OccupationDistance.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "873b7fa4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# packages\n",
    "import numpy as np  # matrix and vector operations\n",
    "import matplotlib.pyplot as plt  # visualization\n",
    "#import fssa as fssa  # scaling collapse \n",
    "from utils import *  # all things ID of MBL\n",
    "from time import time\n",
    "import glob\n",
    "import numpy.ma as ma\n",
    "from plotting import *\n",
    "import scipy as sp\n",
    "import seaborn as sns\n",
    "from tqdm import tqdm\n",
    "\n",
    "import matplotlib as mpl\n",
    "fig_size_dim    = 4\n",
    "golden_ratio    = (1+np.sqrt(5))/2\n",
    "fig_size        = (fig_size_dim, fig_size_dim/golden_ratio)\n",
    "\n",
    "def plot_style():\n",
    "    font_size       = 12\n",
    "    dpi             = 200\n",
    "\n",
    "    params = {'figure.figsize': fig_size,\n",
    "              'figure.dpi': dpi,\n",
    "              'savefig.dpi': dpi,\n",
    "              'font.size': font_size,\n",
    "              'font.family': \"sans-serif\",\n",
    "              'font.sans-serif': [\"Helvetica\"],\n",
    "              'figure.titlesize': font_size,\n",
    "              'legend.fontsize': font_size,\n",
    "              'axes.labelsize': font_size,\n",
    "              'axes.titlesize': font_size,\n",
    "              'xtick.labelsize': font_size,\n",
    "              'ytick.labelsize': font_size,\n",
    "              'text.usetex': True,\n",
    "             }\n",
    "\n",
    "    plt.rcParams.update(params)\n",
    "plot_style()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "4d4355e3-cf06-4d41-ab4b-f35d13108a1b",
   "metadata": {},
   "outputs": [],
   "source": [
    "## Global variables\n",
    "# List of available system sizes\n",
    "Ls = [8, 10, 12, 14]\n",
    "# List of available disorder strengths\n",
    "Ws = np.concatenate([np.arange(1,2.6,0.2), np.arange(2.6,4.55,0.05), np.arange(4.7,6.2,0.2)])\n",
    "\n",
    "# Download data\n",
    "# TODO (curl from ERDA)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "id": "7b5187b0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'0011': 0, '0101': 1, '0110': 2, '1001': 3, '1010': 4, '1100': 5}\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[0., 0., 0., 0., 0., 0.],\n",
       "       [0., 1., 0., 0., 0., 0.],\n",
       "       [0., 0., 1., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 1.]])"
      ]
     },
     "execution_count": 164,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def ODPM(i, j, L):\n",
    "    num_states = binomial(L)\n",
    "    C = np.zeros((num_states,num_states))\n",
    "    (s2i, i2s) = basisStates(L)\n",
    "    \n",
    "   # print(s2i)\n",
    "    \n",
    "    for key in s2i.keys():\n",
    "#         if (key[i] == '1') and (key[j] == '0'):\n",
    "#             new_state = [x for x in key]\n",
    "#             new_state[i] = '0'\n",
    "#             new_state[j] = '1'\n",
    "#             new_state = ''.join(new_state)\n",
    "#             C[s2i[new_state], s2i[key]], C[s2i[key], s2i[new_state]] = 1, 1\n",
    "            \n",
    "        if( i == j ):\n",
    "            if (key[i] == '1'):\n",
    "                C[ s2i[key], s2i[key]] = 1\n",
    "            if (key[i] == '0'):\n",
    "                C[ s2i[key], s2i[key]] = 0\n",
    "            \n",
    "    return C\n",
    "\n",
    "def n(i, L):\n",
    "    num_states = binomial(L)\n",
    "    C = np.zeros((num_states,num_states))\n",
    "    (s2i, i2s) = basisStates(L)\n",
    "    print(s2i)\n",
    "    for key in s2i.keys():\n",
    "        if (key[i] == '1'):\n",
    "            C[ s2i[key], s2i[key]] = 1\n",
    "            \n",
    "    return C\n",
    "\n",
    "n(1,4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "id": "5b86291b-172a-4f6a-98e5-6d838ce42721",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "from utils import *\n",
    "\n",
    "result = {}\n",
    "num_seeds = 100\n",
    "tmp_ws = [1, 12]\n",
    "\n",
    "for L in [10]:\n",
    "    \n",
    "    ODPMs = {}\n",
    "    for i in range(L):\n",
    "        ODPMs[i] = {}\n",
    "        for j in range(L):\n",
    "            ODPMs[i][j] = ODPM(i,j,L)\n",
    "    \n",
    "    result[L] = {}\n",
    "    for W in tmp_ws:\n",
    "        result[L][W] = []\n",
    "        for seed in range(num_seeds):    \n",
    "            H = constructHamiltonian(L = L, W = W, U = 1.0, t = 1.0, seed=seed)\n",
    "            eigvals, eigvecs = np.linalg.eigh(H)\n",
    "            \n",
    "            #print(eigvals)\n",
    "            epsilon = 2*(eigvals - np.min(eigvals))/(np.max(eigvals) - np.min(eigvals))\n",
    "            #print(epsilon)\n",
    "            \n",
    "            idx = np.argmin( np.abs(epsilon - 1) )\n",
    "            ev = eigvecs[:,idx]\n",
    "            \n",
    "            rho = np.zeros( (L, L) )\n",
    "            for i in range(L):\n",
    "                for j in range(L):\n",
    "                    rho[i,j] = np.dot(ev.conj().T, np.dot(ODPMs[i][j], ev))\n",
    "                    \n",
    "            # Diagonalie ODPM\n",
    "            occupations, _ = np.linalg.eigh(rho)\n",
    "        \n",
    "            result[L][W].append( occupations )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "id": "490ab992-16fc-4e33-b2f0-ce91f85a9b95",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([347.,  60.,  38.,  33.,  23.,  24.,  29.,  41.,  65., 340.]),\n",
       " array([6.28127323e-05, 1.00054299e-01, 2.00045785e-01, 3.00037271e-01,\n",
       "        4.00028756e-01, 5.00020242e-01, 6.00011728e-01, 7.00003214e-01,\n",
       "        7.99994700e-01, 8.99986186e-01, 9.99977672e-01]),\n",
       " <BarContainer object of 10 artists>)"
      ]
     },
     "execution_count": 172,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x494.427 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plt.hist( np.array([result[10][1][a] for a in range(1,num_seeds)]).flatten() )\n",
    "plt.hist( np.array(result[10][12]).flatten() )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "id": "bf1ac358",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fe10809f050>]"
      ]
     },
     "execution_count": 173,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x494.427 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot( np.sort(np.mean(result[10][1], axis=0))[::-1] )\n",
    "plt.plot( np.sort(np.mean(result[10][12], axis=0))[::-1] )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "02c4f6ce",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "40d3a090f54c6569ab1632332b64b2c03c39dcf918b08424e98f38b5ae0af88f"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}