leads to smaller dimensions for the 3x3 layer, improving computation time.\n",
    "\n",
    "### Training\n",
    "\n",
    "_Batch normalization_ is applied after each convolution (before the activation). Models are trained on batches of 256 images using a _stochastic gradient descent_ with a fixed momentum of 0.9 and an adaptive learning rate that starts at 0.1 and is divided by 10 when the error plateaus. A weight decay of 0.0001 is added for regularization. Dropout is not used.\n",
    "\n",
    "As an example, here's the 50-layer model in Keras:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "input_1 (InputLayer)            (None, 224, 224, 3)  0                                            \n",
      "__________________________________________________________________________________________________\n",
      "conv1 (Conv2D)                  (None, 112, 112, 64) 9472        input_1[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "bn_conv1 (BatchNormalization)   (None, 112, 112, 64) 256         conv1[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "activation_1 (Activation)       (None, 112, 112, 64) 0           bn_conv1[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2D)  (None, 55, 55, 64)   0           activation_1[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "res2a_branch2a (Conv2D)         (None, 55, 55, 64)   4160        max_pooling2d_1[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "bn2a_branch2a (BatchNormalizati (None, 55, 55, 64)   256         res2a_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_2 (Activation)       (None, 55, 55, 64)   0           bn2a_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res2a_branch2b (Conv2D)         (None, 55, 55, 64)   36928       activation_2[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "bn2a_branch2b (BatchNormalizati (None, 55, 55, 64)   256         res2a_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_3 (Activation)       (None, 55, 55, 64)   0           bn2a_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res2a_branch2c (Conv2D)         (None, 55, 55, 256)  16640       activation_3[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "res2a_branch1 (Conv2D)          (None, 55, 55, 256)  16640       max_pooling2d_1[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "bn2a_branch2c (BatchNormalizati (None, 55, 55, 256)  1024        res2a_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "bn2a_branch1 (BatchNormalizatio (None, 55, 55, 256)  1024        res2a_branch1[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "add_1 (Add)                     (None, 55, 55, 256)  0           bn2a_branch2c[0][0]              \n",
      "                                                                 bn2a_branch1[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "activation_4 (Activation)       (None, 55, 55, 256)  0           add_1[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "res2b_branch2a (Conv2D)         (None, 55, 55, 64)   16448       activation_4[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "bn2b_branch2a (BatchNormalizati (None, 55, 55, 64)   256         res2b_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_5 (Activation)       (None, 55, 55, 64)   0           bn2b_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res2b_branch2b (Conv2D)         (None, 55, 55, 64)   36928       activation_5[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "bn2b_branch2b (BatchNormalizati (None, 55, 55, 64)   256         res2b_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_6 (Activation)       (None, 55, 55, 64)   0           bn2b_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res2b_branch2c (Conv2D)         (None, 55, 55, 256)  16640       activation_6[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "bn2b_branch2c (BatchNormalizati (None, 55, 55, 256)  1024        res2b_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_2 (Add)                     (None, 55, 55, 256)  0           bn2b_branch2c[0][0]              \n",
      "                                                                 activation_4[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "activation_7 (Activation)       (None, 55, 55, 256)  0           add_2[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "res2c_branch2a (Conv2D)         (None, 55, 55, 64)   16448       activation_7[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "bn2c_branch2a (BatchNormalizati (None, 55, 55, 64)   256         res2c_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_8 (Activation)       (None, 55, 55, 64)   0           bn2c_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res2c_branch2b (Conv2D)         (None, 55, 55, 64)   36928       activation_8[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "bn2c_branch2b (BatchNormalizati (None, 55, 55, 64)   256         res2c_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_9 (Activation)       (None, 55, 55, 64)   0           bn2c_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res2c_branch2c (Conv2D)         (None, 55, 55, 256)  16640       activation_9[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "bn2c_branch2c (BatchNormalizati (None, 55, 55, 256)  1024        res2c_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_3 (Add)                     (None, 55, 55, 256)  0           bn2c_branch2c[0][0]              \n",
      "                                                                 activation_7[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "activation_10 (Activation)      (None, 55, 55, 256)  0           add_3[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "res3a_branch2a (Conv2D)         (None, 28, 28, 128)  32896       activation_10[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3a_branch2a (BatchNormalizati (None, 28, 28, 128)  512         res3a_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_11 (Activation)      (None, 28, 28, 128)  0           bn3a_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res3a_branch2b (Conv2D)         (None, 28, 28, 128)  147584      activation_11[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3a_branch2b (BatchNormalizati (None, 28, 28, 128)  512         res3a_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_12 (Activation)      (None, 28, 28, 128)  0           bn3a_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res3a_branch2c (Conv2D)         (None, 28, 28, 512)  66048       activation_12[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res3a_branch1 (Conv2D)          (None, 28, 28, 512)  131584      activation_10[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3a_branch2c (BatchNormalizati (None, 28, 28, 512)  2048        res3a_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "bn3a_branch1 (BatchNormalizatio (None, 28, 28, 512)  2048        res3a_branch1[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "add_4 (Add)                     (None, 28, 28, 512)  0           bn3a_branch2c[0][0]              \n",
      "                                                                 bn3a_branch1[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "activation_13 (Activation)      (None, 28, 28, 512)  0           add_4[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "res3b_branch2a (Conv2D)         (None, 28, 28, 128)  65664       activation_13[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3b_branch2a (BatchNormalizati (None, 28, 28, 128)  512         res3b_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_14 (Activation)      (None, 28, 28, 128)  0           bn3b_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res3b_branch2b (Conv2D)         (None, 28, 28, 128)  147584      activation_14[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3b_branch2b (BatchNormalizati (None, 28, 28, 128)  512         res3b_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_15 (Activation)      (None, 28, 28, 128)  0           bn3b_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res3b_branch2c (Conv2D)         (None, 28, 28, 512)  66048       activation_15[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3b_branch2c (BatchNormalizati (None, 28, 28, 512)  2048        res3b_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_5 (Add)                     (None, 28, 28, 512)  0           bn3b_branch2c[0][0]              \n",
      "                                                                 activation_13[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_16 (Activation)      (None, 28, 28, 512)  0           add_5[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "res3c_branch2a (Conv2D)         (None, 28, 28, 128)  65664       activation_16[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3c_branch2a (BatchNormalizati (None, 28, 28, 128)  512         res3c_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_17 (Activation)      (None, 28, 28, 128)  0           bn3c_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res3c_branch2b (Conv2D)         (None, 28, 28, 128)  147584      activation_17[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3c_branch2b (BatchNormalizati (None, 28, 28, 128)  512         res3c_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_18 (Activation)      (None, 28, 28, 128)  0           bn3c_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res3c_branch2c (Conv2D)         (None, 28, 28, 512)  66048       activation_18[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3c_branch2c (BatchNormalizati (None, 28, 28, 512)  2048        res3c_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_6 (Add)                     (None, 28, 28, 512)  0           bn3c_branch2c[0][0]              \n",
      "                                                                 activation_16[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_19 (Activation)      (None, 28, 28, 512)  0           add_6[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "res3d_branch2a (Conv2D)         (None, 28, 28, 128)  65664       activation_19[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3d_branch2a (BatchNormalizati (None, 28, 28, 128)  512         res3d_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_20 (Activation)      (None, 28, 28, 128)  0           bn3d_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res3d_branch2b (Conv2D)         (None, 28, 28, 128)  147584      activation_20[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3d_branch2b (BatchNormalizati (None, 28, 28, 128)  512         res3d_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_21 (Activation)      (None, 28, 28, 128)  0           bn3d_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res3d_branch2c (Conv2D)         (None, 28, 28, 512)  66048       activation_21[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3d_branch2c (BatchNormalizati (None, 28, 28, 512)  2048        res3d_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_7 (Add)                     (None, 28, 28, 512)  0           bn3d_branch2c[0][0]              \n",
      "                                                                 activation_19[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_22 (Activation)      (None, 28, 28, 512)  0           add_7[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "res4a_branch2a (Conv2D)         (None, 14, 14, 256)  131328      activation_22[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4a_branch2a (BatchNormalizati (None, 14, 14, 256)  1024        res4a_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_23 (Activation)      (None, 14, 14, 256)  0           bn4a_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4a_branch2b (Conv2D)         (None, 14, 14, 256)  590080      activation_23[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4a_branch2b (BatchNormalizati (None, 14, 14, 256)  1024        res4a_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_24 (Activation)      (None, 14, 14, 256)  0           bn4a_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4a_branch2c (Conv2D)         (None, 14, 14, 1024) 263168      activation_24[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4a_branch1 (Conv2D)          (None, 14, 14, 1024) 525312      activation_22[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4a_branch2c (BatchNormalizati (None, 14, 14, 1024) 4096        res4a_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "bn4a_branch1 (BatchNormalizatio (None, 14, 14, 1024) 4096        res4a_branch1[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "add_8 (Add)                     (None, 14, 14, 1024) 0           bn4a_branch2c[0][0]              \n",
      "                                                                 bn4a_branch1[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "activation_25 (Activation)      (None, 14, 14, 1024) 0           add_8[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "res4b_branch2a (Conv2D)         (None, 14, 14, 256)  262400      activation_25[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4b_branch2a (BatchNormalizati (None, 14, 14, 256)  1024        res4b_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_26 (Activation)      (None, 14, 14, 256)  0           bn4b_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4b_branch2b (Conv2D)         (None, 14, 14, 256)  590080      activation_26[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4b_branch2b (BatchNormalizati (None, 14, 14, 256)  1024        res4b_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_27 (Activation)      (None, 14, 14, 256)  0           bn4b_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4b_branch2c (Conv2D)         (None, 14, 14, 1024) 263168      activation_27[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4b_branch2c (BatchNormalizati (None, 14, 14, 1024) 4096        res4b_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_9 (Add)                     (None, 14, 14, 1024) 0           bn4b_branch2c[0][0]              \n",
      "                                                                 activation_25[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_28 (Activation)      (None, 14, 14, 1024) 0           add_9[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "res4c_branch2a (Conv2D)         (None, 14, 14, 256)  262400      activation_28[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4c_branch2a (BatchNormalizati (None, 14, 14, 256)  1024        res4c_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_29 (Activation)      (None, 14, 14, 256)  0           bn4c_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4c_branch2b (Conv2D)         (None, 14, 14, 256)  590080      activation_29[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4c_branch2b (BatchNormalizati (None, 14, 14, 256)  1024        res4c_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_30 (Activation)      (None, 14, 14, 256)  0           bn4c_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4c_branch2c (Conv2D)         (None, 14, 14, 1024) 263168      activation_30[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4c_branch2c (BatchNormalizati (None, 14, 14, 1024) 4096        res4c_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_10 (Add)                    (None, 14, 14, 1024) 0           bn4c_branch2c[0][0]              \n",
      "                                                                 activation_28[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_31 (Activation)      (None, 14, 14, 1024) 0           add_10[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "res4d_branch2a (Conv2D)         (None, 14, 14, 256)  262400      activation_31[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4d_branch2a (BatchNormalizati (None, 14, 14, 256)  1024        res4d_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_32 (Activation)      (None, 14, 14, 256)  0           bn4d_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4d_branch2b (Conv2D)         (None, 14, 14, 256)  590080      activation_32[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4d_branch2b (BatchNormalizati (None, 14, 14, 256)  1024        res4d_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_33 (Activation)      (None, 14, 14, 256)  0           bn4d_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4d_branch2c (Conv2D)         (None, 14, 14, 1024) 263168      activation_33[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4d_branch2c (BatchNormalizati (None, 14, 14, 1024) 4096        res4d_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_11 (Add)                    (None, 14, 14, 1024) 0           bn4d_branch2c[0][0]              \n",
      "                                                                 activation_31[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_34 (Activation)      (None, 14, 14, 1024) 0           add_11[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "res4e_branch2a (Conv2D)         (None, 14, 14, 256)  262400      activation_34[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4e_branch2a (BatchNormalizati (None, 14, 14, 256)  1024        res4e_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_35 (Activation)      (None, 14, 14, 256)  0           bn4e_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4e_branch2b (Conv2D)         (None, 14, 14, 256)  590080      activation_35[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4e_branch2b (BatchNormalizati (None, 14, 14, 256)  1024        res4e_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_36 (Activation)      (None, 14, 14, 256)  0           bn4e_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4e_branch2c (Conv2D)         (None, 14, 14, 1024) 263168      activation_36[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4e_branch2c (BatchNormalizati (None, 14, 14, 1024) 4096        res4e_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_12 (Add)                    (None, 14, 14, 1024) 0           bn4e_branch2c[0][0]              \n",
      "                                                                 activation_34[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_37 (Activation)      (None, 14, 14, 1024) 0           add_12[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "res4f_branch2a (Conv2D)         (None, 14, 14, 256)  262400      activation_37[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4f_branch2a (BatchNormalizati (None, 14, 14, 256)  1024        res4f_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_38 (Activation)      (None, 14, 14, 256)  0           bn4f_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4f_branch2b (Conv2D)         (None, 14, 14, 256)  590080      activation_38[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4f_branch2b (BatchNormalizati (None, 14, 14, 256)  1024        res4f_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_39 (Activation)      (None, 14, 14, 256)  0           bn4f_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4f_branch2c (Conv2D)         (None, 14, 14, 1024) 263168      activation_39[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4f_branch2c (BatchNormalizati (None, 14, 14, 1024) 4096        res4f_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_13 (Add)                    (None, 14, 14, 1024) 0           bn4f_branch2c[0][0]              \n",
      "                                                                 activation_37[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_40 (Activation)      (None, 14, 14, 1024) 0           add_13[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "res5a_branch2a (Conv2D)         (None, 7, 7, 512)    524800      activation_40[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn5a_branch2a (BatchNormalizati (None, 7, 7, 512)    2048        res5a_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_41 (Activation)      (None, 7, 7, 512)    0           bn5a_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res5a_branch2b (Conv2D)         (None, 7, 7, 512)    2359808     activation_41[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn5a_branch2b (BatchNormalizati (None, 7, 7, 512)    2048        res5a_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_42 (Activation)      (None, 7, 7, 512)    0           bn5a_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res5a_branch2c (Conv2D)         (None, 7, 7, 2048)   1050624     activation_42[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res5a_branch1 (Conv2D)          (None, 7, 7, 2048)   2099200     activation_40[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn5a_branch2c (BatchNormalizati (None, 7, 7, 2048)   8192        res5a_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "bn5a_branch1 (BatchNormalizatio (None, 7, 7, 2048)   8192        res5a_branch1[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "add_14 (Add)                    (None, 7, 7, 2048)   0           bn5a_branch2c[0][0]              \n",
      "                                                                 bn5a_branch1[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "activation_43 (Activation)      (None, 7, 7, 2048)   0           add_14[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "res5b_branch2a (Conv2D)         (None, 7, 7, 512)    1049088     activation_43[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn5b_branch2a (BatchNormalizati (None, 7, 7, 512)    2048        res5b_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_44 (Activation)      (None, 7, 7, 512)    0           bn5b_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res5b_branch2b (Conv2D)         (None, 7, 7, 512)    2359808     activation_44[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn5b_branch2b (BatchNormalizati (None, 7, 7, 512)    2048        res5b_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_45 (Activation)      (None, 7, 7, 512)    0           bn5b_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res5b_branch2c (Conv2D)         (None, 7, 7, 2048)   1050624     activation_45[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn5b_branch2c (BatchNormalizati (None, 7, 7, 2048)   8192        res5b_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_15 (Add)                    (None, 7, 7, 2048)   0           bn5b_branch2c[0][0]              \n",
      "                                                                 activation_43[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_46 (Activation)      (None, 7, 7, 2048)   0           add_15[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "res5c_branch2a (Conv2D)         (None, 7, 7, 512)    1049088     activation_46[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn5c_branch2a (BatchNormalizati (None, 7, 7, 512)    2048        res5c_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_47 (Activation)      (None, 7, 7, 512)    0           bn5c_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res5c_branch2b (Conv2D)         (None, 7, 7, 512)    2359808     activation_47[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn5c_branch2b (BatchNormalizati (None, 7, 7, 512)    2048        res5c_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_48 (Activation)      (None, 7, 7, 512)    0           bn5c_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res5c_branch2c (Conv2D)         (None, 7, 7, 2048)   1050624     activation_48[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn5c_branch2c (BatchNormalizati (None, 7, 7, 2048)   8192        res5c_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_16 (Add)                    (None, 7, 7, 2048)   0           bn5c_branch2c[0][0]              \n",
      "                                                                 activation_46[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_49 (Activation)      (None, 7, 7, 2048)   0           add_16[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "avg_pool (AveragePooling2D)     (None, 1, 1, 2048)   0           activation_49[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "flatten_1 (Flatten)             (None, 2048)         0           avg_pool[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "fc1000 (Dense)                  (None, 1000)         2049000     flatten_1[0][0]                  \n",
      "==================================================================================================\n",
      "Total params: 25,636,712\n",
      "Trainable params: 25,583,592\n",
      "Non-trainable params: 53,120\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "\n",
    "model = tf.keras.applications.ResNet50(weights=None)\n",
    "model.summary()"
   ]
  }
 ],
 "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
}