{ "cells": [ { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": false }, "outputs": [], "source": [ "%matplotlib inline" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\nCreated on Tue Nov 9 19:03:18 2021\n\n@author: maout\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": false }, "outputs": [], "source": [ "import numpy as np\n\n\n#from matplotlib import pyplot as plt\n\nfrom scipy.spatial.distance import cdist\nimport time\nimport ot\nimport numba\nimport math\nimport random\nimport sys\nimport pickle\n\ndim = 2\nh = 0.001\nt1 = 0\nt2 = 1.5\nT = t2-t1\ntimegrid = np.arange(0,T+h/2,h)\nN = 2000#0#0#0#2000\n#g = 1\nk = timegrid.size\nM = 80\n\ny2 = np.ones(dim) \nsigmas = np.array([0.1,0.5,1])\nrep_bridge = 5 #10 different bridge instances for every setting\nreps = 20 ##instanses for stochastic path evaluation of each bridge\n# dy0 = np.array([np.pi/4, np.pi/2, np.pi, 3*np.pi/2]) \n# y0s = np.zeros((dim,rep_bridge, dy0.size))\nrandom.seed(22)\n# for i in range(rep_bridge):\n# y0s[0, i, :] = np.random.uniform( low=0, high=2*np.pi,size=1 ) \n# y0s[1, i, :] = (y0s[0, i, :] + dy0) %(2* np.pi)\n\n\n\nKs = np.linspace(0,4,11) \ndy0 = np.array([np.pi/4, np.pi/2]) \ny0s = np.zeros((dim,rep_bridge, dy0.size))\nRttcont = np.zeros(( rep_bridge, Ks.size, dy0.size, sigmas.size, timegrid.size,reps ))*np.nan\nRttnon = np.zeros(( rep_bridge, Ks.size, dy0.size, sigmas.size, timegrid.size,reps ))*np.nan\nused_us = np.zeros((dim, rep_bridge, Ks.size, dy0.size, sigmas.size,timegrid.size, reps ))*np.nan\nbis = [2,3,4]\nfor bi in bis:#range(rep_bridge):\n ###loop over bridge iterations\n \n #for ki,K in enumerate(Ks): ##loop over interaction strengths\n for ki,K in enumerate(Ks):\n \n \n for yi in range(dy0.size): ###loop over initial distances\n \n \n \n for gii,g in enumerate(sigmas):\n \n #naming = 'kuramot\\\\2N_systematic_Kuramoto_homogeneous_repb_%d_ki_%d_yi_%d__gi_%d_N_%d_M_%d'%(bi, ki,yi,gii,N, M)\n naming = 'kurgamot\\\\New_2N_systematic_Kuramoto_homogeneous_repb_%d_ki_%d_yi_%d__gi_%d_N_%d_M_%d'%(bi, ki,yi,gii,N, M)\n \n try:\n file = open(naming+'.dat','rb')\n to_save = pickle.load(file) \n # naming2 = 'kuramot\\\\2N_systematic_Kuramoto_homogeneous_repb_%d_ki_%d_yi_%d__gi_%d_N_%d_M_%d_UNCONTROLLED'%(bi, ki,yi,gii,N, M)\n # file2 = open(naming2+'.dat','rb')\n # to_save2 = pickle.load(file2)\n \n #Fcont = to_save['Fcont'] \n Rttcont[bi,ki,yi, gii,:,:] = to_save['Rttcont'] \n #Fnon = to_save2['Fnon'] \n Rttnon[bi,ki,yi, gii,:,:] = to_save['Rttnon'] \n Kin = to_save['K'] \n y0s[:, bi, yi] = to_save['y0'] \n used_us[:,bi,ki,yi, gii,:] = to_save['used_us'] \n except FileNotFoundError:\n i=0\n print(naming)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": false }, "outputs": [], "source": [ "from scipy.spatial.distance import cdist\nimport time\nimport ot\nimport numba\nimport math\nimport random\nimport sys\nimport pickle\n\ndim = 2\nh = 0.001\nt1 = 0\nt2 = 1.5\nT = t2-t1\ntimegrid = np.arange(0,T+h/2,h)\nN = 2000#0#0#0#2000\n#g = 1\nk = timegrid.size\nM = 80\nbi = 2\nki = 3\ngii = 2\nyi = 1\ng=1\nnaming = 'kurgamot\\\\New_2N_systematic_Kuramoto_homogeneous_repb_%d_ki_%d_yi_%d__gi_%d_N_%d_M_%d'%(bi, ki,yi,gii,N, M)\n\nfile = open(naming+'.dat','rb')\nto_save = pickle.load(file) \nRttconti = to_save['Rttcont'] \nFcont = to_save['Fcont']\nFnon = to_save['Fnon'] \nRttnoni = to_save['Rttnon'] \nused_u = to_save['used_us']" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": false }, "outputs": [], "source": [ "import seaborn as sns\nfrom matplotlib import pyplot as plt\nrepi = 0\npali = sns.diverging_palette(145, 300, s=60, as_cmap=True)\nfrom mpl_toolkits.axes_grid1.inset_locator import zoomed_inset_axes, mark_inset\nimport matplotlib.gridspec as gridspec\nplt.figure(figsize= (8,2.)),\nplt.subplot(1,3,1),\n#cols = [pali(0), pali(0.99) ]\n#cols2 = [pali(0.3),pali(0.7)]\ncols = [pali(0.99), pali(0.8) ]\ncols2 = [pali(0.),pali(0.2)]\nfor di in range(dim):\n plt.plot(timegrid[:-1],Fcont[di,:-1,repi],'-',lw=4,c=cols[di],label=r'$\\theta_{%d}$'%(di+1))\n plt.plot(timegrid[:-1],Fnon[di,:-1,repi],'.',lw=3,c=cols2[di],alpha=0.7,zorder=0)\nfor di in range(dim):\n plt.plot(timegrid[:-1],Fnon[di,:-1, repi],'-',lw=3,c=cols2[di],zorder=0,label=r'$\\theta_{%d}$'%(di+1)) #this line is only for the legend entry\n \n \nplt.xlabel('time')\nplt.ylabel(r'phase $\\theta$')\nax = plt.gca()\nax.tick_params(axis='both',which='both',direction='out', length=3, width=1,colors='#4f4949',zorder=3)\nax.tick_params(bottom=True, top=False, left=True, right=False)\n\nax.set_yticks([0,np.pi/2,np.pi, 3*np.pi/2, 2*np.pi])\nax.set_yticklabels([r'$0$',r'$\\pi/2$',r'$\\pi$', r'${3 \\pi}/{2}$', r'$2 \\pi$'])\n\n\n# legend = plt.legend(frameon = 1,prop={'size': 15})\n# frame = legend.get_frame()\n# frame.set_facecolor('white')\n# frame.set_edgecolor('white')\nplt.xticks([0,0.5,1,1.5])\nlegend = ax.legend() \nlegend.get_frame().set_linewidth(1.8)\nlegend.get_frame().set_facecolor('white')\nlegend.get_frame().set_edgecolor('white')\nhandles, labels = ax.get_legend_handles_labels()\n\nif True:#g==0.5:\n ax.legend(handles, labels, title='controlled uncontrolled',\n handletextpad=0.5, columnspacing=3.2,handlelength=0.8, bbox_to_anchor=[0.0, 0.785],\n loc=3, ncol=2, frameon=True,fontsize = 'small',shadow=None,framealpha =0,edgecolor ='#0a0a0a')\n \nelse:\n ax.legend(handles, labels, title='controlled uncontrolled',\n handletextpad=0.5, columnspacing=3.2,handlelength=0.8,\n loc=3, ncol=2, frameon=True,fontsize = 8,shadow=None,framealpha =0,edgecolor ='#0a0a0a')\n \nplt.setp(plt.gca().get_legend().get_title(), fontsize='10')" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": false }, "outputs": [], "source": [ "plt.subplot(1,3,2),\nplt.plot(timegrid[:-1],Rttnoni[:-1,repi],lw=3,c= cols2[0],zorder=3,label='uncontrolled'),\nplt.plot(timegrid[:-1],Rttconti[:-1,repi],lw=3, c=cols[0],zorder=4,label='controlled'),\n\nplt.plot([0,timegrid[-1]], [1/np.sqrt(dim),1/np.sqrt(dim) ],linestyle=(0,(4,3)),lw=3, c='#ee9222',alpha=0.5,label='independent'),\nplt.plot([0,timegrid[-1]], [1,1 ],linestyle=(0,(4,3)),c='grey',lw=3,dash_capstyle = \"round\"),\nplt.xlabel('time')\nplt.xticks([0,0.5,1,1.5])\nplt.ylabel(r'order param. $R$') \nax7 = plt.gca()\nax7.tick_params(axis='both',which='both',direction='out', length=3, width=1,colors='#4f4949',zorder=3)\nax7.tick_params(bottom=True, top=False, left=True, right=False)\n\nlegend = ax7.legend() \nlegend.get_frame().set_linewidth(1.8)\nlegend.get_frame().set_facecolor('white')\nlegend.get_frame().set_edgecolor('white')\nhandles, labels = ax7.get_legend_handles_labels()\nhandles = [ handles[-2],handles[0] , handles[-1]]\nlabels= [ labels[-2],labels[0] , labels[-1]]\nif True:\n ax7.legend(handles, labels, \n handletextpad=0.5, columnspacing=0.3,handlelength=0.5, bbox_to_anchor=[-0.4, 0.99],\n loc=3, ncol=3, frameon=True,fontsize = 'small',shadow=None,framealpha =0,edgecolor ='#0a0a0a') \n\n\n###########\n# axins = zoomed_inset_axes(ax,2, loc=1, bbox_to_anchor=(2400,800))\n# mark_inset(ax, axins, loc1=2, loc2=1, fc=\"none\", ec=\"0.5\")\n# axins.set_xlim([0.5,1.2])\n# axins.set_ylim([0.95,1.01])\n\n# # # Plot zoom window\n# axins.plot(timegrid[:tti],Rtt[:tti],lw=3,c= cols2[0],zorder=3),\n# axins.plot(timegrid[:tti],Rttcont[:tti],lw=3, c=cols[0],zorder=4),\n# axins.plot([0,timegrid[tti]], [1,1 ],linestyle=(0,(4,3)),c='grey',lw=3,dash_capstyle = \"round\"),\n# axins.tick_params(bottom=False, top=False, left=True, right=False)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": false }, "outputs": [], "source": [ "plt.subplot(1,3,3),\n#for ti in range(2):\nfor di in range(dim):\n plt.plot(timegrid[:],h*g**2*used_u[di,:,repi],lw=2.5, c=cols[di],label=r'$u_{%d}$'%(di+1))\nplt.xticks([0,0.5,1,1.5])\nplt.xlabel('time')\nplt.ylabel(r'control $u$') \nax = plt.gca()\nax.tick_params(axis='both',which='both',direction='out', length=3, width=1,colors='#4f4949',zorder=3)\nax.tick_params(bottom=True, top=False, left=True, right=False)\nlegend = ax.legend() \nlegend.get_frame().set_linewidth(1.8)\nlegend.get_frame().set_facecolor('white')\nlegend.get_frame().set_edgecolor('white')\nhandles, labels = ax.get_legend_handles_labels()\nax.legend(handles, labels, title=None,\n handletextpad=0.5, columnspacing=0,handlelength=1, bbox_to_anchor=[0.50, 0.35],\n loc=1, ncol=1, frameon=True,fontsize = 'small',shadow=None,framealpha =0,edgecolor ='#0a0a0a')\n\n\nnaming2 = 'Kuramoto_2N_trajectories' \n##%%\n#plt.savefig( 'naming.png', bbox_inches='tight')\n#plt.savefig( 'naming.pdf', bbox_inches='tight')\nplt.savefig( naming2+'.png', bbox_inches='tight')\nplt.savefig(naming2 +'.pdf', bbox_inches='tight')" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": false }, "outputs": [], "source": [ "from matplotlib import pyplot as plt \n\n#figure and plotting settings\nimport seaborn as sns\nplt.rcParams['figure.dpi'] = 300\nplt.rcParams['savefig.dpi'] = 300\nplt.rcParams['savefig.facecolor'] = (1,1,1,0)\nplt.rcParams['savefig.bbox'] = \"tight\"\nplt.rcParams['font.size'] = 10*1.2\n# plt.rcParams['font.family'] = 'sans-serif' # not available in Colab\n# plt.rcParams['font.sans-serif'] = 'Helvetica' # not available in Colab\nplt.rcParams['pdf.fonttype'] = 42\nplt.rcParams['xtick.labelsize'] = 10*1.2\nplt.rcParams['ytick.labelsize'] = 10*1.2\nplt.rcParams['axes.labelsize'] = 12*1.2\nplt.rcParams['axes.linewidth'] = 2\nplt.rcParams['axes.spines.top'] = False\nplt.rcParams['axes.spines.right'] = False\n#plt.rcParams['axes.Axes.tick_params'] = True\n#plt.rcParams['axes.solid_capstyle'] = 'round'\nplt.rc('axes',edgecolor='#4f4949')\nplt.rcParams['figure.frameon'] = False\nplt.rcParams['figure.subplot.hspace'] = 0.51\nplt.rcParams['figure.subplot.wspace'] = 0.51\nplt.rcParams['figure.subplot.left'] = 0.1\nplt.rcParams['figure.subplot.right'] = 0.9\nplt.rcParams['figure.subplot.top'] = 0.9\nplt.rcParams['figure.subplot.bottom'] = 0.1\nplt.rcParams['lines.solid_capstyle'] = 'round'\nplt.rcParams['lines.solid_joinstyle'] = 'round'\n#plt.rcParams['xtick.major.size'] = 20\n#plt.rcParams['xtick.major.width'] = 4\nplt.rcParams['text.usetex'] = True" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": false }, "outputs": [], "source": [ "bi=2 \n###bi=22 , k=4 and k=5 is good for plotting \n# R [ bi, ki, yi, gi, :,:]\nplt.figure(),\nfor ki in range(Ks.size):\n plt.subplot(2,6,ki+1)\n plt.title(ki)\n yi=0\n #for yi in range(dy0.size):\n #plt.subplot(1,dy0.size,yi+1)\n plt.plot(timegrid[:-2], np.nanmean(Rttcont[2:4,ki, :, :2,:-2 ], axis=(0,2,-1)).T ) \n plt.plot(timegrid[:-2], np.nanmean(Rttnon[2:4,ki, :, :2,:-2 ], axis=(0,2,-1)).T ,'--',alpha=0.4 ) \n plt.ylim(0.5,1.1)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": false }, "outputs": [], "source": [ "# R [ bi, ki, yi, gi, :,:]\nbi=3\nplt.figure(),\nfor yi in range(dy0.size):\n plt.subplot(1,dy0.size,yi+1)\n ax = plt.gca()\n # color = next(ax._get_lines.prop_cycler)['color']\n # plt.plot(Ks, np.nanmean(np.nanmean(Rttcont[bi,:, yi, 0 ], axis=-2), axis=(-1)) , c=color)\n # plt.plot(Ks, np.nanmean(np.nanmean(Rttnon[bi,:, yi, 0 ], axis=-2), axis=(-1)) ,'--', c=color)\n color = next(ax._get_lines.prop_cycler)['color']\n plt.plot(Ks[0::2], np.nanmean(np.nanmean(Rttcont[2:5,0::2, yi, 1,:-2 ], axis=-2), axis=(0,-1)), c=color,marker='.' ) \n plt.plot(Ks[0::2], np.nanmean(np.nanmean(Rttnon[2:5,0::2, yi, 1,:-2 ], axis=-2), axis=(0,-1)) ,'--', c=color,marker='.' )\n color = next(ax._get_lines.prop_cycler)['color']\n plt.plot(Ks[0::2], np.nanmean(np.nanmean(Rttcont[2:5,0::2, yi, 2 ,:-2], axis=-2), axis=(0,-1)) , c=color,marker='.' ) \n plt.plot(Ks[0::2], np.nanmean(np.nanmean(Rttnon[2:5,0::2, yi, 2,:-2 ], axis=-2), axis=(0,-1)) ,'--', c=color,marker='.' )\n plt.ylim(0.5,1)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": false }, "outputs": [], "source": [ "plt.figure()\nplt.plot(Rttcont[2,3,1,2,:-2]),\nplt.plot(Rttnon[2,3,1,2,:-2],'grey')" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": false }, "outputs": [], "source": [ "plt.figure()\nplt.plot(Rttcont[2,5,1,1,:-2]),\nplt.plot(Rttnon[2,5,1,1,:-2],'grey',alpha=0.5)\n\nplt.figure()\nplt.plot(Rttcont[2,5,1,2,:-2]),\nplt.plot(Rttnon[2,5,1,2,:-2],'grey',alpha=0.5)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": false }, "outputs": [], "source": [ "bis= 2\nbie = 5\n\nplt.figure(),\nfor yi in range(dy0.size):\n plt.subplot(1,dy0.size,yi+1)\n ax = plt.gca()\n # color = next(ax._get_lines.prop_cycler)['color']\n # plt.plot(Ks, np.nanmean(np.nanmean(Rttcont[bi,:, yi, 0 ], axis=-2), axis=(-1)) , c=color)\n # plt.plot(Ks, np.nanmean(np.nanmean(Rttnon[bi,:, yi, 0 ], axis=-2), axis=(-1)) ,'--', c=color)\n color = next(ax._get_lines.prop_cycler)['color']\n plt.plot(Ks[1::2], np.nanmean(np.nanmean(Rttcont[bis:bie,1::2, yi, 1,:-2 ], axis=-2), axis=(0,-1)), c=color,marker='.' ) \n plt.plot(Ks[1::2], np.nanmean(np.nanmean(Rttnon[bis:bie,1::2, yi, 1,:-2 ], axis=-2), axis=(0,-1)) ,'--', c=color,marker='.' )\n color = next(ax._get_lines.prop_cycler)['color']\n plt.plot(Ks[1::2], np.nanmean(np.nanmean(Rttcont[bis:bie,1::2, yi, 2 ,:-2], axis=-2), axis=(0,-1)) , c=color,marker='.' ) \n plt.plot(Ks[1::2], np.nanmean(np.nanmean(Rttnon[bis:bie,1::2, yi, 2,:-2 ], axis=-2), axis=(0,-1)) ,'--', c=color,marker='.' )\n plt.ylim(0.5,1)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": false }, "outputs": [], "source": [ "bi=3\nplt.figure(),\nfor yi in range(0,4):\n plt.subplot(1,4,yi+1)\n ax = plt.gca()\n # color = next(ax._get_lines.prop_cycler)['color']\n # plt.plot(Ks[1::2], np.nanmean(np.nanmean(Rttcont[yi+0:5,1::2, :, 0,:-2 ], axis=-2), axis=(0,2,-1)), c=color,marker='.' ) \n # plt.plot(Ks[1::2], np.nanmean(np.nanmean(Rttnon[yi+0:5,1::2, :, 0,:-2 ], axis=-2), axis=(0,2,-1)) ,'--', c=color,marker='.' )\n # plt.plot(Ks, np.nanmean(np.nanmean(Rttnon[bi,:, yi, 0 ], axis=-2), axis=(-1)) ,'--', c=color)\n color = next(ax._get_lines.prop_cycler)['color']\n plt.plot(Ks[1:-1:2], np.nanmean(np.nanmean(Rttcont[yi+0:5,1::2, :, 1,:-2 ], axis=-2), axis=(0,2,-1)), c=color,marker='.' ) \n plt.plot(Ks[1:-1:2], np.nanmean(np.nanmean(Rttnon[yi+0:5,1::2, :, 1,:-2 ], axis=-2), axis=(0,2,-1)) ,'--', c=color,marker='.' )\n color = next(ax._get_lines.prop_cycler)['color']\n plt.plot(Ks[1:-1:2], np.nanmean(np.nanmean(Rttcont[yi+0:5,1::2, :, 2 ,:-2], axis=-2), axis=(0,2,-1)) , c=color,marker='.' ) \n plt.plot(Ks[1:-1:2], np.nanmean(np.nanmean(Rttnon[yi+0:5, 1::2, :, 2,:-2 ], axis=-2), axis=(0,2,-1)) ,'--', c=color,marker='.' )\n plt.ylim(0.5,1)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": false }, "outputs": [], "source": [ "#figure and plotting settings\nimport seaborn as sns\nplt.rcParams['figure.dpi'] = 300\nplt.rcParams['savefig.dpi'] = 300\nplt.rcParams['savefig.facecolor'] = (1,1,1,0)\nplt.rcParams['savefig.bbox'] = \"tight\"\nplt.rcParams['font.size'] = 10*1.2\n# plt.rcParams['font.family'] = 'sans-serif' # not available in Colab\n# plt.rcParams['font.sans-serif'] = 'Helvetica' # not available in Colab\nplt.rcParams['pdf.fonttype'] = 42\nplt.rcParams['xtick.labelsize'] = 10*1.2\nplt.rcParams['ytick.labelsize'] = 10*1.2\nplt.rcParams['axes.labelsize'] = 12*1.2\nplt.rcParams['axes.linewidth'] = 2\nplt.rcParams['axes.spines.top'] = False\nplt.rcParams['axes.spines.right'] = False\n#plt.rcParams['axes.Axes.tick_params'] = True\n#plt.rcParams['axes.solid_capstyle'] = 'round'\nplt.rc('axes',edgecolor='#4f4949')\nplt.rcParams['figure.frameon'] = False\nplt.rcParams['figure.subplot.hspace'] = 0.51\nplt.rcParams['figure.subplot.wspace'] = 0.51\nplt.rcParams['figure.subplot.left'] = 0.1\nplt.rcParams['figure.subplot.right'] = 0.9\nplt.rcParams['figure.subplot.top'] = 0.9\nplt.rcParams['figure.subplot.bottom'] = 0.1\nplt.rcParams['lines.solid_capstyle'] = 'round'\nplt.rcParams['lines.solid_joinstyle'] = 'round'\nplt.rcParams['xtick.minor.size'] = 10\n#plt.rcParams['xtick.major.width'] = 4\nplt.rcParams['text.usetex'] = True" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": false }, "outputs": [], "source": [ "pali = sns.diverging_palette(145, 300, s=80, as_cmap=True)\ncols = [pali(0.99), pali(0.95), pali(0.85), pali(0.80), pali(0.75), pali(0.70) ]\ncols2 = [pali(0.),pali(0.05),pali(0.1),pali(0.15),pali(0.2),pali(0.25)]\n\nfrecmap = plt.get_cmap( 'plasma')\n# minw = np.abs(np.min(ws))+0.3\n# maxw = np.max(ws)\n# intervalw = maxw +minw +0.4\n# print([ ( (wsi + np.pi/2)/(np.pi) ) for wsi in ws ])\n# wscols = [ frecmap( (wsi + minw)/(intervalw) ) for wsi in ws ]\n\nfig9 = plt.figure(constrained_layout=False, figsize=(6,3))\ngs1 = fig9.add_gridspec(nrows=4, ncols=8, wspace=1.5, hspace=1.2)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "ax01 = fig9.add_subplot(gs1[0:2,0:2 ]) \n\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": false }, "outputs": [], "source": [ "fig9.text(0.7, -0.05, 'time', ha='center',fontsize=16)\nfig9.text(0.45, 0.5, r'order param. $R$', va='center', rotation='vertical',fontsize=14)\n\nfig9.text(0.7, 0.95, r'coupling $J= 2.0$', ha='center',fontsize=10)\nfig9.text(0.95, 0.95, r'noise', ha='center',fontsize=10)\nfig9.text(0.95, 0.73, r'$\\sigma= 0.5$', ha='center',fontsize=10)\nfig9.text(0.95, 0.23, r'$\\sigma= 1.0$', ha='center',fontsize=10)\n# plt.ylabel(r'order param. $R$') \n# plt.xlabel('time')\n# plt.ylabel(r'order param. $R$') \n\nax02 = fig9.add_subplot(gs1[0:4,0:4 ]) \n\n\n\n\n#color = next(ax02._get_lines.prop_cycler)['color']\nplt.plot(Ks[1:-1:2], np.nanmean(np.nanmean(Rttcont[2:5,1::2, :, 1,:-2 ], axis=-2), axis=(0,2,-1)), c=cols[5],marker='^',label=r'$\\sigma=0.5$',zorder=5 ,lw=2,markersize=6,markeredgecolor='#4f4949') \n\n#color = next(ax02._get_lines.prop_cycler)['color']\nplt.plot(Ks[1:-1:2], np.nanmean(np.nanmean(Rttcont[2:5,1::2, :, 2 ,:-2], axis=-2), axis=(0,2,-1)) , c=cols[1],marker='.',label=r'$\\sigma=1.0$' ,zorder=5,lw=2,markersize=10,markeredgecolor='#4f4949') \nplt.plot(Ks[1:-1:2], np.nanmean(np.nanmean(Rttnon[2:5,1::2, :, 1,:-2 ], axis=-2), axis=(0,2,-1)) ,'--', c=cols2[5],marker='^',label=r'$\\sigma=0.5$',lw=2 ,markersize=6,markeredgecolor='#4f4949') \nplt.plot(Ks[1:-1:2], np.nanmean(np.nanmean(Rttnon[2:5, 1::2, :, 2,:-2 ], axis=-2), axis=(0,2,-1)) ,'--', c=cols2[1],marker='.' ,label=r'$\\sigma=1.0$',lw=2,markersize=10,markeredgecolor='#4f4949')\nplt.ylim(0.5,1) \n#plt.xlim(0,4.01)\n\nlegend = ax02.legend() \nlegend.get_frame().set_linewidth(1.8)\nlegend.get_frame().set_facecolor('white')\nlegend.get_frame().set_edgecolor('white')\nhandles, labels = ax02.get_legend_handles_labels()\n#handles = [ handles[-2],handles[0] , handles[-1]]\n#labels= [ labels[-2],labels[0] , labels[-1]]\nif True:\n ax02.legend(handles, labels, title=r'controlled $\\,$ uncontrolled',\n handletextpad=0.5, columnspacing=0.5,handlelength=1.3, bbox_to_anchor=[-0.0, 0.0],\n loc=3, ncol=2, frameon=True,fontsize = 10,shadow=None,framealpha =0,edgecolor ='#0a0a0a') \n\nplt.setp(plt.gca().get_legend().get_title(), fontsize='10') \nplt.ylabel(r'order param. $R$') \nplt.xlabel(r'coupling $J$')\n\nax1 = fig9.add_subplot(gs1[0:2,4:6 ] ) \n\nplt.plot(timegrid[:-2],Rttcont[2,5,1,1,:-2],c=cols[5],alpha=1),\nplt.plot(timegrid[:-2],np.mean(Rttcont[2,5,1,1,:-2],axis=1), '--',c='#4f4949')\n\nax1.set_ylim(0.0,1.01)\n#plt.yticks([0.5,1.0])\nplt.xticks([0.5,1.5])\nax1b = fig9.add_subplot(gs1[0:2,6:8 ] , sharey = ax1) \n\nplt.plot(timegrid[:-2],Rttnon[2,5,1,1,:-2],c=cols2[5],alpha=0.5)\nax1b.set_ylim(0.0,1.01)\nplt.plot(timegrid[:-2],np.mean(Rttnon[2,5,1,1,:-2],axis=1), '--',c='#4f4949')\n#plt.ylabel(r'order param. $R$') \nplt.xticks([0.5,1.5])\n#plt.yticks([0.5,1.0])\nax2 = fig9.add_subplot(gs1[ 2:4,4:6], sharex = ax1)\nax2.set_ylim(0.0,1.01)\nplt.plot(timegrid[:-2],Rttcont[2,5,1,2,:-2],c=cols[1],alpha=1),\nplt.plot(timegrid[:-2],np.mean(Rttcont[2,5,1,2,:-2],axis=1), '--',c='#4f4949')\n#plt.xlabel('time')\n#plt.ylabel(r'order param. $R$') \nplt.xticks([0.5,1.5])\n#plt.yticks([0.5,1.0])\nax2b = fig9.add_subplot(gs1[ 2:4,6:8], sharex = ax1b, sharey = ax2) \nplt.plot(timegrid[:-2],Rttnon[2,5,1,2,:-2],c=cols2[1],alpha=0.5)\nplt.plot(timegrid[:-2],np.mean(Rttnon[2,5,1,2,:-2],axis=1), '--',c='#4f4949')\nax2b.set_ylim(0.0,1.01)\nplt.xticks([0.5,1.5])\n\n\n\n#plt.savefig(\"Kuramoto_2.png\", bbox_inches='tight')\n#plt.savefig(\"Kuramoto_2.pdf\", bbox_inches='tight')" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": false }, "outputs": [], "source": [ "# R [ bi, ki, yi, gi, :,:]\n\nonset_cont = np.zeros((3,Ks.size,2,3,20)) *np.nan\nonset_uncont = np.zeros((3,Ks.size,2,3,20)) *np.nan\n\ncont_synched_durations_m = np.zeros((3,Ks.size,2,3,20)) *np.nan\ncont_synched_durations_std = np.zeros((3,Ks.size,2,3,20)) *np.nan\ncont_synched_durations_max = np.zeros((3,Ks.size,2,3,20)) *np.nan\ncont_synched_durations_sum = np.zeros((3,Ks.size,2,3,20)) *np.nan\n\nuncont_synched_durations_m = np.zeros((3,Ks.size,2,3,20)) *np.nan\nuncont_synched_durations_std = np.zeros((3,Ks.size,2,3,20)) *np.nan\nuncont_synched_durations_max = np.zeros((3,Ks.size,2,3,20)) *np.nan\nuncont_synched_durations_sum = np.zeros((3,Ks.size,2,3,20)) *np.nan\n\ncounter_cont = np.zeros((3,Ks.size,2,3))\ncounter_uncont = np.zeros((3,Ks.size,2,3))\n\nthreshold = 0.99\nmin_duri = 20\nfor bi in range(3):\n bii = bi+2 ###bi should start from 2 for the data matrices since I don't have simulation results for bi:0-1\n for ki in range(Ks.size):\n for yi in range(2):#\n for gi in range(1,3):#3 \n for repi in range(20): \n positions = np.where( Rttcont[bii, ki, yi, gi,1:-2, repi] >=threshold)[0] ###this one detects if there is at all any synchronisation\n synch_or_not = Rttcont[bii, ki, yi, gi,1:-2, repi] >=threshold ##this is boolean indicating the synchronised positions\n \n if positions.size==0: \n onset_cont[bi,ki,yi,gi,repi] = np.nan\n else:\n diffpos = np.diff(synch_or_not.astype(int) ) #difference between consecutive steps\n \n diffpos2 = np.zeros(diffpos.size+1) # extend by one step at the beginning to detect \n diffpos2[1:] = diffpos\n starts = np.argwhere(diffpos2 == 1)\n stops = np.argwhere(diffpos2 == -1) \n if starts.size == 0:\n onset_cont[bi,ki,yi,gi,repi] = np.nan\n elif stops.size == 0:\n onset_cont[bi,ki,yi,gi,repi] = timegrid[ starts[0] ]\n counter_cont[bi,ki,yi,gi] += 1\n dur = timegrid[ -1 ] - timegrid[ starts[0] ]\n cont_synched_durations_m[bi,ki,yi,gi,repi] = dur\n cont_synched_durations_std[bi,ki,yi,gi,repi] = 0\n cont_synched_durations_max[bi,ki,yi,gi,repi] = dur\n cont_synched_durations_sum[bi,ki,yi,gi,repi] = np.sum( synch_or_not )/ (timegrid.size-1 - starts[0] )\n else:\n durations = stops[:,0] - starts[:stops.size, 0] \n synchned_more_than_50 = np.where( durations>=min_duri )[0]\n if synchned_more_than_50.size == 0:\n onset_cont[bi,ki,yi,gi,repi] = np.nan\n else: \n onset_cont[bi,ki,yi,gi,repi] = timegrid[ starts[synchned_more_than_50[0]] ]\n counter_cont[bi,ki,yi,gi] += 1 \n cont_synched_durations_m[bi,ki,yi,gi,repi] = np.mean( durations )\n cont_synched_durations_std[bi,ki,yi,gi,repi] = np.std( durations )\n cont_synched_durations_max[bi,ki,yi,gi,repi] = np.max( durations )\n cont_synched_durations_sum[bi,ki,yi,gi,repi] = np.sum( synch_or_not[ starts[synchned_more_than_50[0]][0] : ] )/(timegrid.size-1 -starts[synchned_more_than_50[0]] ) ##timesteps in synchronied\n\n \n \n positions = np.where( Rttnon[bii, ki, yi, gi, 1:-2,repi] >=threshold)[0]\n synch_or_not = Rttnon[bii, ki, yi, gi,1:-2, repi] >=threshold \n if positions.size ==0: \n onset_uncont[bi,ki,yi,gi,repi] = np.nan\n else:\n diffpos = np.diff(synch_or_not.astype(int) ) #difference between consecutive steps\n \n diffpos2 = np.zeros(diffpos.size+1) # extend by one step at the beginning to detect \n diffpos2[1:] = diffpos\n starts = np.argwhere(diffpos2 == 1)\n stops = np.argwhere(diffpos2 == -1)\n #start_stop =[starts, stops - starts]\n if starts.size == 0:\n onset_uncont[bi,ki,yi,gi,repi] = np.nan\n elif stops.size == 0:\n onset_uncont[bi,ki,yi,gi,repi] = timegrid[ starts[0] ]\n counter_uncont[bi,ki,yi,gi] += 1\n dur = timegrid[ -1 ] - timegrid[ starts[0] ]\n uncont_synched_durations_m[bi,ki,yi,gi,repi] = dur\n uncont_synched_durations_std[bi,ki,yi,gi,repi] = 0\n uncont_synched_durations_max[bi,ki,yi,gi,repi] = dur\n uncont_synched_durations_sum[bi,ki,yi,gi,repi] = np.sum( synch_or_not )/ (timegrid.size-1 - starts[0] )\n else:\n durations = stops[:,0] - starts[:stops.size, 0] \n synchned_more_than_50 = np.where( durations>=min_duri )[0]\n \n if synchned_more_than_50.size == 0:\n onset_uncont[bi,ki,yi,gi,repi] = np.nan\n else: \n onset_uncont[bi,ki,yi,gi,repi] = timegrid[starts[synchned_more_than_50[0]] ]\n counter_uncont[bi,ki,yi,gi] += 1\n uncont_synched_durations_m[bi,ki,yi,gi,repi] = np.mean( durations )\n uncont_synched_durations_std[bi,ki,yi,gi,repi] = np.std( durations )\n uncont_synched_durations_max[bi,ki,yi,gi,repi] = np.max( durations )\n uncont_synched_durations_sum[bi,ki,yi,gi,repi] = np.sum( synch_or_not[ starts[synchned_more_than_50[0]][0] : ] ) /(timegrid.size-1 -starts[synchned_more_than_50[0]] )" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": false }, "outputs": [], "source": [ "count_perc = np.sum(counter_uncont[0:3,1::2,:1, :],axis=0)/60 #counts the percentage of synchronised uncontrolled network for the annotations \n \nplt.figure(figsize=(6.5,2)) \nyi = 0\ngi = 1\nplt.subplot(1,2,1)\nax1 = plt.gca()\n#er1 = plt.errorbar(Ks[1:-1:2], np.nanmean(onset_cont[0:3,1::2,yi, 1], axis=(0,-1)), yerr=np.nanstd(onset_cont[0:3,1::2,yi, 1], axis=(0,-1)), marker=\"o\", linestyle=\"-\",uplims=True,lolims=True,)\n#er2 = plt.errorbar(Ks[1:-1:2], np.nanmean(onset_uncont[0:3,1::2,yi, 1], axis=(0,-1)), yerr=np.nanstd(onset_uncont[0:3,1::2,yi, 1], axis=(0,-1)), marker=\"o\", linestyle=\"-\",uplims=True,lolims=True)\nplt.plot(Ks[1:-1:2], np.nanmean(onset_cont[0:3,1::2,:1, 1], axis=(0,2,-1)) , c=cols[5],marker='^',label=r'$\\sigma=0.5$',zorder=5 ,lw=2.8,markersize=6,markeredgecolor='#4f4949') \nplt.plot(Ks[1:-1:2], np.nanmean(onset_uncont[0:3,1::2,:1, 1], axis=(0,2,-1)),'--', c=cols2[5],marker='^',label=r'$\\sigma=0.5$' ,zorder=5,lw=2.8,markersize=6,markeredgecolor='#4f4949') \n\nplt.text(0.29, 0.49,\"%.2f\"%count_perc[0, 0 ,1], fontsize =8 , color = \"grey\")\n\nplt.text(0.97, 0.57,\"%.2f\"%count_perc[1, 0 ,1], fontsize =8 , color = \"grey\")\n\nplt.text(2.05, 0.44,\"%.2f\"%count_perc[2, 0 ,1], fontsize =8 , color = \"grey\")\n\n\nplt.text(0.29, 0.28,\"%.2f\"%count_perc[0, 0 ,2], fontsize =8 , color = \"grey\")\n\nplt.text(1.1, 0.37,\"%.2f\"%count_perc[1, 0 ,2], fontsize =8 , color = \"grey\")\n\nplt.text(1.89, 0.28,\"%.2f\"%count_perc[2, 0 ,2], fontsize =8 , color = \"grey\")\n\n\nplt.plot(Ks[1:-1:2], np.nanmean(onset_cont[0:3,1::2,:1, 2], axis=(0,2,-1)) , c=cols[1],marker='.',label=r'$\\sigma=1.0$',lw=2.8 ,markersize=12,markeredgecolor='#4f4949') \nplt.plot(Ks[1:-1:2], np.nanmean(onset_uncont[0:3,1::2,:1, 2], axis=(0,2,-1)),'--', c=cols2[0],marker='.' ,label=r'$\\sigma=1.0$',lw=2.8,markersize=12,markeredgecolor='#4f4949') \n\n#plt.yscale('log') \nplt.ylabel(r'onset of synchrony $t^{syn}$') \nplt.xlabel(r'coupling $J$') \n\nax6 = plt.gca()\n\nax6.tick_params(axis='both',which='both',direction='in', length=3, width=1,colors='#4f4949',zorder=3)\nax6.tick_params(bottom=True, top=True, left=True, right=True)\nax6.spines['top'].set_visible(True)\nax6.spines['right'].set_visible(True)\nax6.minorticks_on()\nax6.tick_params(axis='both',which='major',direction='in', length=3.5, width=1,colors='#4f4949',zorder=3)\nax6.tick_params(axis='both',which='minor',direction='in', length=2.5, width=0.5,colors='#4f4949',zorder=3)\nax6.tick_params(bottom=True, top=True, left=True, right=True)\nax6.tick_params(axis='both', which='minor', bottom=True, top=True, left=True, right=True)\n\n\nplt.subplot(1,2,2)\n\nplt.plot(Ks[1:-1:2], np.nanmean(cont_synched_durations_sum[0:3,1::2,:1, 1], axis=(0,2,-1)) , c=cols[5],marker='^',label=r'$\\sigma=0.5$',zorder=2 ,lw=2.8,markersize=6,markeredgecolor='#4f4949') \nplt.plot(Ks[1:-1:2], np.nanmean(uncont_synched_durations_sum[0:3,1::2,:1, 1], axis=(0,2,-1)),'--', c=cols2[5],marker='^',label=r'$\\sigma=0.5$' ,zorder=2,lw=2.8,markersize=6,markeredgecolor='#4f4949') \n\n\nplt.plot(Ks[1:-1:2], np.nanmean(cont_synched_durations_sum[0:3,1::2,:1, 2], axis=(0,2,-1)) , c=cols[1],marker='.',label=r'$\\sigma=1.0$',lw=2.8 ,markersize=12,markeredgecolor='#4f4949') \nplt.plot(Ks[1:-1:2], np.nanmean(uncont_synched_durations_sum[0:3,1::2,:1, 2], axis=(0,2,-1)),'--', c=cols2[0],marker='.' ,label=r'$\\sigma=1.0$',lw=2.8,markersize=12,markeredgecolor='#4f4949') \n\nplt.ylabel('$\\%$ time synchronised\\nafter $t^{syn}$', multialignment='center',labelpad=-0.1) # \"Mat\\nTTp\\n123\"\nplt.xlabel(r'coupling $J$') \nplt.yticks([0, 0.5,1])\nax7 = plt.gca()\nax7.tick_params(axis='both',which='both',direction='in', length=3, width=1,colors='#4f4949',zorder=3)\nax7.tick_params(bottom=True, top=True, left=True, right=True)\nax7.spines['top'].set_visible(True)\nax7.spines['right'].set_visible(True)\nax7.minorticks_on()\nax7.tick_params(axis='both',which='major',direction='in', length=3.5, width=1,colors='#4f4949',zorder=3)\nax7.tick_params(axis='both',which='minor',direction='in', length=2.5, width=0.5,colors='#4f4949',zorder=3)\nax7.tick_params(bottom=True, top=True, left=True, right=True)\nax7.tick_params(axis='both', which='minor', bottom=True, top=True, left=True, right=True)\n\nlegend = ax7.legend() \nlegend.get_frame().set_linewidth(1.8)\nlegend.get_frame().set_facecolor('white')\nlegend.get_frame().set_edgecolor('white')\nhandles, labels = ax7.get_legend_handles_labels()\nhandles = [ handles[0],handles[2] , handles[1], handles[3]]\nlabels= [ labels[0],labels[2] , labels[1], labels[3]]\nif True:\n ax7.legend(handles, labels, title=r'$\\,$controlled $\\,\\,$ uncontrolled',\n handletextpad=0.5, columnspacing=1,handlelength=0.45, bbox_to_anchor=[-0.65, 0.99],\n loc=3, ncol=2, frameon=True,fontsize = 'small',shadow=None,framealpha =0,edgecolor ='#0a0a0a') \nplt.setp(plt.gca().get_legend().get_title(), fontsize='10') \nplt.subplots_adjust(wspace=0.325)#, hspace=0)\nplt.savefig(\"Kuramoto_2_onset.png\", bbox_inches='tight')\nplt.savefig(\"Kuramoto_2_onset.pdf\", bbox_inches='tight')\n\"\"\"\nplt.subplot(1,2,2)#,sharey=ax1)\n\n# plt.plot(Ks[1:-1:2], np.nanmean(onset_cont[:,1::2,yi, 1], axis=(0,-1)) , c=cols[5],marker='^',label=r'$\\sigma=0.5$',zorder=5 ,lw=2.8,markersize=6,markeredgecolor='#4f4949') \n# plt.plot(Ks[1:-1:2], np.nanmean(onset_uncont[:,1::2,yi, 1], axis=(0,-1)),'--', c=cols2[5],marker='.',label=r'$\\sigma=1.0$' ,zorder=5,lw=2.8,markersize=10,markeredgecolor='#4f4949') \nplt.plot(Ks[1:-1:2], np.nanmean(onset_cont[0:3,1::2,0, 2], axis=(0,-1)) , c=cols[1],marker='^',label=r'$\\sigma=0.5$',lw=2.8 ,markersize=6,markeredgecolor='#4f4949') \nplt.plot(Ks[1:-1:2], np.nanmean(onset_uncont[0:3,1::2,0, 2], axis=(0,-1)),'--', c=cols2[2],marker='.' ,label=r'$\\sigma=1.0$',lw=2.8,markersize=12,markeredgecolor='#4f4949') \n\n\nplt.plot(Ks[1:-1:2], np.nanmean(onset_cont[0:3,1::2,1, 2], axis=(0,-1)) , c=cols[1],marker='d',label=r'$\\sigma=0.5$',lw=2.8 ,markersize=6,markeredgecolor='#4f4949') \nplt.plot(Ks[1:-1:2], np.nanmean(onset_uncont[0:3,1::2,1, 2], axis=(0,-1)),'--', c=cols2[2],marker='p' ,label=r'$\\sigma=1.0$',lw=2.8,markersize=7,markeredgecolor='#4f4949') \n\n#plt.yscale('log') \nplt.ylabel(r'onset of synchrony $t^{syn}$') \nplt.xlabel(r'coupling $J$') \n\n\"\"\"" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": false }, "outputs": [], "source": [ "ki=1\nyi=0\ngi=1\nbi=1\nfor repi in range(1):\n positions = np.where( Rttnon[bii, ki, yi, gi, 1:-2,repi] >=threshold)[0]\n print(positions)\n if positions.size ==0: \n onset_uncont[bi,ki,yi,gi,repi] = np.nan\n else:\n diffpos = np.diff(positions)\n positions_consecutive = np.where(diffpos < 5)[0]\n if positions_consecutive.size ==0:\n onset_uncont[bi,ki,yi,gi,repi] = np.nan\n else:\n onset_uncont[bi,ki,yi,gi,repi] = timegrid[positions[positions_consecutive[0]] ]\n print(onset_uncont[bi,ki,yi,gi,repi])\n # print()\n # print()" ] } ], "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": "3.9.15" } }, "nbformat": 4, "nbformat_minor": 0 }