diff --git a/CH01/CH01_SEC04_1_Linear.ipynb b/CH01/CH01_SEC04_1_Linear.ipynb
index 3b87a38..26e01c5 100644
--- a/CH01/CH01_SEC04_1_Linear.ipynb
+++ b/CH01/CH01_SEC04_1_Linear.ipynb
@@ -29,13 +29,13 @@
     "a = a.reshape(-1, 1)\n",
     "b = x*a + np.random.randn(*a.shape) # Add noise\n",
     "\n",
-    "plt.plot(a, x*a, Color='k', LineWidth=2, label='True line') # True relationship\n",
-    "plt.plot(a, b, 'x', Color='r', MarkerSize = 10, label='Noisy data') # Noisy measurements\n",
+    "plt.plot(a, x*a, color='k', linewidth=2, label='True line') # True relationship\n",
+    "plt.plot(a, b, 'x', color='r', MarkerSize = 10, label='Noisy data') # Noisy measurements\n",
     "\n",
     "U, S, VT = np.linalg.svd(a,full_matrices=False)\n",
     "xtilde = VT.T @ np.linalg.inv(np.diag(S)) @ U.T @ b # Least-square fit\n",
     "\n",
-    "plt.plot(a,xtilde * a,'--',Color='b',LineWidth=4, label='Regression line')\n",
+    "plt.plot(a,xtilde * a,'--',color='b',linewidth=4, label='Regression line')\n",
     "\n",
     "plt.xlabel('a')\n",
     "plt.ylabel('b')\n",
diff --git a/CH01/CH01_SEC04_2_Cement.ipynb b/CH01/CH01_SEC04_2_Cement.ipynb
index 80378d2..41c43f4 100644
--- a/CH01/CH01_SEC04_2_Cement.ipynb
+++ b/CH01/CH01_SEC04_2_Cement.ipynb
@@ -33,8 +33,8 @@
     "U, S, VT = np.linalg.svd(A,full_matrices=0)\n",
     "x = VT.T @ np.linalg.inv(np.diag(S)) @ U.T @ b\n",
     "\n",
-    "plt.plot(b, Color='k', LineWidth=2, label='Heat Data') # True relationship\n",
-    "plt.plot(A@x, '-o', Color='r', LineWidth=1.5, MarkerSize=6, label='Regression')\n",
+    "plt.plot(b, color='k', linewidth=2, label='Heat Data') # True relationship\n",
+    "plt.plot(A@x, '-o', color='r', linewidth=1.5, markersize=6, label='Regression')\n",
     "plt.legend()\n",
     "plt.show()"
    ]
diff --git a/CH01/CH01_SEC04_3_Housing.ipynb b/CH01/CH01_SEC04_3_Housing.ipynb
index dd1efd1..29a2e57 100644
--- a/CH01/CH01_SEC04_3_Housing.ipynb
+++ b/CH01/CH01_SEC04_3_Housing.ipynb
@@ -42,16 +42,16 @@
     "fig = plt.figure()\n",
     "ax1 = fig.add_subplot(121)\n",
     "\n",
-    "plt.plot(b, Color='k', LineWidth=2, label='Housing Value') # True relationship\n",
-    "plt.plot(A@x, '-o', Color='r', LineWidth=1.5, MarkerSize=6, label='Regression')\n",
+    "plt.plot(b, color='k', linewidth=2, label='Housing Value') # True relationship\n",
+    "plt.plot(A@x, '-o', color='r', linewidth=1.5, markersize=6, label='Regression')\n",
     "plt.xlabel('Neighborhood')\n",
     "plt.ylabel('Median Home Value [$1k]')\n",
     "plt.legend()\n",
     "\n",
     "ax2 = fig.add_subplot(122)\n",
     "sort_ind = np.argsort(H[:,-1])\n",
-    "plt.plot(b[sort_ind], Color='k', LineWidth=2, label='Housing Value') # True relationship\n",
-    "plt.plot(A[sort_ind,:]@x, '-o', Color='r', LineWidth=1.5, MarkerSize=6, label='Regression')\n",
+    "plt.plot(b[sort_ind], color='k', linewidth=2, label='Housing Value') # True relationship\n",
+    "plt.plot(A[sort_ind,:]@x, '-o', color='r', linewidth=1.5, markersize=6, label='Regression')\n",
     "plt.xlabel('Neighborhood')\n",
     "plt.legend()\n",
     "\n",
diff --git a/CH01/CH01_SEC05_1_PCAGaussian.ipynb b/CH01/CH01_SEC05_1_PCAGaussian.ipynb
index a0d0032..74a09d4 100644
--- a/CH01/CH01_SEC05_1_PCAGaussian.ipynb
+++ b/CH01/CH01_SEC05_1_PCAGaussian.ipynb
@@ -2,12 +2,12 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 52,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 1152x576 with 2 Axes>"
       ]
@@ -36,7 +36,7 @@
     "\n",
     "fig = plt.figure()\n",
     "ax1 = fig.add_subplot(121)\n",
-    "ax1.plot(X[0,:],X[1,:], '.', Color='k')\n",
+    "ax1.plot(X[0,:],X[1,:], '.', color='k')\n",
     "ax1.grid()\n",
     "plt.xlim((-6, 8))\n",
     "plt.ylim((-6,8))\n",
@@ -50,7 +50,7 @@
     "U, S, VT = np.linalg.svd(B/np.sqrt(nPoints),full_matrices=0)\n",
     "\n",
     "ax2 = fig.add_subplot(122)\n",
-    "ax2.plot(X[0,:],X[1,:], '.', Color='k')   # Plot data to overlay PCA\n",
+    "ax2.plot(X[0,:],X[1,:], '.', color='k')   # Plot data to overlay PCA\n",
     "ax2.grid()\n",
     "plt.xlim((-6, 8))\n",
     "plt.ylim((-6,8))\n",
@@ -60,15 +60,15 @@
     "# 1-std confidence interval\n",
     "Xstd = U @ np.diag(S) @ np.array([np.cos(theta),np.sin(theta)])\n",
     "\n",
-    "ax2.plot(Xavg[0] + Xstd[0,:], Xavg[1] + Xstd[1,:],'-',color='r',LineWidth=3)\n",
-    "ax2.plot(Xavg[0] + 2*Xstd[0,:], Xavg[1] + 2*Xstd[1,:],'-',color='r',LineWidth=3)\n",
-    "ax2.plot(Xavg[0] + 3*Xstd[0,:], Xavg[1] + 3*Xstd[1,:],'-',color='r',LineWidth=3)\n",
+    "ax2.plot(Xavg[0] + Xstd[0,:], Xavg[1] + Xstd[1,:],'-',color='r',linewidth=3)\n",
+    "ax2.plot(Xavg[0] + 2*Xstd[0,:], Xavg[1] + 2*Xstd[1,:],'-',color='r',linewidth=3)\n",
+    "ax2.plot(Xavg[0] + 3*Xstd[0,:], Xavg[1] + 3*Xstd[1,:],'-',color='r',linewidth=3)\n",
     "\n",
     "# Plot principal components U[:,0]S[0] and U[:,1]S[1]\n",
     "ax2.plot(np.array([Xavg[0], Xavg[0]+U[0,0]*S[0]]),\n",
-    "         np.array([Xavg[1], Xavg[1]+U[1,0]*S[0]]),'-',color='cyan',LineWidth=5)\n",
+    "         np.array([Xavg[1], Xavg[1]+U[1,0]*S[0]]),'-',color='cyan',linewidth=5)\n",
     "ax2.plot(np.array([Xavg[0], Xavg[0]+U[0,1]*S[1]]),\n",
-    "         np.array([Xavg[1], Xavg[1]+U[1,1]*S[1]]),'-',color='cyan',LineWidth=5)\n",
+    "         np.array([Xavg[1], Xavg[1]+U[1,1]*S[1]]),'-',color='cyan',linewidth=5)\n",
     "\n",
     "plt.show()"
    ]
@@ -97,7 +97,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.3"
+   "version": "3.8.5"
   }
  },
  "nbformat": 4,
diff --git a/CH01/CH01_SEC05_2_OvarianCancer.ipynb b/CH01/CH01_SEC05_2_OvarianCancer.ipynb
index e740d4c..f184486 100644
--- a/CH01/CH01_SEC05_2_OvarianCancer.ipynb
+++ b/CH01/CH01_SEC05_2_OvarianCancer.ipynb
@@ -96,7 +96,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.3"
+   "version": "3.8.5"
   }
  },
  "nbformat": 4,
diff --git a/CH01/CH01_SEC06_2_3_4.ipynb b/CH01/CH01_SEC06_2_3_4.ipynb
index da94a2b..4f33b7d 100644
--- a/CH01/CH01_SEC06_2_3_4.ipynb
+++ b/CH01/CH01_SEC06_2_3_4.ipynb
@@ -214,8 +214,8 @@
     "PCACoordsP1 = U[:,PCAmodes-np.ones_like(PCAmodes)].T @ P1\n",
     "PCACoordsP2 = U[:,PCAmodes-np.ones_like(PCAmodes)].T @ P2\n",
     "\n",
-    "plt.plot(PCACoordsP1[0,:],PCACoordsP1[1,:],'d',Color='k',label='Person 2')\n",
-    "plt.plot(PCACoordsP2[0,:],PCACoordsP2[1,:],'^',Color='r',label='Person 7')\n",
+    "plt.plot(PCACoordsP1[0,:],PCACoordsP1[1,:],'d',color='k',label='Person 2')\n",
+    "plt.plot(PCACoordsP2[0,:],PCACoordsP2[1,:],'^',color='r',label='Person 7')\n",
     "\n",
     "plt.legend()\n",
     "plt.show()"
diff --git a/CH01/CH01_SEC07_1.ipynb b/CH01/CH01_SEC07_1.ipynb
index cef8288..6487171 100644
--- a/CH01/CH01_SEC07_1.ipynb
+++ b/CH01/CH01_SEC07_1.ipynb
@@ -173,10 +173,10 @@
     "\n",
     "fig1,ax1 = plt.subplots(1)\n",
     "\n",
-    "ax1.semilogy(S,'-o', color='k', LineWidth=2)\n",
-    "ax1.semilogy(np.diag(S[:(r+1)]),'o', color='r', LineWidth=2)\n",
-    "ax1.plot(np.array([-20, N+20]),np.array([cutoff, cutoff]),'--', color='r', LineWidth=2)\n",
-    "rect = patches.Rectangle((-5,20),100,200,LineWidth=2,LineStyle='--',FaceColor='none',EdgeColor='k')\n",
+    "ax1.semilogy(S,'-o', color='k', linewidth=2)\n",
+    "ax1.semilogy(np.diag(S[:(r+1)]),'o', color='r', linewidth=2)\n",
+    "ax1.plot(np.array([-20, N+20]),np.array([cutoff, cutoff]),'--', color='r', linewidth=2)\n",
+    "rect = patches.Rectangle((-5,20),100,200,linewidth=2,linestyle='--',facecolor='none',edgecolor='k')\n",
     "ax1.add_patch(rect)\n",
     "plt.xlim((-10,610))\n",
     "plt.ylim((0.003,300))\n",
@@ -185,22 +185,22 @@
     "\n",
     "fig2,ax2 = plt.subplots(1)\n",
     "\n",
-    "ax2.semilogy(S,'-o', color='k', LineWidth=2)\n",
-    "ax2.semilogy(np.diag(S[:(r+1)]),'o', color='r', LineWidth=2)\n",
-    "ax2.plot(np.array([-20, N+20]),np.array([cutoff, cutoff]),'--', color='r', LineWidth=2)\n",
+    "ax2.semilogy(S,'-o', color='k', linewidth=2)\n",
+    "ax2.semilogy(np.diag(S[:(r+1)]),'o', color='r', linewidth=2)\n",
+    "ax2.plot(np.array([-20, N+20]),np.array([cutoff, cutoff]),'--', color='r', linewidth=2)\n",
     "plt.xlim((-5,100))\n",
     "plt.ylim((20,200))\n",
     "ax2.grid()\n",
     "plt.show()\n",
     "\n",
     "fig3,ax3 = plt.subplots(1)\n",
-    "ax3.plot(cdS,'-o',color='k',LineWidth=2)\n",
-    "ax3.plot(cdS[:(r90+1)],'o',color='b',LineWidth=2)\n",
-    "ax3.plot(cdS[:(r+1)],'o',color='r',LineWidth=2)\n",
+    "ax3.plot(cdS,'-o',color='k',linewidth=2)\n",
+    "ax3.plot(cdS[:(r90+1)],'o',color='b',linewidth=2)\n",
+    "ax3.plot(cdS[:(r+1)],'o',color='r',linewidth=2)\n",
     "plt.xticks(np.array([0, 300, r90, 600]))\n",
     "plt.yticks(np.array([0, 0.5, 0.9, 1]))\n",
     "plt.xlim((-10,610))\n",
-    "ax3.plot(np.array([r90, r90, -10]),np.array([0, 0.9, 0.9]),'--',color='b',LineWidth=2)\n",
+    "ax3.plot(np.array([r90, r90, -10]),np.array([0, 0.9, 0.9]),'--',color='b',linewidth=2)\n",
     "\n",
     "ax3.grid()\n",
     "plt.show()\n",
diff --git a/CH01/CH01_SEC08_RSVD.ipynb b/CH01/CH01_SEC08_RSVD.ipynb
index 83c401e..d5b4aa2 100644
--- a/CH01/CH01_SEC08_RSVD.ipynb
+++ b/CH01/CH01_SEC08_RSVD.ipynb
@@ -136,14 +136,14 @@
     "                 [1,0,0],\n",
     "                 [2/3,0,0]])\n",
     "\n",
-    "plt.plot(S,'o-',color='k',LineWidth=2,label='SVD')\n",
+    "plt.plot(S,'o-',color='k',linewidth=2,label='SVD')\n",
     "\n",
     "Y = X\n",
     "for q in range(1,6):\n",
     "    Y = X.T @ Y\n",
     "    Y = X @ Y\n",
     "    Uq, Sq, VTq = np.linalg.svd(Y,full_matrices=0)\n",
-    "    plt.plot(Sq,'-o',color=tuple(color_list[2*q+1]),LineWidth=2,label='rSVD, q = '+str(q))\n",
+    "    plt.plot(Sq,'-o',color=tuple(color_list[2*q+1]),linewidth=2,label='rSVD, q = '+str(q))\n",
     "\n",
     "plt.legend()\n",
     "plt.show()"
diff --git a/CH01/CH01_SEC09_Tensor.ipynb b/CH01/CH01_SEC09_Tensor.ipynb
index 90dbcf8..eafec79 100644
--- a/CH01/CH01_SEC09_Tensor.ipynb
+++ b/CH01/CH01_SEC09_Tensor.ipynb
@@ -77464,9 +77464,9 @@
     "A1, A2, A3 = parafac(A,2)\n",
     "\n",
     "fig, axs = plt.subplots(3,1)\n",
-    "axs[0].plot(y,A1,LineWidth=2)\n",
-    "axs[1].plot(x,A2,LineWidth=2)\n",
-    "axs[2].plot(t,A3,LineWidth=2)\n",
+    "axs[0].plot(y,A1,linewidth=2)\n",
+    "axs[1].plot(x,A2,linewidth=2)\n",
+    "axs[2].plot(t,A3,linewidth=2)\n",
     "plt.show()"
    ]
   },
@@ -77501,7 +77501,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.3"
+   "version": "3.8.5"
   }
  },
  "nbformat": 4,
diff --git a/CH02/CH02_SEC01_0_InnerProduct.ipynb b/CH02/CH02_SEC01_0_InnerProduct.ipynb
index 373e8b7..c24da72 100644
--- a/CH02/CH02_SEC01_0_InnerProduct.ipynb
+++ b/CH02/CH02_SEC01_0_InnerProduct.ipynb
@@ -40,10 +40,10 @@
     "ff = f_interp(xf)  \n",
     "gf = g_interp(xf)\n",
     "\n",
-    "plt.plot(xf[10:-10],ff[10:-10],color='k',LineWidth=2)\n",
+    "plt.plot(xf[10:-10],ff[10:-10],color='k',linewidth=2)\n",
     "plt.plot(x[1:-2],f[1:-2],'o',color='b')\n",
     "\n",
-    "plt.plot(xf[10:-10],gf[10:-10],color='k',LineWidth=2)\n",
+    "plt.plot(xf[10:-10],gf[10:-10],color='k',linewidth=2)\n",
     "plt.plot(x[1:-2],g[1:-2],'o',color='r')\n",
     "\n",
     "plt.show()\n"
diff --git a/CH02/CH02_SEC01_1_FourierSines.ipynb b/CH02/CH02_SEC01_1_FourierSines.ipynb
index 7c5e037..532be24 100644
--- a/CH02/CH02_SEC01_1_FourierSines.ipynb
+++ b/CH02/CH02_SEC01_1_FourierSines.ipynb
@@ -39,7 +39,7 @@
     "f[2*nquart:3*nquart] = np.ones(nquart) - (4/n)*np.arange(0,nquart)\n",
     "\n",
     "fig, ax = plt.subplots()\n",
-    "ax.plot(x,f,'-',color='k',LineWidth=2)\n",
+    "ax.plot(x,f,'-',color='k',linewidth=2)\n",
     "\n",
     "# Compute Fourier series\n",
     "name = \"Accent\"\n",
@@ -99,13 +99,13 @@
     "r = np.max(np.where(ERR > thresh))\n",
     "\n",
     "fig, axs = plt.subplots(2,1)\n",
-    "axs[0].semilogy(np.arange(kmax),A,color='k',LineWidth=2)\n",
-    "axs[0].semilogy(r,A[r],'o',color='b',MarkerSize=10)\n",
+    "axs[0].semilogy(np.arange(kmax),A,color='k',linewidth=2)\n",
+    "axs[0].semilogy(r,A[r],'o',color='b',markersize=10)\n",
     "plt.sca(axs[0])\n",
     "plt.title('Fourier Coefficients')\n",
     "\n",
-    "axs[1].semilogy(np.arange(kmax),ERR,color='k',LineWidth=2)\n",
-    "axs[1].semilogy(r,ERR[r],'o',color='b',MarkerSize=10)\n",
+    "axs[1].semilogy(np.arange(kmax),ERR,color='k',linewidth=2)\n",
+    "axs[1].semilogy(r,ERR[r],'o',color='b',markersize=10)\n",
     "plt.sca(axs[1])\n",
     "plt.title('Error')\n",
     "\n",
diff --git a/CH02/CH02_SEC01_2_Gibbs.ipynb b/CH02/CH02_SEC01_2_Gibbs.ipynb
index 1377024..76264fc 100644
--- a/CH02/CH02_SEC01_2_Gibbs.ipynb
+++ b/CH02/CH02_SEC01_2_Gibbs.ipynb
@@ -41,8 +41,8 @@
     "    Bk = np.sum(f * np.sin(2*np.pi*k*x/L)) * dx * 2 / L\n",
     "    fFS = fFS + Ak*np.cos(2*k*np.pi*x/L) + Bk*np.sin(2*k*np.pi*x/L)\n",
     "    \n",
-    "plt.plot(x,f,color='k',LineWidth=2)\n",
-    "plt.plot(x,fFS,'-',color='r',LineWidth=1.5)\n",
+    "plt.plot(x,f,color='k',linewidth=2)\n",
+    "plt.plot(x,fFS,'-',color='r',linewidth=1.5)\n",
     "plt.show()"
    ]
   },
diff --git a/CH02/CH02_SEC01_2_Gibbs_Movie.ipynb b/CH02/CH02_SEC01_2_Gibbs_Movie.ipynb
index cc66d2d..b03cd8f 100644
--- a/CH02/CH02_SEC01_2_Gibbs_Movie.ipynb
+++ b/CH02/CH02_SEC01_2_Gibbs_Movie.ipynb
@@ -41074,8 +41074,8 @@
     "fFS = A0/2 * np.ones_like(f)\n",
     "\n",
     "fig,ax = plt.subplots()\n",
-    "plt.plot(x,f,color='k',LineWidth=2)\n",
-    "fFS_plot, = plt.plot([],[],color='r',LineWidth=2)\n",
+    "plt.plot(x,f,color='k',linewidth=2)\n",
+    "fFS_plot, = plt.plot([],[],color='r',linewidth=2)\n",
     "\n",
     "all_fFS = np.zeros((len(fFS),101))\n",
     "all_fFS[:,0] = fFS\n",
diff --git a/CH02/CH02_SEC02_2_Denoise.ipynb b/CH02/CH02_SEC02_2_Denoise.ipynb
index 99ddbe7..4e5bcd2 100644
--- a/CH02/CH02_SEC02_2_Denoise.ipynb
+++ b/CH02/CH02_SEC02_2_Denoise.ipynb
@@ -70,20 +70,20 @@
     "fig,axs = plt.subplots(3,1)\n",
     "\n",
     "plt.sca(axs[0])\n",
-    "plt.plot(t,f,color='r',LineWidth=1.5,label='Noisy')\n",
-    "plt.plot(t,f_clean,color='k',LineWidth=2,label='Clean')\n",
+    "plt.plot(t,f,color='r',linewidth=1.5,label='Noisy')\n",
+    "plt.plot(t,f_clean,color='k',linewidth=2,label='Clean')\n",
     "plt.xlim(t[0],t[-1])\n",
     "plt.legend()\n",
     "\n",
     "plt.sca(axs[1])\n",
-    "plt.plot(t,f_clean,color='k',LineWidth=1.5,label='Clean')\n",
-    "plt.plot(t,ffilt,color='b',LineWidth=2,label='Filtered')\n",
+    "plt.plot(t,f_clean,color='k',linewidth=1.5,label='Clean')\n",
+    "plt.plot(t,ffilt,color='b',linewidth=2,label='Filtered')\n",
     "plt.xlim(t[0],t[-1])\n",
     "plt.legend()\n",
     "\n",
     "plt.sca(axs[2])\n",
-    "plt.plot(freq[L],PSD[L],color='r',LineWidth=2,label='Noisy')\n",
-    "plt.plot(freq[L],PSDclean[L],color='b',LineWidth=1.5,label='Filtered')\n",
+    "plt.plot(freq[L],PSD[L],color='r',linewidth=2,label='Noisy')\n",
+    "plt.plot(freq[L],PSDclean[L],color='b',linewidth=1.5,label='Filtered')\n",
     "plt.xlim(freq[L[0]],freq[L[-1]])\n",
     "plt.legend()\n",
     "\n",
diff --git a/CH02/CH02_SEC02_3_SpectralDerivative.ipynb b/CH02/CH02_SEC02_3_SpectralDerivative.ipynb
index 9f8f157..079d22a 100644
--- a/CH02/CH02_SEC02_3_SpectralDerivative.ipynb
+++ b/CH02/CH02_SEC02_3_SpectralDerivative.ipynb
@@ -68,9 +68,9 @@
    ],
    "source": [
     "## Plots\n",
-    "plt.plot(x,df.real,color='k',LineWidth=2,label='True Derivative')\n",
-    "plt.plot(x,dfFD.real,'--',color='b',LineWidth=1.5,label='Finite Diff.')\n",
-    "plt.plot(x,dfFFT.real,'--',color='r',LineWidth=1.5,label='FFT Derivative')\n",
+    "plt.plot(x,df.real,color='k',linewidth=2,label='True Derivative')\n",
+    "plt.plot(x,dfFD.real,'--',color='b',linewidth=1.5,label='Finite Diff.')\n",
+    "plt.plot(x,dfFFT.real,'--',color='r',linewidth=1.5,label='FFT Derivative')\n",
     "plt.legend()\n",
     "plt.show()"
    ]
diff --git a/CH02/CH02_SEC05_HAAR.ipynb b/CH02/CH02_SEC05_HAAR.ipynb
index 78992e7..9032074 100644
--- a/CH02/CH02_SEC05_HAAR.ipynb
+++ b/CH02/CH02_SEC05_HAAR.ipynb
@@ -53,13 +53,13 @@
     "f22 = np.pad(f22, (2, 2), 'constant')\n",
     "\n",
     "fig,axs = plt.subplots(3,1)\n",
-    "axs[0].plot(x,f10,color='k',LineWidth=2)\n",
+    "axs[0].plot(x,f10,color='k',linewidth=2)\n",
     "axs[0].set_xlim(-0.2,1.2)\n",
     "axs[0].set_ylim(-1.75,1.75)\n",
-    "axs[1].plot(x,f21,color='k',LineWidth=2)\n",
+    "axs[1].plot(x,f21,color='k',linewidth=2)\n",
     "axs[1].set_xlim(-0.2,1.2)\n",
     "axs[1].set_ylim(-1.75,1.75)\n",
-    "axs[2].plot(x,f22,color='k',LineWidth=2)\n",
+    "axs[2].plot(x,f22,color='k',linewidth=2)\n",
     "axs[2].set_xlim(-0.2,1.2)\n",
     "axs[2].set_ylim(-1.75,1.75)\n",
     "plt.show()"
@@ -86,7 +86,7 @@
    "source": [
     "x = np.arange(-5,5,0.001)\n",
     "fMexHat = (1-np.power(x,2)) * np.exp(-np.power(x,2)/2)\n",
-    "plt.plot(x,fMexHat,color='k',LineWidth=2)\n",
+    "plt.plot(x,fMexHat,color='k',linewidth=2)\n",
     "plt.show()"
    ]
   },
diff --git a/CH03/CH03_SEC03_1_Underdetermined.ipynb b/CH03/CH03_SEC03_1_Underdetermined.ipynb
index 87ef301..08c444c 100644
--- a/CH03/CH03_SEC03_1_Underdetermined.ipynb
+++ b/CH03/CH03_SEC03_1_Underdetermined.ipynb
@@ -60,9 +60,9 @@
    "source": [
     "fig,axs = plt.subplots(2,2)\n",
     "axs = axs.reshape(-1)\n",
-    "axs[0].plot(s_L1,color='b',LineWidth=1.5)\n",
+    "axs[0].plot(s_L1,color='b',linewidth=1.5)\n",
     "axs[0].set_ylim(-0.2,0.2)\n",
-    "axs[1].plot(s_L2,color='r',LineWidth=1.5)\n",
+    "axs[1].plot(s_L2,color='r',linewidth=1.5)\n",
     "axs[1].set_ylim(-0.2,0.2)\n",
     "axs[2].hist(s_L1,bins=np.arange(-0.105,0.105,0.01),rwidth=0.9)\n",
     "axs[3].hist(s_L2,bins=np.arange(-0.105,0.105,0.01),rwidth=0.9)\n",
diff --git a/CH03/CH03_SEC03_2_AudioCS.ipynb b/CH03/CH03_SEC03_2_AudioCS.ipynb
index 9d0bfb7..bc50362 100644
--- a/CH03/CH03_SEC03_2_AudioCS.ipynb
+++ b/CH03/CH03_SEC03_2_AudioCS.ipynb
@@ -112,23 +112,23 @@
     "fig,axs = plt.subplots(2,2)\n",
     "axs = axs.reshape(-1)\n",
     "\n",
-    "axs[1].plot(freq[:L],PSD[:L],color='k',LineWidth=2)\n",
+    "axs[1].plot(freq[:L],PSD[:L],color='k',linewidth=2)\n",
     "axs[1].set_xlim(0, 1024)\n",
     "axs[1].set_ylim(0, 1200)\n",
     "\n",
-    "axs[0].plot(t,x,color='k',LineWidth=2)\n",
-    "axs[0].plot(perm/n,y,color='r',marker='x',LineWidth=0,ms=12,mew=4)\n",
+    "axs[0].plot(t,x,color='k',linewidth=2)\n",
+    "axs[0].plot(perm/n,y,color='r',marker='x',linewidth=0,ms=12,mew=4)\n",
     "axs[0].set_xlim(time_window[0],time_window[1])\n",
     "axs[0].set_ylim(-2, 2)\n",
     "\n",
-    "axs[2].plot(t,xrecon,color='r',LineWidth=2)\n",
+    "axs[2].plot(t,xrecon,color='r',linewidth=2)\n",
     "axs[2].set_xlim(time_window[0],time_window[1])\n",
     "axs[2].set_ylim(-2, 2)\n",
     "\n",
     "xtrecon = np.fft.fft(xrecon,n) # computes the (fast) discrete fourier transform\n",
     "PSDrecon = xtrecon * np.conj(xtrecon)/n # Power spectrum (how much power in each freq)\n",
     "\n",
-    "axs[3].plot(freq[:L],PSDrecon[:L],color='r',LineWidth=2)\n",
+    "axs[3].plot(freq[:L],PSDrecon[:L],color='r',linewidth=2)\n",
     "axs[3].set_xlim(0, 1024)\n",
     "axs[3].set_ylim(0, 1200)\n",
     "\n",
diff --git a/CH04/CH04_SEC01_LinearRegression.ipynb b/CH04/CH04_SEC01_LinearRegression.ipynb
index 34305df..c6d3f7d 100644
--- a/CH04/CH04_SEC01_LinearRegression.ipynb
+++ b/CH04/CH04_SEC01_LinearRegression.ipynb
@@ -80,9 +80,9 @@
     "\n",
     "plt.figure()\n",
     "plt.plot(xf,y1,color='k',label='E_\\infty')\n",
-    "plt.plot(xf,y2,'--',color='k',LineWidth=2,label='E_1')\n",
-    "plt.plot(xf,y3,color='k',LineWidth=2,label='E_2')\n",
-    "plt.plot(x,y,'o',color='r',LineWidth=2)\n",
+    "plt.plot(xf,y2,'--',color='k',linewidth=2,label='E_1')\n",
+    "plt.plot(xf,y3,color='k',linewidth=2,label='E_2')\n",
+    "plt.plot(x,y,'o',color='r',linewidth=2)\n",
     "\n",
     "plt.ylim(0,4)\n",
     "plt.legend()\n",
@@ -142,9 +142,9 @@
     "\n",
     "plt.figure()\n",
     "plt.plot(xf,y1,color='k',label='E_\\infty')\n",
-    "plt.plot(xf,y2,'--',color='k',LineWidth=2,label='E_1')\n",
-    "plt.plot(xf,y3,color='k',LineWidth=2,label='E_2')\n",
-    "plt.plot(x,y,'o',color='r',LineWidth=2)\n",
+    "plt.plot(xf,y2,'--',color='k',linewidth=2,label='E_1')\n",
+    "plt.plot(xf,y3,color='k',linewidth=2,label='E_2')\n",
+    "plt.plot(x,y,'o',color='r',linewidth=2)\n",
     "\n",
     "plt.ylim(0,4)\n",
     "plt.legend()\n",
diff --git a/CH04/CH04_SEC02_1_GradientDescent.ipynb b/CH04/CH04_SEC02_1_GradientDescent.ipynb
index 9f7f9dc..5930fbf 100644
--- a/CH04/CH04_SEC02_1_GradientDescent.ipynb
+++ b/CH04/CH04_SEC02_1_GradientDescent.ipynb
@@ -150,7 +150,7 @@
     "fig,ax = plt.subplots(1,1,subplot_kw={'projection': '3d'})\n",
     "ax.plot_surface(X, Y, Fquad,linewidth=0,color='k',alpha=0.3)\n",
     "ax.scatter(x,y,f,'o',color='r',s=200)\n",
-    "ax.plot(x,y,f,':',color='k',LineWidth=3)\n",
+    "ax.plot(x,y,f,':',color='k',linewidth=3)\n",
     "ax.contour(X, Y, Fquad, zdir='z', offset=ax.get_zlim()[0], cmap='gray')\n",
     "ax.view_init(elev=40, azim=-140)\n",
     "plt.show()"
@@ -279,11 +279,11 @@
    "source": [
     "fig,ax = plt.subplots(1,1,subplot_kw={'projection': '3d'})\n",
     "ax.plot_surface(X, Y, F-0.2,cmap='binary',alpha=0.5)\n",
-    "ax.plot(x1,y1,f1,'o',color='r',MarkerSize=10)\n",
+    "ax.plot(x1,y1,f1,'o',color='r',markersize=10)\n",
     "ax.plot(x1,y1,f1,':',color='k')\n",
-    "ax.plot(x2,y2,f2,'o',color='m',MarkerSize=10)\n",
+    "ax.plot(x2,y2,f2,'o',color='m',markersize=10)\n",
     "ax.plot(x2,y2,f2,':',color='k')\n",
-    "ax.plot(x3,y3,f3,'o',color='b',MarkerSize=10)\n",
+    "ax.plot(x3,y3,f3,'o',color='b',markersize=10)\n",
     "ax.plot(x3,y3,f3,':',color='k')\n",
     "ax.view_init(elev=40, azim=-100)\n",
     "plt.show()"
@@ -402,11 +402,11 @@
    "source": [
     "fig,ax = plt.subplots(1,1,subplot_kw={'projection': '3d'})\n",
     "ax.plot_surface(X, Y, F-0.2,cmap='binary',alpha=0.5)\n",
-    "ax.plot(x1,y1,f1,'o',color='r',MarkerSize=10)\n",
+    "ax.plot(x1,y1,f1,'o',color='r',markersize=10)\n",
     "ax.plot(x1,y1,f1,':',color='k')\n",
-    "ax.plot(x2,y2,f2,'o',color='m',MarkerSize=10)\n",
+    "ax.plot(x2,y2,f2,'o',color='m',markersize=10)\n",
     "ax.plot(x2,y2,f2,':',color='k')\n",
-    "ax.plot(x3,y3,f3,'o',color='b',MarkerSize=10)\n",
+    "ax.plot(x3,y3,f3,'o',color='b',markersize=10)\n",
     "ax.plot(x3,y3,f3,':',color='k')\n",
     "ax.view_init(elev=40, azim=-100)\n",
     "plt.show()"
diff --git a/CH04/CH04_SEC05_0_Fig4p16_Pareto.ipynb b/CH04/CH04_SEC05_0_Fig4p16_Pareto.ipynb
index cf89360..d24982f 100644
--- a/CH04/CH04_SEC05_0_Fig4p16_Pareto.ipynb
+++ b/CH04/CH04_SEC05_0_Fig4p16_Pareto.ipynb
@@ -45,9 +45,9 @@
     "y3 = np.reshape(y3,-1)\n",
     "x3 = np.tile(x,(1,5)).reshape(-1)\n",
     "\n",
-    "plt.plot(x,y,color='k',LineWidth=2)\n",
+    "plt.plot(x,y,color='k',linewidth=2)\n",
     "\n",
-    "rect = Rectangle((0.5,0.4), 0.9, 1.6,LineWidth=1,edgecolor='k',facecolor='grey',alpha=0.6) \n",
+    "rect = Rectangle((0.5,0.4), 0.9, 1.6,linewidth=1,edgecolor='k',facecolor='grey',alpha=0.6) \n",
     "ax.add_patch(rect)\n",
     "\n",
     "plt.scatter(x2,y2,100,color='magenta',edgecolors='k')\n",
diff --git a/CH04/CH04_SEC05_1_CrossValidate.ipynb b/CH04/CH04_SEC05_1_CrossValidate.ipynb
index e9c09d4..e017c25 100644
--- a/CH04/CH04_SEC05_1_CrossValidate.ipynb
+++ b/CH04/CH04_SEC05_1_CrossValidate.ipynb
@@ -43,8 +43,8 @@
     "ftrain = np.power(x1,2) # Train parabola x = [0,4]\n",
     "ftest = np.power(x2,2)  # Test parabola x = [4,8]\n",
     "\n",
-    "plt.plot(x1,ftrain,color='r',LineWidth=2)\n",
-    "plt.plot(x2,ftest,color='b',LineWidth=2)\n",
+    "plt.plot(x1,ftrain,color='r',linewidth=2)\n",
+    "plt.plot(x2,ftest,color='b',linewidth=2)\n",
     "plt.show()\n",
     "M = 30 # number of model terms\n",
     "Eni = np.zeros((100,M))\n",
diff --git a/CH04/CH04_SEC07_1_ModelValidation.ipynb b/CH04/CH04_SEC07_1_ModelValidation.ipynb
index 6907aeb..a6e11b2 100644
--- a/CH04/CH04_SEC07_1_ModelValidation.ipynb
+++ b/CH04/CH04_SEC07_1_ModelValidation.ipynb
@@ -64,10 +64,10 @@
     "g3 = g3/np.trapz(g3,x)\n",
     "\n",
     "plt.figure()\n",
-    "plt.plot(x,f,LineWidth=2,label='f')\n",
-    "plt.plot(x,g1,LineWidth=2,label='g1')\n",
-    "plt.plot(x,g2,LineWidth=2,label='g2')\n",
-    "plt.plot(x,g3,LineWidth=2,label='g3')\n",
+    "plt.plot(x,f,linewidth=2,label='f')\n",
+    "plt.plot(x,g1,linewidth=2,label='g1')\n",
+    "plt.plot(x,g2,linewidth=2,label='g2')\n",
+    "plt.plot(x,g3,linewidth=2,label='g3')\n",
     "plt.legend()\n",
     "plt.show()"
    ]
diff --git a/CH05/CH05_SEC03_1_Kmeans.ipynb b/CH05/CH05_SEC03_1_Kmeans.ipynb
index 69fb82a..e2a9bbc 100644
--- a/CH05/CH05_SEC03_1_Kmeans.ipynb
+++ b/CH05/CH05_SEC03_1_Kmeans.ipynb
@@ -211,7 +211,7 @@
     "fig,axs = plt.subplots(2)\n",
     "axs[0].plot(x[:n1],y[:n1],'ro')\n",
     "axs[0].plot(x3[:n1],y3[:n1],'bo')\n",
-    "axs[0].plot(xsep,ysep,c='k',LineWidth=2)\n",
+    "axs[0].plot(xsep,ysep,c='k',linewidth=2)\n",
     "axs[0].set_xlim(-2,4)\n",
     "axs[0].set_ylim(-3,2)\n",
     "\n",
@@ -219,7 +219,7 @@
     "\n",
     "axs[1].plot(x[n1:],y[n1:],'ro')\n",
     "axs[1].plot(x3[n1:],y3[n1:],'bo')\n",
-    "axs[1].plot(xsep,ysep,c='k',LineWidth=2)\n",
+    "axs[1].plot(xsep,ysep,c='k',linewidth=2)\n",
     "axs[1].set_xlim(-2,4)\n",
     "axs[1].set_ylim(-3,2)\n",
     "\n",
@@ -279,8 +279,8 @@
    ],
    "source": [
     "plt.bar(range(100),dn['leaves'])\n",
-    "plt.plot(np.array([0, 100]),np.array([50, 50]),'r:',LineWidth=2)\n",
-    "plt.plot(np.array([50.5, 50.5]),np.array([0, 100]),'r:',LineWidth=2)\n",
+    "plt.plot(np.array([0, 100]),np.array([50, 50]),'r:',linewidth=2)\n",
+    "plt.plot(np.array([50.5, 50.5]),np.array([0, 100]),'r:',linewidth=2)\n",
     "\n",
     "plt.show()"
    ]
diff --git a/CH05/CH05_SEC04_1_Dendrogram.ipynb b/CH05/CH05_SEC04_1_Dendrogram.ipynb
index cf6264d..7e05f42 100644
--- a/CH05/CH05_SEC04_1_Dendrogram.ipynb
+++ b/CH05/CH05_SEC04_1_Dendrogram.ipynb
@@ -143,8 +143,8 @@
    ],
    "source": [
     "plt.bar(range(100),dn['leaves'])\n",
-    "plt.plot(np.array([0, 100]),np.array([50, 50]),'r:',LineWidth=2)\n",
-    "plt.plot(np.array([50.5, 50.5]),np.array([0, 100]),'r:',LineWidth=2)\n",
+    "plt.plot(np.array([0, 100]),np.array([50, 50]),'r:',linewidth=2)\n",
+    "plt.plot(np.array([50.5, 50.5]),np.array([0, 100]),'r:',linewidth=2)\n",
     "\n",
     "plt.show()"
    ]
diff --git a/CH05/CH05_SEC05_1_GaussianMixtureModels.ipynb b/CH05/CH05_SEC05_1_GaussianMixtureModels.ipynb
index 299b319..8431689 100644
--- a/CH05/CH05_SEC05_1_GaussianMixtureModels.ipynb
+++ b/CH05/CH05_SEC05_1_GaussianMixtureModels.ipynb
@@ -59,8 +59,8 @@
     "GMModel = GaussianMixture(n_components=2).fit(dogcat)\n",
     "AIC = GMModel.aic(dogcat)\n",
     "\n",
-    "plt.plot(v[:80,1],v[:80,3],'ro',MarkerFaceColor=(0,1,0.2),MarkerEdgeColor='k',ms=12)\n",
-    "plt.plot(v[80:,1],v[80:,3],'bo',MarkerFaceColor=(0.9,0,1),MarkerEdgeColor='k',ms=12)\n",
+    "plt.plot(v[:80,1],v[:80,3],'ro',markerfacecolor=(0,1,0.2),markeredgecolor='k',ms=12)\n",
+    "plt.plot(v[80:,1],v[80:,3],'bo',markerfacecolor=(0.9,0,1),markeredgecolor='k',ms=12)\n",
     "\n",
     "x = np.linspace(-0.15, 0.25)\n",
     "y = np.linspace(-0.25, 0.2)\n",
diff --git a/CH05/CH05_SEC06_1_LDA_Classify.ipynb b/CH05/CH05_SEC06_1_LDA_Classify.ipynb
index e7e96e7..b40cc93 100644
--- a/CH05/CH05_SEC06_1_LDA_Classify.ipynb
+++ b/CH05/CH05_SEC06_1_LDA_Classify.ipynb
@@ -69,8 +69,8 @@
     "fig,axs = plt.subplots(2)\n",
     "axs[0].bar(range(40),test_class)\n",
     "\n",
-    "axs[1].plot(v[:80,1],v[:80,3],'ro',MarkerFaceColor=(0,1,0.2),MarkerEdgeColor='k',ms=12)\n",
-    "axs[1].plot(v[80:,1],v[80:,3],'bo',MarkerFaceColor=(0.9,0,1),MarkerEdgeColor='k',ms=12)\n",
+    "axs[1].plot(v[:80,1],v[:80,3],'ro',markerfacecolor=(0,1,0.2),markeredgecolor='k',ms=12)\n",
+    "axs[1].plot(v[80:,1],v[80:,3],'bo',markerfacecolor=(0.9,0,1),markeredgecolor='k',ms=12)\n",
     "\n",
     "plt.show()"
    ]
@@ -150,8 +150,8 @@
     "fig,axs = plt.subplots(2)\n",
     "axs[0].bar(range(40),test_class)\n",
     "\n",
-    "axs[1].plot(v[:80,1],v[:80,3],'ro',MarkerFaceColor=(0,1,0.2),MarkerEdgeColor='k',ms=12)\n",
-    "axs[1].plot(v[80:,1],v[80:,3],'bo',MarkerFaceColor=(0.9,0,1),MarkerEdgeColor='k',ms=12)\n",
+    "axs[1].plot(v[:80,1],v[:80,3],'ro',markerfacecolor=(0,1,0.2),markeredgecolor='k',ms=12)\n",
+    "axs[1].plot(v[80:,1],v[80:,3],'bo',markerfacecolor=(0.9,0,1),markeredgecolor='k',ms=12)\n",
     "\n",
     "plt.show()"
    ]
@@ -198,7 +198,7 @@
     "    E[jj] = 100*np.sum(np.abs(test_class-truth))/40\n",
     "    \n",
     "plt.bar(range(100),E,color=(0.5,0.5,0.5))\n",
-    "plt.plot(range(100),np.mean(E)*np.ones(100),'r:',LineWidth=3)\n",
+    "plt.plot(range(100),np.mean(E)*np.ones(100),'r:',linewidth=3)\n",
     "plt.show()"
    ]
   },
@@ -242,8 +242,8 @@
     "plt.rcParams['figure.figsize'] = [12, 6]\n",
     "fig,axs = plt.subplots(1,2)\n",
     "for j in range(2):\n",
-    "    axs[j].plot(v[:80,1],v[:80,3],'ro',MarkerFaceColor=(0,1,0.2),MarkerEdgeColor='k',ms=12)\n",
-    "    axs[j].plot(v[80:,1],v[80:,3],'bo',MarkerFaceColor=(0.9,0,1),MarkerEdgeColor='k',ms=12)\n",
+    "    axs[j].plot(v[:80,1],v[:80,3],'ro',markerfacecolor=(0,1,0.2),markeredgecolor='k',ms=12)\n",
+    "    axs[j].plot(v[80:,1],v[80:,3],'bo',markerfacecolor=(0.9,0,1),markeredgecolor='k',ms=12)\n",
     "\n",
     "# Linear Discriminant\n",
     "xtrain = np.concatenate((v[:60,np.array([1,3])],v[80:140,np.array([1,3])]))\n",
@@ -259,7 +259,7 @@
     "\n",
     "\n",
     "x = np.arange(-0.15,0.25,0.005)\n",
-    "axs[0].plot(x,-(L[0]*x+K)/L[1],'k',LineWidth=2)\n",
+    "axs[0].plot(x,-(L[0]*x+K)/L[1],'k',linewidth=2)\n",
     "\n",
     "\n",
     "# Quadratic Discriminant\n",
diff --git a/CH05/CH05_SEC07_1_SVM.ipynb b/CH05/CH05_SEC07_1_SVM.ipynb
index a432121..cb7e68c 100644
--- a/CH05/CH05_SEC07_1_SVM.ipynb
+++ b/CH05/CH05_SEC07_1_SVM.ipynb
@@ -48,8 +48,8 @@
     "x2 = 1.5*np.random.randn(n1) + 1.5\n",
     "y2 = 1.2*np.random.randn(n1) - np.power(x2-1.5,2) + 7\n",
     "\n",
-    "plt.plot(x1,y1,'ro',MarkerFaceColor=(0,1,0.2),MarkerEdgeColor='k',ms=12)\n",
-    "plt.plot(x2,y2,'bo',MarkerFaceColor=(0.9,0,1),MarkerEdgeColor='k',ms=12)\n",
+    "plt.plot(x1,y1,'ro',markerfacecolor=(0,1,0.2),markeredgecolor='k',ms=12)\n",
+    "plt.plot(x2,y2,'bo',markerfacecolor=(0.9,0,1),markeredgecolor='k',ms=12)\n",
     "\n",
     "plt.show()"
    ]
@@ -79,8 +79,8 @@
     "fig = plt.figure()\n",
     "ax = plt.axes(projection='3d')\n",
     "\n",
-    "ax.plot(x1,y1,z1,'ro',MarkerFaceColor=(0,1,0.2),MarkerEdgeColor='k',ms=12)\n",
-    "ax.plot(x2,y2,z2,'bo',MarkerFaceColor=(0.9,0,1),MarkerEdgeColor='k',ms=12)\n",
+    "ax.plot(x1,y1,z1,'ro',markerfacecolor=(0,1,0.2),markeredgecolor='k',ms=12)\n",
+    "ax.plot(x2,y2,z2,'bo',markerfacecolor=(0.9,0,1),markeredgecolor='k',ms=12)\n",
     "\n",
     "ax.view_init(20, -135)\n",
     "\n",
@@ -120,8 +120,8 @@
     "ax = plt.axes(projection='3d')\n",
     "ax.view_init(20, -135)\n",
     "\n",
-    "ax.plot(xr,yr,zr+40,'ro',MarkerFaceColor=(0,1,0.2),MarkerEdgeColor='k',ms=12)\n",
-    "ax.plot(x5,y5,z5+40,'bo',MarkerFaceColor=(0.9,0,1),MarkerEdgeColor='k',ms=12)\n",
+    "ax.plot(xr,yr,zr+40,'ro',markerfacecolor=(0,1,0.2),markeredgecolor='k',ms=12)\n",
+    "ax.plot(x5,y5,z5+40,'bo',markerfacecolor=(0.9,0,1),markeredgecolor='k',ms=12)\n",
     "\n",
     "\n",
     "x = np.arange(-10,10.5,0.5)\n",
@@ -131,8 +131,8 @@
     "\n",
     "ax.plot_surface(X, Y, F3, cmap='gray',linewidth=0, antialiased=True,alpha=0.2)\n",
     "\n",
-    "ax.plot(xr,yr,np.zeros(*xr.shape),'ro',MarkerFaceColor=(179/255,1,179/255),MarkerEdgeColor='k',ms=12)\n",
-    "ax.plot(x5,y5,np.zeros(*x5.shape),'bo',MarkerFaceColor=(240/255,194/255,224/255),MarkerEdgeColor='k',ms=12)\n",
+    "ax.plot(xr,yr,np.zeros(*xr.shape),'ro',markerfacecolor=(179/255,1,179/255),markeredgecolor='k',ms=12)\n",
+    "ax.plot(x5,y5,np.zeros(*x5.shape),'bo',markerfacecolor=(240/255,194/255,224/255),markeredgecolor='k',ms=12)\n",
     "\n",
     "theta = np.linspace(0,2*np.pi,100)\n",
     "xrr = np.sqrt(14)*np.cos(theta)\n",