Spaces:

ashutosh1919
/

quantum-perceptron

Runtime error

App Files Files Community

ashutosh1919 commited on Apr 15, 2023

Commit

4da8a9d

•

1 Parent(s): 2b9363f

Added code to generate data and train model

Browse files

Files changed (11) hide show

.DS_Store +0 -0
TrainPerceptron.ipynb +348 -0
data/sample_space_qubits_4_fweight_626.txt +0 -0
data/train_space_qubits_4_fweight_626.txt +3050 -0
quantum_perceptron/train/__init__.py +2 -0
quantum_perceptron/train/data_gen.py +90 -0
quantum_perceptron/train/training.py +164 -0
quantum_perceptron/utils/__init__.py +1 -0
quantum_perceptron/utils/data_utils.py +42 -1
quantum_perceptron/utils/plot_utils.py +41 -0
requirements.txt +2 -1

.DS_Store ADDED Viewed

Binary file (6.15 kB). View file

TrainPerceptron.ipynb ADDED Viewed

	@@ -0,0 +1,348 @@

+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "9f3fb515-747d-4e93-bf07-de5553a6e958",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "from quantum_perceptron.utils import plot_img_from_data, get_vector_from_int, get_int_from_vector\n",
+    "from quantum_perceptron.train import generate_dataset, PerceptronTrainer\n",
+    "from quantum_perceptron import Perceptron"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "681740ed-ea82-428e-a5c8-9d44cfe92b2e",
+   "metadata": {},
+   "source": [
+    "# Generate Dataset\n",
+    "\n",
+    "Generate dataset to train 4 qubit perceptron."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "eba68da3-013e-4501-bdaa-67c82cef393d",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "fixed_weight = 626\n",
+    "num_qubits = 4\n",
+    "plot_img_from_data(fixed_weight, num_qubits)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "18e6cc2b-51be-4d83-8e3a-4bbd4078bcf7",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 65536/65536 [19:38<00:00, 55.59it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Number of positive samples: 274\n",
+      "Number of negative samples: 65262\n",
+      "Saved data to ./data/sample_space_qubits_4_fweight_626.txt\n",
+      "Saved training data to ./data/train_space_qubits_4_fweight_626.txt\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "generate_dataset(num_qubits=num_qubits, fixed_weight=fixed_weight)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "ecd46307-eb53-46b4-809a-4493b1661d62",
+   "metadata": {},
+   "source": [
+    "# Perceptron Training\n",
+    "\n",
+    "Train generated dataset with 4 qubit perceptron. We will initialize random weight and will train perceptron so that it gradually updates the weight to the `fixed_weight` with which the dataset is generated."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "b6081742-8913-4d46-a6bc-d62bc2eb715b",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "\u001b[34m\u001b[1mwandb\u001b[0m: Currently logged in as: \u001b[33mashutosh1919\u001b[0m. Use \u001b[1m`wandb login --relogin`\u001b[0m to force relogin\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "Tracking run with wandb version 0.14.2"
+      ],
+      "text/plain": [
+       "<IPython.core.display.HTML object>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "Run data is saved locally in <code>/Users/ashutosh1919/Documents/IUB/QuantumProgramming/quantum-perceptron/wandb/run-20230415_181130-p665sffu</code>"
+      ],
+      "text/plain": [
+       "<IPython.core.display.HTML object>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "Syncing run <strong><a href='https://wandb.ai/ashutosh1919/quantum-perceptron/runs/p665sffu' target=\"_blank\">vivid-lake-5</a></strong> to <a href='https://wandb.ai/ashutosh1919/quantum-perceptron' target=\"_blank\">Weights & Biases</a> (<a href='https://wandb.me/run' target=\"_blank\">docs</a>)<br/>"
+      ],
+      "text/plain": [
+       "<IPython.core.display.HTML object>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       " View project at <a href='https://wandb.ai/ashutosh1919/quantum-perceptron' target=\"_blank\">https://wandb.ai/ashutosh1919/quantum-perceptron</a>"
+      ],
+      "text/plain": [
+       "<IPython.core.display.HTML object>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       " View run at <a href='https://wandb.ai/ashutosh1919/quantum-perceptron/runs/p665sffu' target=\"_blank\">https://wandb.ai/ashutosh1919/quantum-perceptron/runs/p665sffu</a>"
+      ],
+      "text/plain": [
+       "<IPython.core.display.HTML object>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "trainer = PerceptronTrainer(\n",
+    "    num_qubits=num_qubits,\n",
+    "    fixed_weight=fixed_weight,\n",
+    "    dataset_path=\"./data/train_space_qubits_4_fweight_626.txt\"\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "c07e3adf-43e8-4f70-9f53-4b7367d7df8f",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Randomly initialized weight before training:\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "print(\"Randomly initialized weight before training:\")\n",
+    "plot_img_from_data(trainer.weight_variable, num_qubits)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "92be5fcf-e82a-45a6-8e26-3b5bb89101c7",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3050/3050 [01:56<00:00, 26.25it/s]\n",
+      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3050/3050 [01:57<00:00, 26.03it/s]\n",
+      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3050/3050 [01:55<00:00, 26.41it/s]\n",
+      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3050/3050 [01:52<00:00, 27.15it/s]\n",
+      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3050/3050 [01:49<00:00, 27.81it/s]\n",
+      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3050/3050 [01:56<00:00, 26.14it/s]\n",
+      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3050/3050 [01:58<00:00, 25.78it/s]\n",
+      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3050/3050 [01:48<00:00, 27.99it/s]\n",
+      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3050/3050 [01:54<00:00, 26.72it/s]\n",
+      " 32%|████████████████████████████████████▋                                                                               | 963/3050 [00:35<01:17, 26.77it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Training converged at step: 28414\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "trainer.train(num_epochs=10)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "84c72b42-e18c-456f-b5a4-f41b577cb50b",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Final weight after training:\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "print(\"Final weight after training:\")\n",
+    "plot_img_from_data(trainer.weight_variable, num_qubits)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "fbecdf4f-68c8-4d4d-9a6f-9ef66bd182f0",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0, 0.5, 'Fidelity Similarity with Actual Weight')"
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "plt.plot(list(range(trainer.num_steps)), sorted(trainer.accumulate_loss))\n",
+    "plt.xlabel(\"Number of Steps\")\n",
+    "plt.ylabel(\"Fidelity Similarity with Actual Weight\")"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "quantum",
+   "language": "python",
+   "name": "quantum"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.16"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

data/sample_space_qubits_4_fweight_626.txt ADDED Viewed

The diff for this file is too large to render. See raw diff

data/train_space_qubits_4_fweight_626.txt ADDED Viewed

	@@ -0,0 +1,3050 @@

+16510,0
+25308,0
+45953,0
+55467,0
+21450,0
+35992,0
+62900,0
+61310,0
+49398,0
+18979,0
+26701,0
+20136,0
+56331,0
+37822,0
+15238,0
+2152,0
+63244,0
+5432,0
+414,0
+45749,0
+3544,0
+44728,0
+21722,0
+9115,0
+3807,0
+53311,0
+3787,0
+61967,0
+4452,0
+18964,0
+59893,0
+2811,0
+29521,0
+13007,0
+29106,0
+23412,0
+27383,0
+1983,0
+165,0
+40881,0
+45338,0
+22596,0
+40586,0
+19972,0
+32613,0
+4480,0
+32157,1
+36179,0
+25427,0
+22474,0
+32506,0
+8218,0
+24510,0
+4788,0
+29550,0
+11340,0
+53316,0
+13535,0
+49919,0
+32083,0
+34868,0
+1282,0
+11218,0
+40575,0
+25965,0
+17983,0
+2715,0
+57126,0
+57445,0
+45625,0
+24164,0
+51804,0
+49918,0
+4279,0
+36585,0
+42160,0
+43284,0
+19469,0
+51251,0
+51532,0
+17484,0
+37128,0
+50201,0
+60642,0
+35273,0
+58605,0
+17795,0
+56139,0
+15794,0
+52954,0
+38035,0
+37477,0
+34246,0
+64976,0
+35765,0
+57960,0
+21443,0
+31474,0
+18717,0
+54265,0
+1736,0
+30864,0
+64128,0
+880,1
+44013,0
+39883,0
+24324,0
+42620,0
+23920,0
+32829,0
+57315,0
+45024,0
+30414,0
+33029,0
+31875,0
+46026,0
+9429,0
+47739,0
+13833,0
+9974,0
+41709,0
+15136,0
+63614,0
+50054,0
+1626,0
+35713,0
+44095,0
+11994,0
+33886,0
+44770,0
+10772,0
+54569,0
+24835,0
+13961,0
+55835,0
+28015,0
+58129,0
+17213,0
+55715,0
+53067,0
+42952,0
+43528,0
+45700,0
+23626,0
+21897,0
+41051,0
+11544,0
+53766,0
+48282,0
+5175,0
+30933,0
+59537,0
+2900,0
+42353,0
+98,1
+33867,0
+20365,0
+23360,0
+11315,0
+4118,0
+56267,0
+14294,0
+19621,0
+1118,0
+37389,0
+33704,0
+36049,0
+17246,0
+44603,0
+33138,0
+638,1
+44092,0
+14210,0
+64895,0
+49563,0
+50963,0
+4629,0
+22156,0
+26848,0
+48255,0
+1499,0
+29132,0
+40092,0
+16573,0
+62720,0
+52657,0
+62647,0
+1779,0
+55821,0
+53784,0
+56653,0
+24126,0
+56329,0
+29820,0
+51293,0
+39659,0
+40564,0
+24198,0
+36954,0
+41059,0
+26816,0
+13292,0
+64897,1
+35736,0
+51583,0
+13226,0
+54430,0
+14199,0
+57418,0
+50372,0
+37410,0
+48606,0
+21486,0
+29042,0
+58557,0
+41101,0
+36968,0
+32025,0
+18832,0
+36991,0
+42059,0
+9590,0
+8294,0
+51460,0
+58315,0
+24547,0
+51543,0
+54181,0
+45311,0
+40415,0
+30941,0
+64679,0
+54814,0
+50421,0
+1911,0
+10063,0
+3845,0
+22911,0
+24561,0
+11491,0
+39579,0
+18445,0
+9446,0
+36472,0
+36884,0
+35529,0
+11901,0
+61777,0
+48365,0
+2737,0
+52371,0
+40292,0
+2098,0
+560,1
+49264,0
+30848,0
+64336,0
+37868,0
+10586,0
+40745,0
+61606,0
+31712,0
+61350,0
+56301,0
+39778,0
+21033,0
+12156,0
+58466,0
+40560,0
+176,0
+18924,0
+61657,0
+36745,0
+26186,0
+18097,0
+13761,0
+8249,0
+12545,0
+52075,0
+25068,0
+40407,0
+9330,0
+22128,0
+41922,0
+63136,0
+43585,0
+23109,0
+19425,0
+31012,0
+729,0
+26901,0
+50161,0
+25803,0
+47501,1
+19259,0
+2337,0
+169,0
+32043,0
+575,0
+7535,0
+37289,0
+25250,0
+35402,0
+7544,0
+27554,0
+6784,0
+24490,0
+26397,0
+16908,0
+26956,0
+45525,0
+49498,0
+60148,0
+24483,0
+62396,0
+19306,0
+56985,0
+16089,0
+26874,0
+54777,0
+21981,0
+56576,0
+642,0
+14868,0
+30913,0
+41584,0
+40537,0
+48377,0
+45414,0
+16446,0
+61840,0
+2084,0
+3980,0
+7753,0
+18594,0
+46280,0
+62317,0
+25220,0
+12354,0
+18557,0
+52631,0
+42830,0
+41278,0
+12388,0
+33017,0
+29077,0
+44985,0
+47759,0
+15103,0
+32900,0
+33144,0
+39094,0
+9821,0
+38946,0
+59688,0
+38534,0
+33248,0
+12743,0
+7292,0
+42399,0
+11323,0
+38984,0
+2696,0
+9233,0
+19959,0
+17502,0
+17702,0
+27282,0
+20030,0
+61575,0
+62882,0
+4873,0
+59248,0
+9580,0
+32805,0
+56417,0
+54452,0
+49888,0
+64518,0
+63956,0
+29315,0
+55998,0
+20727,0
+59384,0
+5923,0
+10741,0
+64519,0
+38954,0
+55044,0
+49691,0
+43016,0
+23664,0
+35565,0
+49143,0
+64543,0
+8629,0
+13646,0
+16080,0
+646,0
+64752,0
+45120,0
+6009,0
+7600,0
+46682,0
+24030,0
+53455,0
+47880,0
+23914,0
+60850,0
+26799,0
+28146,0
+65284,0
+21128,0
+31573,0
+39499,0
+3696,0
+15393,0
+52166,0
+19507,0
+35136,0
+49062,0
+60557,1
+52933,0
+14325,0
+19336,0
+36559,0
+37265,0
+63349,0
+12108,0
+34852,0
+46491,0
+12607,0
+22107,0
+50735,0
+44512,0
+24384,0
+3798,0
+18199,0
+51535,0
+34451,0
+5872,0
+46746,0
+6579,0
+65421,1
+43970,0
+41456,0
+61432,0
+46102,0
+27905,0
+22040,0
+18275,0
+18896,0
+33560,0
+2088,0
+44472,0
+41671,0
+31028,0
+8678,0
+65371,0
+56834,0
+53839,0
+19058,1
+47223,0
+44589,0
+38093,0
+52431,0
+64278,0
+50097,0
+30360,0
+18845,0
+65496,0
+46846,0
+34207,0
+25090,0
+38197,0
+12558,0
+28991,0
+16335,0
+43681,0
+20007,0
+61020,0
+60362,0
+41958,0
+30475,0
+13951,0
+29898,0
+58146,0
+57522,0
+20853,0
+30458,0
+17010,1
+16252,0
+21581,0
+14581,0
+62855,0
+10828,0
+9147,0
+9050,0
+50287,0
+43337,0
+27268,0
+34469,0
+24337,0
+10214,0
+53797,0
+54778,0
+39452,0
+57177,0
+5766,0
+54256,0
+36770,0
+27828,0
+58786,0
+31427,0
+45017,0
+47050,0
+21910,0
+37411,0
+18629,0
+50989,0
+53071,0
+38361,0
+44559,0
+9438,0
+25133,0
+6280,0
+16818,0
+33587,0
+60330,0
+2743,0
+20885,0
+56052,0
+15202,0
+8020,0
+47122,0
+50204,0
+63406,0
+40748,0
+1457,0
+65348,0
+14776,0
+27997,0
+49368,0
+49424,0
+59446,0
+45806,0
+34221,0
+55383,0
+15722,0
+55792,0
+7677,0
+59381,0
+17057,0
+52503,0
+6735,0
+7692,0
+26594,0
+25284,0
+14932,0
+44283,0
+49349,0
+56245,0
+50442,0
+52483,0
+19932,0
+10630,0
+8976,0
+56963,0
+22134,0
+4980,0
+16180,0
+30965,0
+7782,0
+34437,0
+44267,0
+27535,0
+10024,0
+35831,0
+22462,0
+31414,0
+31650,0
+11234,0
+48409,0
+30516,0
+16758,0
+15767,0
+39031,0
+62012,0
+38019,0
+16463,0
+50225,0
+11718,0
+46347,0
+44445,0
+64099,0
+22147,0
+65222,0
+17864,0
+58027,0
+42323,0
+14576,0
+42885,0
+2435,0
+25656,0
+17900,0
+26138,0
+62548,0
+7109,0
+31420,0
+41698,0
+37854,0
+16406,0
+23870,0
+40938,0
+5238,0
+64586,0
+112,1
+59371,0
+53940,0
+21332,0
+56854,0
+53658,0
+37840,0
+33563,0
+65299,0
+62776,0
+27810,0
+24318,0
+40199,0
+14280,0
+11966,0
+41190,0
+6511,0
+49051,0
+63940,0
+21750,0
+40491,0
+29149,0
+38712,0
+55090,0
+60030,0
+19120,0
+1,0
+4022,0
+21891,0
+8233,0
+54641,0
+39130,0
+65102,0
+58867,0
+26012,0
+27235,0
+2641,0
+9604,0
+39084,0
+14658,0
+26831,0
+12017,0
+54367,0
+57513,0
+43275,0
+27658,0
+24783,0
+17225,0
+279,0
+45249,0
+32061,0
+4526,0
+1797,0
+46606,0
+42656,0
+51324,0
+64017,0
+54190,0
+58904,0
+13609,0
+43821,0
+23884,0
+48443,0
+32670,0
+5233,0
+57505,0
+49767,0
+35841,0
+19721,0
+32981,0
+58297,0
+51156,0
+37676,0
+58663,0
+16437,0
+13203,0
+32813,0
+42068,0
+53416,0
+3532,0
+59870,0
+30509,0
+3769,0
+12627,0
+32842,0
+40080,0
+61136,0
+21087,0
+40532,0
+46262,0
+32160,0
+23216,0
+14165,0
+55915,0
+44214,0
+57434,0
+46391,0
+22424,0
+46198,0
+39376,0
+55385,0
+8479,0
+23797,0
+64353,0
+60974,0
+15216,0
+26570,0
+25493,0
+23949,1
+23946,0
+34866,0
+62066,0
+20535,0
+11798,0
+19289,0
+23703,0
+22370,0
+16862,0
+48451,0
+54654,0
+22096,0
+30032,0
+27064,0
+46266,0
+54749,0
+48570,0
+42607,0
+8633,0
+20320,0
+49144,0
+45677,0
+37817,0
+62194,0
+63642,0
+57066,0
+28543,0
+45566,0
+9715,0
+44723,0
+62849,0
+35425,0
+64082,0
+22111,0
+23822,0
+23645,0
+5861,0
+9940,0
+11852,0
+26474,0
+14624,0
+14386,0
+23159,0
+3507,0
+59187,0
+43419,0
+20117,0
+34588,0
+6286,0
+63908,0
+31317,0
+65280,0
+28373,0
+59957,0
+30060,0
+59362,0
+65412,0
+38684,0
+26170,0
+19797,0
+49259,0
+30311,0
+29530,0
+61200,0
+38827,0
+51664,0
+40217,0
+2964,0
+58623,0
+37216,0
+58244,0
+52364,0
+10660,0
+60991,0
+51276,0
+44057,0
+57851,0
+29101,0
+3039,0
+8120,0
+62155,0
+49965,0
+31072,0
+4547,0
+34732,0
+18020,0
+58457,0
+2417,0
+49001,0
+53748,0
+1009,0
+2678,1
+14826,0
+47140,0
+758,1
+37968,0
+48196,0
+5403,0
+40816,0
+10707,0
+49089,0
+20734,0
+43549,0
+23009,0
+54860,0
+7467,0
+52947,0
+6811,0
+22329,0
+5123,0
+22182,0
+21801,0
+30229,0
+11770,0
+53463,0
+50513,0
+54577,0
+51329,0
+18932,0
+18981,0
+58375,0
+39468,0
+64740,0
+44650,0
+39086,0
+47115,0
+2539,0
+57325,0
+48389,0
+39044,0
+57693,0
+13176,0
+44434,0
+11092,0
+53575,0
+47529,0
+14881,0
+28503,0
+16528,0
+42587,0
+42336,0
+63770,0
+34884,0
+34141,0
+31052,0
+60960,0
+42584,0
+34470,0
+12479,0
+861,0
+65138,0
+30255,0
+7846,0
+6171,0
+26350,0
+28957,0
+45290,0
+31733,0
+49609,0
+30468,0
+45568,0
+10566,0
+23279,0
+45769,0
+10767,0
+18313,0
+11860,0
+3488,0
+59967,0
+42444,0
+39945,0
+14060,0
+26205,0
+13912,0
+13123,0
+30552,0
+3277,0
+30439,0
+4605,0
+44948,0
+32405,0
+31559,0
+20298,0
+5767,0
+37987,0
+24604,0
+10895,0
+55995,0
+57902,0
+32948,0
+60869,0
+15124,0
+22670,0
+33232,0
+38991,0
+8463,0
+55319,0
+9533,0
+39314,0
+19112,0
+43287,0
+9074,1
+24834,0
+33040,0
+48070,0
+39175,0
+32131,0
+5211,0
+64693,0
+4052,0
+3182,0
+32609,0
+63790,0
+17061,0
+45209,0
+14568,0
+33581,0
+22859,0
+43271,0
+16050,0
+3611,0
+49711,0
+58430,0
+61665,0
+19351,0
+53862,0
+55423,0
+8157,0
+3763,0
+50744,0
+57273,0
+52320,0
+7095,0
+21191,0
+42650,0
+41173,0
+54405,0
+35442,1
+46912,0
+20101,0
+54445,0
+44158,0
+59680,0
+28309,0
+3585,0
+44330,0
+11625,0
+54843,0
+46723,0
+50545,0
+31014,0
+14865,0
+41873,0
+30752,0
+9299,0
+37309,0
+46673,0
+41220,0
+18257,0
+36465,0
+54869,0
+18581,0
+10082,0
+37929,0
+28296,0
+2340,0
+43461,0
+57166,0
+24031,0
+21976,0
+35195,0
+36801,0
+2916,0
+44439,0
+63310,0
+9856,0
+16555,0
+29571,0
+40851,0
+59883,0
+20042,0
+4003,0
+58932,0
+6659,0
+34988,0
+59518,0
+18253,0
+20758,0
+41967,0
+19356,0
+57626,0
+26801,0
+55563,0
+14428,0
+60228,0
+49680,0
+32313,0
+48549,0
+60820,0
+41447,0
+57644,0
+62114,0
+27935,0
+12502,0
+3218,0
+43558,0
+49845,0
+9216,0
+54295,0
+25687,0
+58701,0
+20657,0
+55814,0
+11202,0
+43165,0
+53548,0
+44333,0
+40473,0
+4574,0
+60889,0
+45940,0
+58181,0
+12296,0
+54851,0
+56290,0
+64315,0
+54022,0
+58242,0
+46989,0
+46808,0
+818,1
+63140,0
+46462,0
+18542,0
+46791,0
+64780,1
+40289,0
+51604,0
+14234,0
+8320,0
+49679,0
+14659,0
+53090,0
+58564,0
+35032,0
+20203,0
+49503,0
+48240,0
+34476,0
+41143,0
+6479,0
+46879,0
+33239,0
+29431,0
+9956,0
+64972,1
+25162,0
+44325,0
+57109,0
+9088,0
+23646,0
+30409,0
+37850,0
+1784,0
+1848,0
+38972,0
+10509,0
+16684,0
+64,0
+61154,0
+16795,0
+722,1
+34832,0
+46652,0
+35211,0
+19448,0
+53123,0
+26893,0
+62793,0
+7002,0
+32571,0
+50844,0
+10135,0
+26552,0
+33235,0
+52810,0
+57234,0
+27687,0
+48956,0
+54144,0
+12588,0
+1360,0
+55224,0
+42320,0
+25983,0
+24118,0
+46770,0
+56518,0
+19433,0
+4793,0
+54364,0
+25901,0
+41997,0
+13338,0
+22541,0
+48257,0
+56860,0
+62433,0
+38791,0
+63721,0
+51419,0
+2593,0
+54164,0
+48026,0
+58848,0
+30438,0
+48676,0
+10504,0
+29473,0
+11756,0
+26918,0
+47789,0
+61571,0
+11864,0
+24280,0
+51211,0
+2692,0
+50898,0
+45847,0
+18737,0
+62030,0
+52616,0
+3363,0
+39350,0
+37646,0
+30441,0
+36207,0
+47590,0
+34834,0
+55672,0
+14540,0
+5258,0
+37001,0
+23113,0
+59188,0
+3632,0
+11935,0
+23927,0
+47042,0
+5330,0
+59703,0
+34916,0
+21453,0
+41875,0
+60474,0
+65161,0
+63119,0
+680,0
+3627,0
+53119,0
+61839,0
+36771,0
+51088,0
+58065,0
+35878,0
+13107,0
+50156,0
+24437,0
+4165,0
+28798,0
+6813,0
+49694,0
+52705,0
+42475,0
+55535,0
+21105,0
+6621,0
+15512,0
+64387,0
+7678,0
+1543,0
+9024,0
+9076,0
+16683,0
+47872,0
+22532,0
+51865,0
+10277,0
+52469,0
+61633,0
+62917,0
+59244,0
+54853,0
+7139,0
+60383,0
+3847,0
+37595,0
+55197,0
+494,0
+31424,0
+27286,0
+17535,0
+16331,0
+54645,0
+14711,0
+25553,0
+32344,0
+56817,0
+23937,0
+28414,0
+26925,0
+59527,0
+25257,0
+25719,0
+15053,0
+43596,0
+63506,0
+14736,0
+5814,0
+3328,0
+23322,0
+27029,0
+55173,0
+18359,0
+14692,0
+9309,0
+23976,0
+1956,0
+29514,0
+28963,0
+48992,0
+7850,0
+20128,0
+5068,0
+26625,0
+60373,0
+51737,0
+20579,0
+2214,0
+39239,0
+40736,0
+23162,0
+62872,0
+26405,0
+7645,0
+10735,0
+52563,0
+7766,0
+56731,0
+14751,0
+47959,0
+41947,0
+31821,0
+29754,0
+2929,0
+44717,0
+860,0
+55588,0
+55166,0
+47780,0
+60596,0
+20305,0
+12365,0
+49757,0
+56556,0
+46358,0
+56062,0
+3154,0
+56030,0
+47826,0
+60325,0
+60804,0
+15960,0
+43760,0
+15738,0
+45716,0
+44997,0
+26693,0
+10764,0
+23972,0
+34273,0
+36007,0
+26578,0
+46058,0
+53822,0
+46161,0
+58556,0
+45289,0
+45181,0
+5956,0
+14008,0
+65474,0
+36042,0
+7464,0
+31616,0
+56665,0
+14216,0
+55033,0
+22316,0
+3394,0
+17827,0
+31850,0
+23842,0
+63390,0
+17001,0
+59818,0
+33680,0
+20661,0
+61704,0
+11489,0
+64051,0
+40724,0
+27920,0
+44542,0
+30099,0
+1598,0
+18647,0
+4076,0
+25611,0
+62099,0
+37293,0
+55978,0
+10873,0
+48176,0
+3378,0
+60048,0
+49037,1
+24121,0
+7791,0
+18407,0
+2504,0
+6660,0
+47843,0
+47150,0
+41050,0
+4774,0
+12679,0
+10490,0
+52784,0
+59707,0
+172,0
+38831,0
+63812,0
+57181,0
+42922,0
+41127,0
+21808,0
+32662,0
+12660,0
+12116,0
+33844,0
+58651,0
+9379,0
+36306,0
+17978,0
+64520,0
+44047,0
+51892,0
+62677,0
+36979,0
+6044,0
+19980,0
+50646,0
+8414,0
+3687,0
+42701,0
+3112,0
+35382,0
+17002,0
+1437,0
+59791,0
+22297,0
+25881,0
+2787,0
+56134,0
+52202,0
+56339,0
+37382,0
+24868,0
+4752,0
+48300,0
+50974,0
+15651,0
+34468,0
+55449,0
+30525,0
+13816,0
+8993,0
+10219,0
+28537,0
+39244,0
+60296,0
+39632,0
+60983,0
+29114,0
+26104,0
+10610,0
+17349,0
+52534,0
+58566,0
+20272,0
+21097,0
+2763,0
+11709,0
+24008,0
+53935,0
+4222,0
+43329,0
+5166,0
+57808,0
+3964,0
+4849,0
+22991,0
+29572,0
+19443,0
+29509,0
+36258,0
+38284,0
+56443,0
+16222,0
+46467,0
+54508,0
+20499,0
+52310,0
+33290,0
+64757,0
+2955,0
+11587,0
+49366,0
+47928,0
+37680,0
+52114,0
+37601,0
+17323,0
+26582,0
+58023,0
+45628,0
+61960,0
+18389,0
+3135,0
+65172,0
+35354,0
+51053,0
+42302,0
+32325,0
+55908,0
+22002,0
+47065,0
+27047,0
+51233,0
+38151,0
+23095,0
+595,1
+26778,0
+28846,0
+2353,0
+55144,0
+7814,0
+39524,0
+51834,0
+39489,0
+61580,0
+3168,0
+19961,0
+46878,0
+58286,0
+55145,0
+659,0
+2577,0
+13135,0
+45522,0
+59,0
+24996,0
+54518,0
+14525,0
+49520,0
+16689,0
+57679,0
+54686,0
+22944,0
+17121,0
+41313,0
+29284,0
+39847,0
+17132,0
+48770,0
+40964,0
+6046,0
+1379,0
+41766,0
+14337,0
+60211,0
+10352,0
+56413,0
+16168,0
+40357,0
+1918,0
+8686,0
+46924,0
+12408,0
+44160,0
+44579,0
+5389,0
+53702,0
+2352,0
+29009,0
+62190,0
+58679,0
+349,0
+17881,0
+8054,0
+22030,0
+61352,0
+299,0
+34785,0
+43828,0
+3344,0
+21513,0
+26399,0
+4591,0
+9633,0
+37802,0
+28690,0
+63958,0
+43762,0
+7391,0
+22252,0
+51146,0
+27610,0
+63213,0
+14780,0
+37370,0
+17846,0
+22726,0
+28170,0
+2054,0
+29762,0
+5605,0
+48099,0
+24096,0
+13158,0
+36592,0
+13965,0
+53537,0
+64847,0
+32247,0
+58688,0
+15534,0
+17788,0
+10711,0
+9598,0
+9973,0
+12466,0
+6758,0
+28559,0
+26130,0
+65208,0
+62975,0
+423,0
+36220,0
+11002,0
+6422,0
+152,0
+41582,0
+20283,0
+18244,0
+11997,0
+7033,0
+26204,0
+37668,0
+33390,0
+42853,0
+45226,0
+15234,0
+8731,0
+598,1
+6091,0
+64669,1
+65329,0
+60646,0
+610,1
+56847,0
+825,0
+58003,0
+49172,0
+34703,0
+16919,0
+19612,0
+50852,0
+49065,0
+5427,0
+40535,0
+48563,0
+34055,0
+51464,0
+38319,0
+5290,0
+10348,0
+18838,0
+5229,0
+6928,0
+23772,0
+6245,0
+65004,0
+20187,0
+10385,0
+17936,0
+58090,0
+49054,0
+34845,0
+19719,0
+8999,0
+41153,0
+330,0
+17006,0
+54389,0
+40641,0
+54557,0
+62344,0
+8758,0
+2046,0
+20069,0
+4762,0
+24842,0
+25565,0
+8336,0
+20141,0
+16744,0
+50386,0
+22234,0
+28908,0
+6602,0
+4138,0
+38372,0
+52335,0
+36988,0
+9181,0
+16806,0
+33366,0
+33748,0
+527,0
+42628,0
+59523,0
+56701,0
+64196,0
+59650,0
+6066,0
+49180,0
+43513,0
+21958,0
+28735,0
+54170,0
+40423,0
+22636,0
+64761,0
+59529,0
+13694,0
+31822,0
+23451,0
+55549,0
+30108,0
+24320,0
+2700,0
+64644,0
+18254,0
+38559,0
+49271,0
+18875,0
+43181,0
+23824,0
+23729,0
+50671,0
+16655,0
+26300,0
+24910,0
+5420,0
+29150,0
+12966,0
+60749,0
+61378,0
+49014,0
+64536,0
+62411,0
+61315,0
+50350,0
+9928,0
+1633,0
+433,0
+46362,0
+63629,1
+23500,0
+25467,0
+24811,0
+39046,0
+51344,0
+3647,0
+1008,0
+51579,0
+42457,0
+11507,0
+44665,0
+63261,0
+1936,0
+44941,0
+4690,1
+18966,0
+19909,0
+52172,0
+38059,0
+39267,0
+42368,0
+40439,0
+31945,0
+3818,0
+55227,0
+56867,0
+61083,0
+17470,0
+38564,0
+4742,0
+31586,0
+61687,0
+11858,0
+40524,0
+5208,0
+35537,0
+6572,0
+53516,0
+12649,0
+24045,0
+15208,0
+44945,0
+54679,0
+52121,0
+15443,0
+42001,0
+46895,0
+60834,0
+7054,0
+28428,0
+6752,0
+916,0
+48937,0
+55440,0
+8110,0
+19522,0
+17217,0
+54017,0
+28940,0
+27940,0
+5664,0
+57376,0
+8754,1
+42029,0
+10406,0
+38105,0
+4419,0
+39806,0
+23683,0
+50818,0
+14713,0
+13865,0
+4149,0
+3048,0
+17855,0
+50838,0
+35434,0
+35330,0
+64649,1
+54618,0
+38744,0
+781,0
+65052,0
+25077,0
+10871,0
+49740,0
+41151,0
+39073,0
+36643,0
+12814,0
+52347,0
+16552,0
+20847,0
+3518,0
+58890,0
+34467,0
+33500,0
+10416,0
+10425,0
+53651,0
+17594,0
+32605,0
+20974,0
+60293,0
+40254,0
+53580,0
+35548,0
+18992,0
+39962,0
+60587,0
+33658,0
+53250,0
+20251,0
+6592,0
+10657,0
+24734,0
+48220,0
+44554,0
+29942,0
+48144,0
+17728,0
+50333,0
+42505,0
+8946,1
+27542,0
+5174,0
+16129,0
+6456,0
+29096,0
+48880,0
+32540,0
+43735,0
+36482,0
+32283,0
+17707,0
+10708,0
+1264,0
+25189,0
+7059,0
+53553,0
+3511,0
+49633,0
+16733,0
+25163,0
+58521,0
+30800,0
+44204,0
+37658,0
+34082,0
+32521,0
+19085,0
+7165,0
+30554,0
+13258,0
+61291,0
+48427,0
+51828,0
+2018,0
+14642,0
+38449,0
+12572,0
+30366,0
+6575,0
+61503,0
+34210,0
+58305,0
+7805,0
+20628,0
+11019,0
+36855,0
+62778,0
+39712,0
+19624,0
+38601,0
+51162,0
+16594,0
+56598,0
+35237,0
+61110,0
+21862,0
+32524,0
+31346,0
+13651,0
+34521,0
+1398,0
+11339,0
+42543,0
+6108,0
+45221,0
+48774,0
+25514,0
+47295,0
+50232,0
+64324,0
+17420,0
+57323,0
+25474,0
+43960,0
+1015,0
+27862,0
+2988,0
+25816,0
+33591,0
+3633,0
+26846,0
+61567,0
+36628,0
+24936,0
+13821,0
+17435,0
+7338,0
+29078,0
+54681,0
+38532,0
+48041,0
+21901,0
+22869,0
+44800,0
+21979,0
+52453,0
+17925,0
+60321,0
+37573,0
+18712,0
+3463,0
+53533,0
+25837,0
+9017,0
+61165,0
+44438,0
+29448,0
+59772,0
+63477,0
+33882,0
+5840,0
+57360,0
+23876,0
+24836,0
+50678,0
+34531,0
+37724,0
+12407,0
+1436,0
+59390,0
+35597,0
+34295,0
+1045,0
+367,0
+44869,0
+39026,0
+25801,0
+59737,0
+55019,0
+39460,0
+56935,0
+51319,0
+3937,0
+63482,0
+39011,0
+6837,0
+38293,0
+41751,0
+1138,1
+10115,0
+61533,0
+49901,0
+29015,0
+9071,0
+26619,0
+717,0
+6684,0
+4224,0
+6428,0
+22642,0
+20728,0
+2713,0
+5851,0
+20294,0
+11796,0
+55008,0
+22564,0
+15323,0
+30846,0
+54496,0
+50063,0
+46449,0
+9547,0
+62559,0
+53138,0
+57868,0
+53370,0
+10044,0
+17422,0
+4312,0
+21436,0
+37247,0
+40520,0
+17877,0
+17130,0
+24577,0
+43530,0
+52963,0
+45559,0
+3972,0
+55068,0
+8252,0
+34516,0
+15671,0
+2758,0
+35074,0
+49091,0
+36281,0
+46022,0
+6848,0
+44300,0
+34155,0
+37205,0
+63317,0
+8882,0
+9567,0
+53202,0
+29663,0
+25700,0
+23169,0
+26955,0
+60685,1
+35491,0
+22979,0
+17434,0
+11657,0
+21256,0
+40932,0
+33304,0
+27292,0
+22600,0
+7317,0
+7245,0
+49594,0
+12840,0
+40977,0
+60028,0
+20526,0
+56749,1
+25738,0
+63343,0
+52966,0
+30058,0
+35089,0
+1842,0
+19529,0
+54068,0
+65058,0
+11676,0
+41430,0
+53177,0
+40392,0
+18998,0
+25535,0
+24624,0
+54092,0
+26064,0
+10101,0
+24432,0
+47992,0
+883,1
+28127,0
+47936,0
+60210,0
+17370,0
+17773,0
+2338,0
+59762,0
+16891,0
+58561,0
+64839,0
+12882,0
+43042,0
+36310,0
+45537,0
+52227,0
+47846,0
+22467,0
+6027,0
+55960,0
+7180,0
+48620,0
+34011,0
+57427,0
+41319,0
+25033,0
+10433,0
+54220,0
+3924,0
+27963,0
+51637,0
+23209,0
+50983,0
+536,0
+44594,0
+48280,0
+15152,0
+31368,0
+16544,0
+20400,0
+35343,0
+28727,0
+41628,0
+30834,0
+31140,0
+61638,0
+65506,0
+6,0
+20947,0
+43884,0
+36321,0
+18161,0
+29029,0
+3538,0
+37272,0
+41084,0
+18644,0
+64235,0
+53242,0
+48572,0
+55372,0
+27250,0
+30744,0
+23618,0
+16650,0
+39513,0
+3329,0
+43458,0
+48357,0
+28225,0
+35868,0
+65044,0
+2610,1
+53754,0
+31273,0
+52285,0
+6301,0
+30776,0
+26539,0
+54396,0
+64908,1
+38421,0
+17991,0
+57357,0
+11213,0
+16898,0
+43547,0
+36356,0
+6937,0
+62746,0
+27805,0
+813,0
+52536,0
+45999,0
+26152,0
+43759,0
+32804,0
+45376,0
+14426,0
+51911,0
+42552,0
+11631,0
+29875,0
+37163,0
+21708,0
+32560,0
+50755,0
+33437,0
+64397,1
+64191,0
+35371,0
+61196,0
+64091,0
+60407,0
+8255,0
+30487,0
+1804,0
+7008,0
+4170,0
+24245,0
+62228,0
+59543,0
+53104,0
+32697,0
+52337,0
+13408,0
+25986,0
+12652,0
+65211,0
+43008,0
+62669,0
+29069,0
+42985,0
+10313,0
+59849,0
+51118,0
+43520,0
+21525,0
+39725,0
+36273,0
+26569,0
+20236,0
+11621,0
+3384,0
+48378,0
+5660,0
+25788,0
+63371,0
+16336,0
+56154,0
+39382,0
+44862,0
+17107,0
+19596,0
+34863,0
+55171,0
+42266,0
+58365,0
+31192,0
+54531,0
+55781,0
+39676,0
+50510,0
+6289,0
+28594,0
+27547,0
+64817,0
+25138,0
+29729,0
+10326,0
+19388,0
+62261,0
+8947,0
+35958,0
+9329,0
+11371,0
+6052,0
+4904,0
+1649,0
+17479,0
+12345,0
+21770,0
+9295,0
+33830,0
+1421,0
+20125,0
+40412,0
+58669,0
+51856,0
+30686,0
+44978,0
+30619,0
+57687,0
+36257,0
+20004,0
+53508,0
+53860,0
+51672,0
+487,0
+9199,0
+12779,0
+7716,0
+40536,0
+3275,0
+50856,0
+44659,0
+35407,0
+8082,0
+33198,0
+19478,0
+21120,0
+41404,0
+12975,0
+58270,0
+7539,0
+60854,0
+32306,0
+59475,0
+43375,0
+36994,0
+40116,0
+15311,0
+3415,0
+2976,0
+62514,0
+44884,0
+37322,0
+63438,0
+17768,0
+51206,0
+31780,0
+50701,0
+48697,0
+36941,0
+61934,0
+43293,0
+16035,0
+11533,0
+41753,0
+35399,0
+27394,0
+52116,0
+60298,0
+62520,0
+39536,0
+64873,0
+25887,0
+61202,0
+36403,0
+51587,0
+15026,0
+54450,0
+18567,0
+17314,0
+14079,0
+31118,0
+60690,0
+46477,1
+53864,0
+32400,0
+37777,0
+64566,0
+46301,0
+35930,0
+18940,0
+34720,0
+3768,0
+18267,0
+44910,0
+72,0
+14356,0
+57252,0
+25366,0
+3456,0
+61886,0
+5633,0
+11563,0
+2217,0
+18283,0
+39459,0
+33105,0
+24640,0
+6651,0
+3936,0
+58908,0
+19617,0
+29175,0
+54156,0
+21163,0
+59268,0
+37463,0
+53116,0
+56321,0
+6612,0
+11912,0
+9451,0
+39674,0
+55552,0
+40740,0
+1155,0
+38659,0
+12499,0
+12733,0
+2382,0
+16484,0
+41040,0
+20367,0
+33291,0
+36188,0
+41689,0
+12741,0
+33613,0
+63733,0
+62903,0
+33905,0
+24724,0
+27904,0
+21596,0
+8901,0
+4410,0
+55186,0
+63578,0
+23782,0
+4807,0
+33809,0
+64973,1
+40918,0
+51514,0
+47793,0
+40555,0
+10184,0
+14066,0
+55228,0
+33501,0
+31117,1
+53142,0
+43334,0
+40728,0
+36787,0
+50762,0
+63887,1
+54705,0
+43622,0
+13827,0
+33821,0
+22319,0
+60935,0
+48274,0
+5323,0
+18014,0
+41726,0
+58239,0
+40527,0
+25665,0
+53050,0
+61705,0
+2216,0
+29782,0
+64589,0
+9569,0
+6343,0
+45234,0
+45964,0
+59234,0
+23699,0
+60037,0
+41846,0
+1372,0
+46919,0
+33951,0
+44967,0
+56601,0
+25971,0
+31601,0
+22476,0
+53163,0
+46521,0
+62562,0
+46672,0
+49652,0
+58108,0
+24086,0
+17252,0
+55183,0
+62836,0
+5512,0
+46184,0
+35838,0
+6539,0
+32456,0
+19371,0
+18095,0
+64193,0
+15997,0
+6085,0
+57692,0
+29553,0
+15327,0
+51945,0
+29925,0
+46448,0
+39184,0
+37778,0
+15286,0
+31556,0
+3697,0
+10864,0
+62246,0
+15353,0
+45632,0
+45350,0
+38398,0
+29255,0
+1057,0
+42946,0
+434,0
+21237,0
+18650,0
+3459,0
+34974,0
+21177,0
+35830,0
+62267,0
+4511,0
+12367,0
+44676,0
+63083,0
+46472,0
+35030,0
+23361,0
+50854,0
+295,0
+26868,0
+41175,0
+64334,0
+37080,0
+36182,0
+7927,0
+24815,0
+12670,0
+10042,0
+31993,0
+55363,0
+25629,0
+26922,0
+2808,0
+61696,0
+52499,0
+8716,0
+3913,0
+34493,0
+17856,0
+61119,0
+18028,0
+14317,0
+44583,0
+34200,0
+26942,0
+34944,0
+9603,0
+1900,0
+47579,0
+51418,0
+19183,0
+44865,0
+39211,0
+13229,0
+37258,0
+31486,0
+59778,0
+7658,0
+45460,0
+18824,0
+34752,0
+56855,0
+35950,0
+22738,0
+20429,0
+10308,0
+37473,0
+5041,0
+47568,0
+30241,0
+46000,0
+45085,0
+62407,0
+31421,0
+45384,0
+7774,0
+42583,0
+48898,0
+32214,0
+34584,0
+37004,0
+11969,0
+49599,0
+43685,0
+64624,0
+28718,0
+30684,0
+61714,0
+37464,0
+59935,0
+55042,0
+64244,0
+42915,0
+9244,0
+14411,0
+28677,0
+9787,0
+2400,0
+10935,0
+55005,0
+11904,0
+10669,0
+40190,0
+26725,0
+23789,0
+13329,0
+35508,0
+13914,0
+55078,0
+46886,0
+15627,0
+624,1
+852,0
+35230,0
+11337,0
+28348,0
+5299,0
+64850,0
+56990,0
+4600,0
+28774,0
+45794,0
+25370,0
+55945,0
+26370,0
+60669,0
+36365,0
+34523,0
+52312,0
+12181,0
+2660,0
+20699,0
+64840,0
+55388,0
+21236,0
+40782,0
+546,1
+16572,0
+27975,0
+8517,0
+55618,0
+54590,0
+30457,0
+47806,0
+63022,0
+61668,0
+33312,0
+61546,0
+65460,0
+58788,0
+10202,0
+248,0
+53833,0
+21345,0
+13814,0
+18211,0
+37047,0
+14855,0
+38081,0
+24270,0
+50505,0
+42173,0
+58203,0
+7682,0
+24219,0
+51055,0
+16421,0
+40387,0
+41629,0
+17798,0
+43333,0
+6465,0
+26912,0
+54884,0
+14508,0
+44759,0
+59626,0
+63082,0
+61280,0
+22507,0
+20174,0
+34582,0
+4074,0
+7910,0
+46029,0
+19750,0
+33735,0
+25856,0
+46127,0
+38194,0
+57496,0
+38129,0
+29527,0
+42724,0
+630,1
+53149,0
+64786,0
+59353,0
+48074,0
+18741,0
+54664,0
+47947,0
+6591,0
+12060,0
+46373,0
+54283,0
+13767,0
+31509,0
+40403,0
+41224,0
+48369,0
+13235,0
+40076,0
+64922,0
+46243,0
+58114,0
+48474,0
+4878,0
+34243,0
+10522,0
+64317,0
+51301,0
+40993,0
+63232,0
+16814,0
+28369,0
+15643,0
+39033,0
+14451,0
+49085,0
+10136,0
+8114,0
+61234,0
+60314,0
+2521,0
+31791,0
+25321,0
+6346,0
+8917,0
+24571,0
+37277,0
+63347,0
+3978,0
+1238,0
+11649,0
+20880,0
+22122,0
+33518,0
+2722,0
+55123,0
+13852,0
+5192,0
+17816,0
+30419,0
+12096,0
+38298,0
+21122,0
+16365,0
+60798,0
+41734,0
+49823,0
+41047,0
+25363,0
+36245,0
+30253,0
+37726,0
+37158,0
+15686,0
+39351,0
+19221,0
+45947,0
+42593,0
+13952,0
+32555,0
+14679,0
+53305,0
+43032,0
+58229,0
+51062,0
+54651,0
+46421,0
+5496,0
+30446,0
+35723,0
+11095,0
+8013,0
+65234,0
+59311,0
+58165,0
+13520,0
+49579,0
+64773,1
+17477,0
+710,0
+57863,0
+36051,0
+48541,1
+23717,0
+20033,0
+48967,0
+1366,0
+20192,0
+20044,0
+56089,0
+38233,0
+52832,0
+52501,0
+42589,0
+8900,0
+2288,0
+17457,0
+46880,0
+48072,0
+8159,0
+55853,0
+36516,0
+15308,0
+18617,0
+47067,0
+27460,0
+11767,0
+43071,0
+28163,0
+56605,0
+50103,0
+51897,0
+47125,0
+35189,0
+27740,0
+64881,0
+110,0
+41235,0
+48170,0
+32577,0
+63635,0
+11043,0
+47059,0
+33265,0
+39018,0
+60434,0
+63203,0
+63176,0
+11269,0
+29000,0
+40904,0
+52866,0
+61593,0
+61856,0
+50175,0
+44463,0
+4291,0
+8182,0
+4503,0
+47914,0
+9895,0
+10641,0
+54745,0
+59633,0
+40927,0
+55477,0
+38156,0
+591,0
+53500,0
+57213,0
+49584,0
+37136,0
+29178,0
+1536,0
+16464,0
+13220,0
+5018,0
+53621,0
+41401,0
+48014,0
+19050,0
+58127,0
+48102,0
+5182,0
+9947,0
+23686,0
+59385,0
+26505,0
+11871,0
+13246,0
+14463,0
+43869,0
+45367,0
+47300,0
+14208,0
+7495,0
+27458,0
+46724,0
+26307,0
+31222,0
+12321,0
+17569,0
+44897,0
+22169,0
+50654,0
+45054,0
+39247,0
+16874,0
+63979,0
+31167,0
+40659,0
+59494,0
+62742,0
+51347,0
+33338,0
+17335,0
+62424,0
+61708,0
+43063,0
+37098,0
+631,1
+16977,0
+19975,0
+48039,0
+31987,0
+53744,0
+17325,0
+8229,0
+36349,0
+60050,0
+35553,0
+38045,0
+21696,0
+32838,0
+28955,0
+26308,0
+33446,0
+8300,0
+4985,0
+63815,0
+47707,0
+59126,0
+35592,0
+53998,0
+32178,0
+20988,0
+60481,0
+57449,0
+45751,0
+8086,0
+15223,0
+11485,0
+8522,0
+1212,0
+60937,0
+19237,0
+30808,0
+33214,0
+43705,0
+64861,0
+56058,0
+37645,0
+40354,0
+58047,0
+58183,0
+7189,0
+26224,0
+8998,0
+45244,0
+35409,0
+39040,0
+65126,0
+50520,0
+20701,0
+62120,0
+60492,0
+50316,0
+2239,0
+10867,0
+63002,0
+17729,0
+19245,0
+57916,0
+29311,0
+22091,0
+26895,0
+12214,0
+61966,0
+48182,0
+44822,0
+45898,0
+32542,0
+20834,0
+59255,0
+45212,0
+7389,0
+57476,0
+20420,0
+32474,0
+1945,0
+7969,0
+49207,0
+5622,0
+24064,0
+37734,0
+46756,0
+48938,0
+28760,0
+526,0
+1648,1
+19688,0
+40128,0
+23652,0
+47871,0
+34379,0
+63513,0
+59052,0
+4763,0
+30469,0
+41288,0
+51886,0
+25864,0
+22650,0
+54931,0
+1349,0
+17631,0
+16294,0
+61390,0
+409,0
+41980,0
+33324,0
+23767,0
+3506,0
+17316,0
+51638,0
+59392,0
+30093,1
+20738,0
+31130,0
+56902,0
+14834,0
+45587,0
+17078,0
+43072,0
+52623,0
+52675,0
+40726,0
+28784,0
+31563,0
+37832,0
+35374,0
+52327,0
+35485,0
+8638,0
+9502,0
+61049,0
+64495,0
+21488,0
+24282,0
+46829,0
+8915,0
+9991,0
+48407,0
+56475,0

quantum_perceptron/train/__init__.py ADDED Viewed

	@@ -0,0 +1,2 @@


1	+ from quantum_perceptron.train.data_gen import *
2	+ from quantum_perceptron.train.training import *

quantum_perceptron/train/data_gen.py ADDED Viewed

	@@ -0,0 +1,90 @@

+import numpy as np
+from tqdm import tqdm
+from quantum_perceptron.utils import (
+    calculate_succ_probability
+)
+from quantum_perceptron.perceptron import Perceptron
+def generate_training_samples(data: np.ndarray,
+                              num_positive_samples: int,
+                              num_negative_samples: int) -> np.ndarray:
+    """
+    From the entire dataset, generate training samples.
+    """
+    pos_inds = np.where(data[:, 1] == 1)[0]
+    neg_inds = np.where(data[:, 1] == 0)[0]
+    if len(pos_inds) < num_positive_samples:
+        num_positive_samples = len(pos_inds)
+    if len(neg_inds) < num_negative_samples:
+        num_negative_samples = len(neg_inds)
+    sampled_neg_inds = np.random.choice(neg_inds,
+                                        num_negative_samples,
+                                        replace=False)
+    sampled_pos_inds = np.random.choice(pos_inds,
+                                        num_positive_samples,
+                                        replace=False)
+    new_data = np.vstack((data[sampled_pos_inds], data[sampled_neg_inds]))
+    np.random.shuffle(new_data)
+    return new_data
+def generate_dataset(num_qubits: int = 4,
+                     fixed_weight: int = 626,
+                     dir_path: str = './data/',
+                     threshold: float = 0.5,
+                     num_runs: int = 8192,
+                     create_training_samples: bool = True,
+                     num_pos_train_samples: int = 50,
+                     num_neg_train_samples: int = 3000):
+    """
+    Generate training dataset with fixed weight.
+    Args:
+        num_qubits: `int` representing number of qubits.
+        fixed_weight: `int` representing the fixed weight value.
+        dir_path: `str` representing the directory path.
+    """
+    num_samples = np.power(2, np.power(2, num_qubits))
+    data = np.empty([num_samples, 2], dtype=np.int64)
+    p = Perceptron(num_qubits, fixed_weight, 0)
+    for i in tqdm(range(num_samples)):
+        p.input = i
+        p.build_circuit()
+        prob = calculate_succ_probability(p.measure_circuit(num_runs))
+        if prob > threshold:
+            label = 1
+        else:
+            label = 0
+        data[i][0] = i
+        data[i][1] = label
+    print("Number of positive samples: {}".format(
+        np.sum(data[:, 1] == 1)
+    ))
+    print("Number of negative samples: {}".format(
+        np.sum(data[:, 1] == 0)
+    ))
+    filename = 'sample_space_qubits_{}_fweight_{}.txt'.format(
+        num_qubits, fixed_weight
+    )
+    np.savetxt(dir_path + filename, data, fmt='%i,%i', delimiter=',')
+    print('Saved data to {}'.format(dir_path + filename))
+    if create_training_samples:
+        train_data = generate_training_samples(
+            data, num_pos_train_samples, num_neg_train_samples
+        )
+        train_filename = 'train_space_qubits_{}_fweight_{}.txt'.format(
+            num_qubits, fixed_weight
+        )
+        np.savetxt(dir_path + train_filename,
+                   train_data,
+                   fmt='%i,%i',
+                   delimiter=',')
+        print('Saved training data to {}'.format(dir_path + train_filename))

quantum_perceptron/train/training.py ADDED Viewed

	@@ -0,0 +1,164 @@

+import os
+from typing import List
+import numpy as np
+from tqdm import tqdm
+import wandb
+from quantum_perceptron.utils import (
+    get_vector_from_int,
+    get_int_from_vector,
+    calculate_succ_probability
+)
+from quantum_perceptron import Perceptron
+class PerceptronTrainer:
+    def __init__(self,
+                 num_qubits: int,
+                 fixed_weight: int,
+                 dataset_path: str,
+                 threshold: float = 0.5,
+                 num_runs: int = 8192,
+                 learning_rate_pos: float = 0.5,
+                 learning_rate_neg: float = 0.5):
+        """
+        This class is used to train the perceptron.
+        Args:
+          num_qubits: `int` representing number of qubits.
+          fixed_weight: `int` representing the fixed weight value.
+          dataset_path: `str` representing the path to the dataset.
+          threshold: `float` representing the threshold value.
+          num_runs: `int` representing number of runs.
+          learning_rate_pos: `float` representing the learning rate for positve
+            samples.
+          learning_rate_neg: `float` representing the learning rate for
+            negativesamples.
+        """
+        self.num_qubits = num_qubits
+        self.fixed_weight = fixed_weight
+        assert os.path.exists(dataset_path), "Dataset path does not exist"
+        self.data = self.read_dataset(dataset_path)
+        self.threshold = threshold
+        self.num_runs = num_runs
+        self.learning_rate_pos = learning_rate_pos
+        self.learning_rate_neg = learning_rate_neg
+        self.perceptron = Perceptron(num_qubits)
+        self.accumulate_loss: List[float] = []
+        self.num_steps = 0
+        # Initializing random weight for the training
+        self.weight_variable = np.random.randint(
+            np.power(2, np.power(2, num_qubits)))
+        wandb.init(
+            project="quantum-perceptron",
+            config={
+                "learning_rate_pos": learning_rate_pos,
+                "learning_rate_neg": learning_rate_neg,
+                "fixed_weight": fixed_weight,
+                "num_qubits": num_qubits,
+                "dataset": dataset_path,
+                "num_runs": num_runs,
+                "threshold": threshold
+            }
+        )
+    def read_dataset(self, filepath: str) -> np.ndarray:
+        """
+        Read dataset from file.
+        """
+        return np.loadtxt(filepath, dtype=np.int64, delimiter=',')
+    def invert_non_matching_bits(self, input: int):
+        """
+        Invert non-matching positions in vector.
+        """
+        input_vector = get_vector_from_int(input, self.num_qubits)
+        weight_vector = get_vector_from_int(self.weight_variable,
+                                            self.num_qubits)
+        non_match_ids = np.where(input_vector != weight_vector)[0]
+        num_select = int(np.ceil(len(non_match_ids) * self.learning_rate_pos))
+        selected_ids = np.random.choice(non_match_ids,
+                                        num_select,
+                                        replace=False)
+        for id in selected_ids:
+            weight_vector[id] *= -1
+        self.weight_variable = get_int_from_vector(weight_vector,
+                                                   self.num_qubits)
+    def invert_matching_bits(self, input: int):
+        """
+        Invert matching positions in vector.
+        """
+        input_vector = get_vector_from_int(input, self.num_qubits)
+        weight_vector = get_vector_from_int(self.weight_variable,
+                                            self.num_qubits)
+        match_ids = np.where(input_vector == weight_vector)[0]
+        num_select = int(np.ceil(len(match_ids) * self.learning_rate_neg))
+        selected_ids = np.random.choice(match_ids,
+                                        num_select,
+                                        replace=False)
+        for id in selected_ids:
+            weight_vector[id] *= -1
+        self.weight_variable = get_int_from_vector(weight_vector,
+                                                   self.num_qubits)
+    def calc_loss(self):
+        """
+        Note that we will only use this loss to generate the plot
+        and not for training the perceptron.
+        """
+        self.perceptron.input = self.weight_variable
+        self.perceptron.weight = self.fixed_weight
+        self.perceptron.build_circuit()
+        loss = calculate_succ_probability(
+            self.perceptron.measure_circuit(self.num_runs))
+        return loss
+    def train_step(self, input: int, label: int):
+        """
+        Training step for a single sample.
+        """
+        self.perceptron.input = input
+        self.perceptron.weight = self.weight_variable
+        self.perceptron.build_circuit()
+        prob = calculate_succ_probability(
+            self.perceptron.measure_circuit(self.num_runs))
+        loss = self.calc_loss()
+        self.accumulate_loss.append(loss)
+        self.num_steps += 1
+        if int(loss) == 1:
+            print("Training converged at step: {}".format(self.num_steps))
+            return True
+        if prob > self.threshold:
+            pred = 1
+        else:
+            pred = 0
+        if label == 1 and pred == 0:
+            self.invert_non_matching_bits(input)
+            wandb.log({"probability": loss, "weight": self.weight_variable})
+        elif label == 0 and pred == 1:
+            self.invert_matching_bits(input)
+            wandb.log({"probability": loss, "weight": self.weight_variable})
+        return False
+    def train_epoch(self, epoch: int):
+        """
+        Train the epoch.
+        """
+        for i in tqdm(range(self.data.shape[0])):
+            input = self.data[i, 0]
+            label = self.data[i, 1]
+            converged = self.train_step(input, label)
+            if converged:
+                return True
+        return False
+    def train(self, num_epochs: int):
+        """
+        Train the perceptron.
+        """
+        for i in range(num_epochs):
+            converged = self.train_epoch(i)
+            if converged:
+                break

quantum_perceptron/utils/__init__.py CHANGED Viewed

@@ -1,2 +1,3 @@
 from quantum_perceptron.utils.data_utils import *
 from quantum_perceptron.utils.quantum_utils import *

 from quantum_perceptron.utils.data_utils import *
 from quantum_perceptron.utils.quantum_utils import *
+from quantum_perceptron.utils.plot_utils import *

quantum_perceptron/utils/data_utils.py CHANGED Viewed

@@ -35,7 +35,7 @@ def get_vector_from_int(data: int, num_qubits: int) -> np.ndarray:
     Args:
       data: `int` representing data value
-        (correspponding toinput or weight vector)
       num_qubits: `int` representing number of qubits.
     Returns: Vector in  form of `np.ndarray`.
@@ -52,6 +52,30 @@ def get_vector_from_int(data: int, num_qubits: int) -> np.ndarray:
     return data_vector
 def get_possible_state_strings(num_bits: int) -> np.ndarray:
     """
     Get all the state bit strings corresponding to given number of bits.
@@ -100,3 +124,20 @@ def get_ones_counts_to_states(states: np.ndarray) -> Dict[int, List[int]]:
         ones_count[ct].append(i)
     return ones_count

     Args:
       data: `int` representing data value
+        (correspponding to input or weight vector)
       num_qubits: `int` representing number of qubits.
     Returns: Vector in  form of `np.ndarray`.
     return data_vector
+def get_int_from_vector(data_vector: np.ndarray, num_qubits: int) -> int:
+    """
+    Get the integer data value from the vector.
+    Args:
+      data_vector: `np.ndarray` representing the vector.
+      num_qubits: `int` representing number of qubits.
+    Returns: `int` representing the data value.
+    """
+    if len(data_vector) != np.power(2, num_qubits):
+        raise ValueError("The vector length is not equal to 2^num_qubits")
+    data_bin_vec = []
+    for i, val in enumerate(data_vector):
+        if val == -1:
+            data_bin_vec.append('1')
+        elif val == 1:
+            data_bin_vec.append('0')
+    data_bin = ''.join(data_bin_vec)
+    return int(data_bin, 2)
 def get_possible_state_strings(num_bits: int) -> np.ndarray:
     """
     Get all the state bit strings corresponding to given number of bits.
         ones_count[ct].append(i)
     return ones_count
+def calculate_succ_probability(counts: Dict[str, int]) -> float:
+    """
+    Calculate the success probability from the counts of the states.
+    Args:
+      counts: `dict` containing the counts of the states.
+    Returns: `float` representing the success probability.
+    """
+    if len(counts) == 0:
+        raise ValueError("The counts dict is empty")
+    total_count = sum(counts.values())
+    succ_count = counts.get('1', 0)
+    return succ_count / total_count

quantum_perceptron/utils/plot_utils.py ADDED Viewed

	@@ -0,0 +1,41 @@

+import numpy as np
+import matplotlib.pyplot as plt
+from quantum_perceptron.utils.data_utils import (
+    get_bin_int,
+    assert_bits,
+    assert_negative
+)
+def get_img_from_data(data: int, num_qubits: int) -> np.ndarray:
+    """
+    Get n x n matrix representing the image of the data where n is
+    num_qubits.
+    Args:
+      data: `int` representing data value
+        (correspponding to input or weight vector)
+      num_qubits: `int` representing number of qubits.
+    Returns: Image in  form of `np.ndarray`.
+    """
+    assert_negative(data)
+    assert_bits(data, num_qubits)
+    bin_str = get_bin_int(data, num_qubits)
+    img = np.zeros((np.power(2, num_qubits)))
+    for i, bit in enumerate(bin_str):
+        if bit == '0':
+            img[i] = 255
+    return img.reshape((num_qubits, num_qubits))
+def plot_img_from_data(data: int, num_qubits: int):
+    """
+    Plot image from data.
+    """
+    img = get_img_from_data(data, num_qubits)
+    ax = plt.imshow(img, cmap='gray')
+    ax.axes.xaxis.set_visible(False)
+    ax.axes.yaxis.set_visible(False)

requirements.txt CHANGED Viewed

@@ -3,4 +3,5 @@ qiskit
 pycodestyle
 pytest
 mypy
-gradio

 pycodestyle
 pytest
 mypy
+gradio
+wandb