{ "cells": [ { "cell_type": "code", "execution_count": 1, "metadata": { "tags": [ "remove-cell" ] }, "outputs": [], "source": [ "import sys\n", "\n", "sys.path.insert(1, \"../../\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Quadratic Assignment Problem\n", "\n", "Quadratic assignment problem (QAP) is the following problem.\n", "\n", "{card} Quadratic assignment problem\n", "Let $N$ be a positive integer. Consider $N$ factories to be built on $N$ candidate sites. Each factory can be built on any of the candidate sites. Every two factories have trucks traveling to and from them, and their transportation volumes are known in advance. How can we minimize the sum of the amount transported x the distance traveled?\n", "\n", "\n", "An application could be to determine the seating chart for a meeting so that people close to each other have seats closely.\n", "\n", "## Formulation\n", "\n", "Let $N$ potential factory locations be denoted by land $0$, land $1$, ... , and $N$ factories are denoted as factories $0$, factories $1$, ..., factories $N-1$. Also let $D_{i, j}$ denote the distance between land $i$ and land $j$, and $F_{k, l}$ denote the transport volume between factory $k$ and factory $l$.\n", "\n", "### Variables\n", "\n", "With $N \\times N$ binary variables $q$, let $q_{i, k}$ represent whether factory $k$ is to be built on land $i$.\n", "\n", "For example, factory $3$ will be built on land $0$ if $q$ has the following value.\n", "\n", "{csv-table} Binary variable table\n", ":header-rows: 1\n", ":stub-columns: 1\n", "\n", "\"\", \"factory 0\", \"factory 1\", \"factory 2\", \"factory 3\", \"factory 4\"\n", "\"land 0\", 0, 0, 0, 1, 0\n", "\"land 1\", 0, 1, 0, 0, 0\n", "\"land 2\", 0, 0, 0, 0, 1\n", "\"land 3\", 1, 0, 0, 0, 0\n", "\"land 4\", 0, 0, 1, 0, 0\n", "\n", "\n", "### Constraints\n", "\n", "Each row and column of the binary variable table must have exactly one variable that is 1, so we place a one-hot constraint on each row and column. Conversely, if these are satisfied, then there is only one way to determine which factory to build on which land.\n", "\n", "### Objective function\n", "\n", "The objective function is the sum of transport volume x distance between factories. This can be expressed in the equation using $q$ as follows.\n", "\n", "$$\n", "\\sum_{q_{i, k} = 1, q_{j, l} = 1} D_{i, j} \\ F_{k, l} = \\sum_{i, j, k, l} q_{i, k} \\ q_{j, l} \\ D_{i, j} \\ F_{k, l}\n", "$$\n", "\n", "### Formulation\n", "\n", "The above formulation, with $N\\times N$ binary variables $q$, can be written as follows.\n", "\n", "\n", "\\begin{align}\n", "\\text{minimize} \\quad &\\sum_{i, j, k, l} q_{i, k} \\ q_{j, l} \\ D_{i, j} \\ F_{k, l} \\\\\n", "\\text{subject to} \\quad &\\sum_k q_{i, k} = 1 \\quad \\text{for} \\quad i \\in \\{0, 1, \\ldots, N - 1\\}, \\\\ \n", " &\\sum_i q_{i, k} = 1 \\quad \\text{for} \\quad k \\in \\{0, 1, \\ldots, N - 1\\}, \\\\\n", " &q_{i, k} \\in \\{0, 1\\} \\quad \\text{for} \\quad i, k \\in \\{0, 1, \\ldots, N - 1\\}.\n", "\\end{align}\n", "\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Problem setting\n", "\n", "Before formulating with the Amplify SDK, we will create a problem. For simplicity, let the number of factories $N=10$." ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "import numpy as np" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "N = 10" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Next, we create a distance matrix $D$ representing the distances between lands. The lands are randomly generated on the Euclidean plane. The distance matrix is created as a two-dimensional {py:class}numpy.ndarray and rounded to integers for ease of presentation. The name of the matrix is distance." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[[ 0 82 86 78 109 52 52 35 19 12]\n", " [ 82 0 20 58 42 78 48 48 68 71]\n", " [ 86 20 0 77 60 93 63 55 75 76]\n", " [ 78 58 77 0 47 37 27 52 59 66]\n", " [109 42 60 47 0 83 60 75 91 96]\n", " [ 52 78 93 37 83 0 29 44 37 43]\n", " [ 52 48 63 27 60 29 0 25 33 39]\n", " [ 35 48 55 52 75 44 25 0 20 23]\n", " [ 19 68 75 59 91 37 33 20 0 6]\n", " [ 12 71 76 66 96 43 39 23 6 0]]\n" ] } ], "source": [ "rng = np.random.default_rng()\n", "\n", "x = rng.integers(0, 100, size=(N,))\n", "y = rng.integers(0, 100, size=(N,))\n", "\n", "distance = (\n", " (\n", " (x[:, np.newaxis] - x[np.newaxis, :]) ** 2\n", " + (y[:, np.newaxis] - y[np.newaxis, :]) ** 2\n", " )\n", " ** 0.5\n", ").astype(int)\n", "\n", "print(distance)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Also, we create a matrix $F$ representing the amount of transport between factories, a random symmetric matrix of dimension 2, named flow." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[[ 0 74 17 38 55 71 67 46 47 7]\n", " [74 0 52 17 94 86 84 66 65 40]\n", " [17 52 0 28 26 60 3 3 90 98]\n", " [38 17 28 0 7 89 92 5 43 97]\n", " [55 94 26 7 0 94 51 46 29 3]\n", " [71 86 60 89 94 0 90 29 66 10]\n", " [67 84 3 92 51 90 0 74 85 26]\n", " [46 66 3 5 46 29 74 0 87 6]\n", " [47 65 90 43 29 66 85 87 0 57]\n", " [ 7 40 98 97 3 10 26 6 57 0]]\n" ] } ], "source": [ "flow = np.zeros((N, N), dtype=int)\n", "for i in range(N):\n", " for j in range(i + 1, N):\n", " flow[i, j] = flow[j, i] = rng.integers(0, 100)\n", "\n", "print(flow)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Formulation with the Amplify SDK\n", "\n", "In the formulation, we can use the {py:class}~amplify.Matrix class for efficient formulation, since a quadratic term consisting of any two binary variables can appear in the objective function.\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Creating variables\n", "\n", "To formulate using the {py:class}~amplify.Matrix class, {py:class}~amplify.VariableGenerator's {py:meth}~amplify.VariableGenerator.matrix method to issue variables." ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "text/latex": [ "\\displaystyle \\begin{aligned}&\\left[\\begin{matrix}q_{0,0}& q_{0,1}& q_{0,2}& q_{0,3}& q_{0,4}& q_{0,5}& q_{0,6}& q_{0,7}& \n", " q_{0,8}& q_{0,9}\\\\q_{1,0}& q_{1,1}& q_{1,2}& q_{1,3}& q_{1,4}& q_{1,5}& q_{1,6}& q_{1,7}& \n", " q_{1,8}& q_{1,9}\\\\q_{2,0}& q_{2,1}& q_{2,2}& q_{2,3}& q_{2,4}& q_{2,5}& q_{2,6}& q_{2,7}& \n", " q_{2,8}& q_{2,9}\\\\q_{3,0}& q_{3,1}& q_{3,2}& q_{3,3}& q_{3,4}& q_{3,5}& q_{3,6}& q_{3,7}& \n", " q_{3,8}& q_{3,9}\\\\q_{4,0}& q_{4,1}& q_{4,2}& q_{4,3}& q_{4,4}& q_{4,5}& q_{4,6}& q_{4,7}& \n", " q_{4,8}& q_{4,9}\\\\q_{5,0}& q_{5,1}& q_{5,2}& q_{5,3}& q_{5,4}& q_{5,5}& q_{5,6}& q_{5,7}& \n", " q_{5,8}& q_{5,9}\\\\q_{6,0}& q_{6,1}& q_{6,2}& q_{6,3}& q_{6,4}& q_{6,5}& q_{6,6}& q_{6,7}& \n", " q_{6,8}& q_{6,9}\\\\q_{7,0}& q_{7,1}& q_{7,2}& q_{7,3}& q_{7,4}& q_{7,5}& q_{7,6}& q_{7,7}& \n", " q_{7,8}& q_{7,9}\\\\q_{8,0}& q_{8,1}& q_{8,2}& q_{8,3}& q_{8,4}& q_{8,5}& q_{8,6}& q_{8,7}& \n", " q_{8,8}& q_{8,9}\\\\q_{9,0}& q_{9,1}& q_{9,2}& q_{9,3}& q_{9,4}& q_{9,5}& q_{9,6}& q_{9,7}& \n", " q_{9,8}& q_{9,9}\\end{matrix}\\right]\\end{aligned}" ], "text/plain": [ "PolyArray([[q_{0,0}, q_{0,1}, q_{0,2}, q_{0,3}, q_{0,4}, q_{0,5}, q_{0,6}, q_{0,7}, \n", " q_{0,8}, q_{0,9}],\n", " [q_{1,0}, q_{1,1}, q_{1,2}, q_{1,3}, q_{1,4}, q_{1,5}, q_{1,6}, q_{1,7}, \n", " q_{1,8}, q_{1,9}],\n", " [q_{2,0}, q_{2,1}, q_{2,2}, q_{2,3}, q_{2,4}, q_{2,5}, q_{2,6}, q_{2,7}, \n", " q_{2,8}, q_{2,9}],\n", " [q_{3,0}, q_{3,1}, q_{3,2}, q_{3,3}, q_{3,4}, q_{3,5}, q_{3,6}, q_{3,7}, \n", " q_{3,8}, q_{3,9}],\n", " [q_{4,0}, q_{4,1}, q_{4,2}, q_{4,3}, q_{4,4}, q_{4,5}, q_{4,6}, q_{4,7}, \n", " q_{4,8}, q_{4,9}],\n", " [q_{5,0}, q_{5,1}, q_{5,2}, q_{5,3}, q_{5,4}, q_{5,5}, q_{5,6}, q_{5,7}, \n", " q_{5,8}, q_{5,9}],\n", " [q_{6,0}, q_{6,1}, q_{6,2}, q_{6,3}, q_{6,4}, q_{6,5}, q_{6,6}, q_{6,7}, \n", " q_{6,8}, q_{6,9}],\n", " [q_{7,0}, q_{7,1}, q_{7,2}, q_{7,3}, q_{7,4}, q_{7,5}, q_{7,6}, q_{7,7}, \n", " q_{7,8}, q_{7,9}],\n", " [q_{8,0}, q_{8,1}, q_{8,2}, q_{8,3}, q_{8,4}, q_{8,5}, q_{8,6}, q_{8,7}, \n", " q_{8,8}, q_{8,9}],\n", " [q_{9,0}, q_{9,1}, q_{9,2}, q_{9,3}, q_{9,4}, q_{9,5}, q_{9,6}, q_{9,7}, \n", " q_{9,8}, q_{9,9}]])" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from amplify import VariableGenerator\n", "\n", "gen = VariableGenerator()\n", "matrix = gen.matrix(\"Binary\", N, N) # coefficient matrix\n", "q = matrix.variable_array # variables\n", "\n", "q" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Creating the objective function\n", "\n", "The matrix created above is an instance of the class {py:class}~amplify.Matrix, which has the following three properties.\n", "\n", "* {py:attr}~amplify.Matrix.quadratic\n", "* {py:attr}~amplify.Matrix.linear\n", "* {py:attr}~amplify.Matrix.constant\n", "\n", "{py:attr}~amplify.Matrix.quadratic is {py:class}numpy.ndarray representing the coefficients of the second order terms, and its {py:attr}~numpy.ndarray.shape is (N, N, N, N) this time. quadratic[i, k, j, l]{l=python} corresponds to the coefficients of q[i, k] * q[j, l]{l=python}. That is, {py:attr}~amplify.Matrix.quadratic must be set to a 4-dimensional NumPy array such that quadratic[i, k, j, l] = distance[i, j] * flow[k, l]{l=python}\n", "\n", "{py:attr}~amplify.Matrix.linear and {py:attr}~amplify.Matrix.constant represent the coefficient and constant terms of the linear term, respectively, but since the objective function used in this problem contains only second order terms, we will not set them." ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "tags": [ "remove-output" ] }, "outputs": [ { "data": { "text/plain": [ "array([[[[ 0., 0., 0., ..., 0., 0., 0.],\n", " [ 0., 6068., 1394., ..., 3772., 3854., 574.],\n", " [ 0., 6364., 1462., ..., 3956., 4042., 602.],\n", " ...,\n", " [ 0., 2590., 595., ..., 1610., 1645., 245.],\n", " [ 0., 1406., 323., ..., 874., 893., 133.],\n", " [ 0., 888., 204., ..., 552., 564., 84.]],\n", "\n", " [[ 0., 0., 0., ..., 0., 0., 0.],\n", " [6068., 0., 4264., ..., 5412., 5330., 3280.],\n", " [6364., 0., 4472., ..., 5676., 5590., 3440.],\n", " ...,\n", " [2590., 0., 1820., ..., 2310., 2275., 1400.],\n", " [1406., 0., 988., ..., 1254., 1235., 760.],\n", " [ 888., 0., 624., ..., 792., 780., 480.]],\n", "\n", " [[ 0., 0., 0., ..., 0., 0., 0.],\n", " [1394., 4264., 0., ..., 246., 7380., 8036.],\n", " [1462., 4472., 0., ..., 258., 7740., 8428.],\n", " ...,\n", " [ 595., 1820., 0., ..., 105., 3150., 3430.],\n", " [ 323., 988., 0., ..., 57., 1710., 1862.],\n", " [ 204., 624., 0., ..., 36., 1080., 1176.]],\n", "\n", " ...,\n", "\n", " [[ 0., 0., 0., ..., 0., 0., 0.],\n", " [3772., 5412., 246., ..., 0., 7134., 492.],\n", " [3956., 5676., 258., ..., 0., 7482., 516.],\n", " ...,\n", " [1610., 2310., 105., ..., 0., 3045., 210.],\n", " [ 874., 1254., 57., ..., 0., 1653., 114.],\n", " [ 552., 792., 36., ..., 0., 1044., 72.]],\n", "\n", " [[ 0., 0., 0., ..., 0., 0., 0.],\n", " [3854., 5330., 7380., ..., 7134., 0., 4674.],\n", " [4042., 5590., 7740., ..., 7482., 0., 4902.],\n", " ...,\n", " [1645., 2275., 3150., ..., 3045., 0., 1995.],\n", " [ 893., 1235., 1710., ..., 1653., 0., 1083.],\n", " [ 564., 780., 1080., ..., 1044., 0., 684.]],\n", "\n", " [[ 0., 0., 0., ..., 0., 0., 0.],\n", " [ 574., 3280., 8036., ..., 492., 4674., 0.],\n", " [ 602., 3440., 8428., ..., 516., 4902., 0.],\n", " ...,\n", " [ 245., 1400., 3430., ..., 210., 1995., 0.],\n", " [ 133., 760., 1862., ..., 114., 1083., 0.],\n", " [ 84., 480., 1176., ..., 72., 684., 0.]]],\n", "\n", "\n", " [[[ 0., 6068., 1394., ..., 3772., 3854., 574.],\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " [ 0., 1480., 340., ..., 920., 940., 140.],\n", " ...,\n", " [ 0., 3552., 816., ..., 2208., 2256., 336.],\n", " [ 0., 5032., 1156., ..., 3128., 3196., 476.],\n", " [ 0., 5254., 1207., ..., 3266., 3337., 497.]],\n", "\n", " [[6068., 0., 4264., ..., 5412., 5330., 3280.],\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " [1480., 0., 1040., ..., 1320., 1300., 800.],\n", " ...,\n", " [3552., 0., 2496., ..., 3168., 3120., 1920.],\n", " [5032., 0., 3536., ..., 4488., 4420., 2720.],\n", " [5254., 0., 3692., ..., 4686., 4615., 2840.]],\n", "\n", " [[1394., 4264., 0., ..., 246., 7380., 8036.],\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " [ 340., 1040., 0., ..., 60., 1800., 1960.],\n", " ...,\n", " [ 816., 2496., 0., ..., 144., 4320., 4704.],\n", " [1156., 3536., 0., ..., 204., 6120., 6664.],\n", " [1207., 3692., 0., ..., 213., 6390., 6958.]],\n", "\n", " ...,\n", "\n", " [[3772., 5412., 246., ..., 0., 7134., 492.],\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " [ 920., 1320., 60., ..., 0., 1740., 120.],\n", " ...,\n", " [2208., 3168., 144., ..., 0., 4176., 288.],\n", " [3128., 4488., 204., ..., 0., 5916., 408.],\n", " [3266., 4686., 213., ..., 0., 6177., 426.]],\n", "\n", " [[3854., 5330., 7380., ..., 7134., 0., 4674.],\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " [ 940., 1300., 1800., ..., 1740., 0., 1140.],\n", " ...,\n", " [2256., 3120., 4320., ..., 4176., 0., 2736.],\n", " [3196., 4420., 6120., ..., 5916., 0., 3876.],\n", " [3337., 4615., 6390., ..., 6177., 0., 4047.]],\n", "\n", " [[ 574., 3280., 8036., ..., 492., 4674., 0.],\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " [ 140., 800., 1960., ..., 120., 1140., 0.],\n", " ...,\n", " [ 336., 1920., 4704., ..., 288., 2736., 0.],\n", " [ 476., 2720., 6664., ..., 408., 3876., 0.],\n", " [ 497., 2840., 6958., ..., 426., 4047., 0.]]],\n", "\n", "\n", " [[[ 0., 6364., 1462., ..., 3956., 4042., 602.],\n", " [ 0., 1480., 340., ..., 920., 940., 140.],\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " ...,\n", " [ 0., 4070., 935., ..., 2530., 2585., 385.],\n", " [ 0., 5550., 1275., ..., 3450., 3525., 525.],\n", " [ 0., 5624., 1292., ..., 3496., 3572., 532.]],\n", "\n", " [[6364., 0., 4472., ..., 5676., 5590., 3440.],\n", " [1480., 0., 1040., ..., 1320., 1300., 800.],\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " ...,\n", " [4070., 0., 2860., ..., 3630., 3575., 2200.],\n", " [5550., 0., 3900., ..., 4950., 4875., 3000.],\n", " [5624., 0., 3952., ..., 5016., 4940., 3040.]],\n", "\n", " [[1462., 4472., 0., ..., 258., 7740., 8428.],\n", " [ 340., 1040., 0., ..., 60., 1800., 1960.],\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " ...,\n", " [ 935., 2860., 0., ..., 165., 4950., 5390.],\n", " [1275., 3900., 0., ..., 225., 6750., 7350.],\n", " [1292., 3952., 0., ..., 228., 6840., 7448.]],\n", "\n", " ...,\n", "\n", " [[3956., 5676., 258., ..., 0., 7482., 516.],\n", " [ 920., 1320., 60., ..., 0., 1740., 120.],\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " ...,\n", " [2530., 3630., 165., ..., 0., 4785., 330.],\n", " [3450., 4950., 225., ..., 0., 6525., 450.],\n", " [3496., 5016., 228., ..., 0., 6612., 456.]],\n", "\n", " [[4042., 5590., 7740., ..., 7482., 0., 4902.],\n", " [ 940., 1300., 1800., ..., 1740., 0., 1140.],\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " ...,\n", " [2585., 3575., 4950., ..., 4785., 0., 3135.],\n", " [3525., 4875., 6750., ..., 6525., 0., 4275.],\n", " [3572., 4940., 6840., ..., 6612., 0., 4332.]],\n", "\n", " [[ 602., 3440., 8428., ..., 516., 4902., 0.],\n", " [ 140., 800., 1960., ..., 120., 1140., 0.],\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " ...,\n", " [ 385., 2200., 5390., ..., 330., 3135., 0.],\n", " [ 525., 3000., 7350., ..., 450., 4275., 0.],\n", " [ 532., 3040., 7448., ..., 456., 4332., 0.]]],\n", "\n", "\n", " ...,\n", "\n", "\n", " [[[ 0., 2590., 595., ..., 1610., 1645., 245.],\n", " [ 0., 3552., 816., ..., 2208., 2256., 336.],\n", " [ 0., 4070., 935., ..., 2530., 2585., 385.],\n", " ...,\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " [ 0., 1480., 340., ..., 920., 940., 140.],\n", " [ 0., 1702., 391., ..., 1058., 1081., 161.]],\n", "\n", " [[2590., 0., 1820., ..., 2310., 2275., 1400.],\n", " [3552., 0., 2496., ..., 3168., 3120., 1920.],\n", " [4070., 0., 2860., ..., 3630., 3575., 2200.],\n", " ...,\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " [1480., 0., 1040., ..., 1320., 1300., 800.],\n", " [1702., 0., 1196., ..., 1518., 1495., 920.]],\n", "\n", " [[ 595., 1820., 0., ..., 105., 3150., 3430.],\n", " [ 816., 2496., 0., ..., 144., 4320., 4704.],\n", " [ 935., 2860., 0., ..., 165., 4950., 5390.],\n", " ...,\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " [ 340., 1040., 0., ..., 60., 1800., 1960.],\n", " [ 391., 1196., 0., ..., 69., 2070., 2254.]],\n", "\n", " ...,\n", "\n", " [[1610., 2310., 105., ..., 0., 3045., 210.],\n", " [2208., 3168., 144., ..., 0., 4176., 288.],\n", " [2530., 3630., 165., ..., 0., 4785., 330.],\n", " ...,\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " [ 920., 1320., 60., ..., 0., 1740., 120.],\n", " [1058., 1518., 69., ..., 0., 2001., 138.]],\n", "\n", " [[1645., 2275., 3150., ..., 3045., 0., 1995.],\n", " [2256., 3120., 4320., ..., 4176., 0., 2736.],\n", " [2585., 3575., 4950., ..., 4785., 0., 3135.],\n", " ...,\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " [ 940., 1300., 1800., ..., 1740., 0., 1140.],\n", " [1081., 1495., 2070., ..., 2001., 0., 1311.]],\n", "\n", " [[ 245., 1400., 3430., ..., 210., 1995., 0.],\n", " [ 336., 1920., 4704., ..., 288., 2736., 0.],\n", " [ 385., 2200., 5390., ..., 330., 3135., 0.],\n", " ...,\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " [ 140., 800., 1960., ..., 120., 1140., 0.],\n", " [ 161., 920., 2254., ..., 138., 1311., 0.]]],\n", "\n", "\n", " [[[ 0., 1406., 323., ..., 874., 893., 133.],\n", " [ 0., 5032., 1156., ..., 3128., 3196., 476.],\n", " [ 0., 5550., 1275., ..., 3450., 3525., 525.],\n", " ...,\n", " [ 0., 1480., 340., ..., 920., 940., 140.],\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " [ 0., 444., 102., ..., 276., 282., 42.]],\n", "\n", " [[1406., 0., 988., ..., 1254., 1235., 760.],\n", " [5032., 0., 3536., ..., 4488., 4420., 2720.],\n", " [5550., 0., 3900., ..., 4950., 4875., 3000.],\n", " ...,\n", " [1480., 0., 1040., ..., 1320., 1300., 800.],\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " [ 444., 0., 312., ..., 396., 390., 240.]],\n", "\n", " [[ 323., 988., 0., ..., 57., 1710., 1862.],\n", " [1156., 3536., 0., ..., 204., 6120., 6664.],\n", " [1275., 3900., 0., ..., 225., 6750., 7350.],\n", " ...,\n", " [ 340., 1040., 0., ..., 60., 1800., 1960.],\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " [ 102., 312., 0., ..., 18., 540., 588.]],\n", "\n", " ...,\n", "\n", " [[ 874., 1254., 57., ..., 0., 1653., 114.],\n", " [3128., 4488., 204., ..., 0., 5916., 408.],\n", " [3450., 4950., 225., ..., 0., 6525., 450.],\n", " ...,\n", " [ 920., 1320., 60., ..., 0., 1740., 120.],\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " [ 276., 396., 18., ..., 0., 522., 36.]],\n", "\n", " [[ 893., 1235., 1710., ..., 1653., 0., 1083.],\n", " [3196., 4420., 6120., ..., 5916., 0., 3876.],\n", " [3525., 4875., 6750., ..., 6525., 0., 4275.],\n", " ...,\n", " [ 940., 1300., 1800., ..., 1740., 0., 1140.],\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " [ 282., 390., 540., ..., 522., 0., 342.]],\n", "\n", " [[ 133., 760., 1862., ..., 114., 1083., 0.],\n", " [ 476., 2720., 6664., ..., 408., 3876., 0.],\n", " [ 525., 3000., 7350., ..., 450., 4275., 0.],\n", " ...,\n", " [ 140., 800., 1960., ..., 120., 1140., 0.],\n", " [ 0., 0., 0., ..., 0., 0., 0.],\n", " [ 42., 240., 588., ..., 36., 342., 0.]]],\n", "\n", "\n", " [[[ 0., 888., 204., ..., 552., 564., 84.],\n", " [ 0., 5254., 1207., ..., 3266., 3337., 497.],\n", " [ 0., 5624., 1292., ..., 3496., 3572., 532.],\n", " ...,\n", " [ 0., 1702., 391., ..., 1058., 1081., 161.],\n", " [ 0., 444., 102., ..., 276., 282., 42.],\n", " [ 0., 0., 0., ..., 0., 0., 0.]],\n", "\n", " [[ 888., 0., 624., ..., 792., 780., 480.],\n", " [5254., 0., 3692., ..., 4686., 4615., 2840.],\n", " [5624., 0., 3952., ..., 5016., 4940., 3040.],\n", " ...,\n", " [1702., 0., 1196., ..., 1518., 1495., 920.],\n", " [ 444., 0., 312., ..., 396., 390., 240.],\n", " [ 0., 0., 0., ..., 0., 0., 0.]],\n", "\n", " [[ 204., 624., 0., ..., 36., 1080., 1176.],\n", " [1207., 3692., 0., ..., 213., 6390., 6958.],\n", " [1292., 3952., 0., ..., 228., 6840., 7448.],\n", " ...,\n", " [ 391., 1196., 0., ..., 69., 2070., 2254.],\n", " [ 102., 312., 0., ..., 18., 540., 588.],\n", " [ 0., 0., 0., ..., 0., 0., 0.]],\n", "\n", " ...,\n", "\n", " [[ 552., 792., 36., ..., 0., 1044., 72.],\n", " [3266., 4686., 213., ..., 0., 6177., 426.],\n", " [3496., 5016., 228., ..., 0., 6612., 456.],\n", " ...,\n", " [1058., 1518., 69., ..., 0., 2001., 138.],\n", " [ 276., 396., 18., ..., 0., 522., 36.],\n", " [ 0., 0., 0., ..., 0., 0., 0.]],\n", "\n", " [[ 564., 780., 1080., ..., 1044., 0., 684.],\n", " [3337., 4615., 6390., ..., 6177., 0., 4047.],\n", " [3572., 4940., 6840., ..., 6612., 0., 4332.],\n", " ...,\n", " [1081., 1495., 2070., ..., 2001., 0., 1311.],\n", " [ 282., 390., 540., ..., 522., 0., 342.],\n", " [ 0., 0., 0., ..., 0., 0., 0.]],\n", "\n", " [[ 84., 480., 1176., ..., 72., 684., 0.],\n", " [ 497., 2840., 6958., ..., 426., 4047., 0.],\n", " [ 532., 3040., 7448., ..., 456., 4332., 0.],\n", " ...,\n", " [ 161., 920., 2254., ..., 138., 1311., 0.],\n", " [ 42., 240., 588., ..., 36., 342., 0.],\n", " [ 0., 0., 0., ..., 0., 0., 0.]]]])" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "np.einsum(\"ij,kl->ikjl\", distance, flow, out=matrix.quadratic)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Creating constraints\n", "\n", "Impose a one-hot constraint on each row and column of the variable array q created in [](#Creating variables).\n" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [], "source": [ "from amplify import one_hot\n", "\n", "constraints = one_hot(q, axis=1) + one_hot(q, axis=0)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "(qap-model)=\n", "\n", "### Creating a combinatorial optimization model\n", "\n", "Let's combine the objective function and constraints to create a model." ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [], "source": [ "penalty_weight = np.max(distance) * np.max(flow) * (N - 1)\n", "model = matrix + penalty_weight * constraints" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The penalty_weight is applied to the constraints to give weight to the constraints. In Amplify AE, the solver used in this example, if you do not specify appropriate weights for the constraints, the solver will search in the direction of making the objective function smaller rather than trying to satisfy the constraints, and you will not be able to find a feasible solution. See [](penalty.md) for details." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Creating a solver client\n", "\n", "Now, we will create a solver client to perform combinatorial optimization using Amplify AE. The solver client class corresponding to Amplify AE is {py:class}~amplify.FixstarsClient class." ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [], "source": [ "from amplify import FixstarsClient\n", "\n", "client = FixstarsClient()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We also need to set the API token required to run Amplify AE.\n", "\n", "{tip}\n", "After [user registration](https://amplify.fixstars.com/en/register), you can obtain a free API token that can be used for evaluation and validation purposes.\n", "" ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "tags": [ "skip-execution" ] }, "outputs": [], "source": [ "client.token = \"xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx\"" ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "tags": [ "remove-cell" ] }, "outputs": [], "source": [ "client.token = \"\"" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We will set the solver's timeout." ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [], "source": [ "import datetime\n", "\n", "client.parameters.timeout = datetime.timedelta(seconds=1)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Executing the solver\n", "\n", "Finally, we will execute the solver using the created combinatorial optimization model and the solver client to find the solution to the quadratic programming problem." ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [], "source": [ "from amplify import solve\n", "\n", "result = solve(model, client)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The objective function value based on the best solution is shown below." ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "204954.0" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "result.best.objective" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The values of the variables in the optimal solution can be obtained in the form of a NumPy multidimensional array as follows." ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[[0. 0. 0. 0. 1. 0. 0. 0. 0. 0.]\n", " [0. 0. 0. 1. 0. 0. 0. 0. 0. 0.]\n", " [0. 0. 0. 0. 0. 0. 0. 0. 0. 1.]\n", " [0. 0. 0. 0. 0. 0. 0. 0. 1. 0.]\n", " [0. 0. 1. 0. 0. 0. 0. 0. 0. 0.]\n", " [0. 0. 0. 0. 0. 0. 0. 1. 0. 0.]\n", " [0. 0. 0. 0. 0. 0. 1. 0. 0. 0.]\n", " [0. 0. 0. 0. 0. 1. 0. 0. 0. 0.]\n", " [0. 1. 0. 0. 0. 0. 0. 0. 0. 0.]\n", " [1. 0. 0. 0. 0. 0. 0. 0. 0. 0.]]\n" ] } ], "source": [ "q_values = q.evaluate(result.best.values)\n", "\n", "print(q_values)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Checking the results\n", "\n", "We will visualize the results using matplotlib." ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [], "source": [ "import matplotlib.pyplot as plt\n", "import itertools" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "