added code used to plot

Author: tgiani

plot_analysis.py

@@ -0,0 +1,111 @@
+from __future__ import division
+import numpy as np
+import matplotlib.pyplot as plt
+import matplotlib as mpl
+import sys
+
+
+# data
+epochs3, mag3, mag_err3, en3, en_err3, susc3, susc_err3, cv3, cv_err3, log_likelihood_mean3, ll_up3, ll_down3   = np.loadtxt("./test3_ft_0.001_k5/analysis/analysis_1.8_L8/data.dat",usecols=(0,1,2,3,4,5,6,7,8,9,10,11), unpack=True, skiprows=1)
+
+
+epochs2, mag2, mag_err2, en2, en_err2, susc2, susc_err2, cv2, cv_err2, log_likelihood_mean2, ll_up2, ll_down2   = np.loadtxt("./test3_ft_0.01_k5/analysis/analysis_1.8_L8/data.dat",usecols=(0,1,2,3,4,5,6,7,8,9,10,11), unpack=True, skiprows=1)
+
+
+epochs1, mag1, mag_err1, en1, en_err1, susc1, susc_err1, cv1, cv_err1, log_likelihood_mean1, ll_up1, ll_down1   = np.loadtxt("./test3/analysis/analysis_1.8_L8/data.dat",usecols=(0,1,2,3,4,5,6,7,8,9,10,11), unpack=True, skiprows=1)
+
+
+epochs = np.concatenate((epochs1, epochs2, epochs3), axis=0)
+mag = np.concatenate((mag1,mag2,mag3 ), axis=0)
+mag_err = np.concatenate((mag_err1,mag_err2,mag_err3), axis=0)
+en = np.concatenate((en1, en2, en3), axis=0)
+en_err = np.concatenate((en_err1, en_err2, en_err3), axis=0)
+susc = np.concatenate((susc1, susc2, susc3), axis=0)
+susc_err = np.concatenate((susc_err1, susc_err2,susc_err3), axis=0)
+cv = np.concatenate((cv1, cv2, cv3), axis=0)
+cv_err = np.concatenate((cv_err1, cv_err2, cv_err3), axis=0)
+log_likelihood_mean = np.concatenate((log_likelihood_mean1, log_likelihood_mean2, log_likelihood_mean3), axis=0)
+ll_up = np.concatenate((ll_up1, ll_up2, ll_up3), axis=0)
+ll_down = np.concatenate((ll_down1, ll_down2, ll_down3), axis=0)
+
+
+
+# expected values from magneto
+npoints = epochs.size
+
+mag_ = 0.95669*np.ones(npoints)
+en_  = -1.85910*np.ones(npoints)
+chi_ = 0.1167*np.ones(npoints)  
+cv_  = 0.44058*np.ones(npoints)
+
+
+## Observables vs number of epoch ##
+
+plt.figure(figsize=(15, 5))
+plt.scatter(epochs, mag)
+plt.errorbar(epochs, mag, yerr = mag_err, elinewidth = 0.8)
+plt.plot(epochs, mag_)
+plt.ylabel("m", fontsize=18)
+plt.xlabel("epoch", fontsize=18)
+plt.suptitle('Magnetization vs number of epochs', fontsize=20)
+#plt.show()
+plt.savefig("./test3/analysis/analysis_1.8_L8/mag_1.8.png")
+plt.close()
+
+plt.figure(figsize=(15, 5))
+plt.scatter(epochs, en)
+plt.errorbar(epochs, en, yerr = en_err, elinewidth = 0.8)
+plt.plot(epochs, en_)
+plt.ylabel("energy", fontsize=18)
+plt.xlabel("epoch", fontsize=18)
+plt.suptitle("Energy vs number of epochs", fontsize=20)
+#plt.show()
+plt.savefig("/home/s1792848/Documents/RBM/rbm_ising/figs/energy_1.8.png")
+plt.close()
+
+
+plt.figure(figsize=(15, 5))
+plt.scatter(epochs, susc)
+plt.errorbar(epochs, susc, yerr = susc_err, elinewidth = 0.8)
+plt.plot(epochs, chi_)
+plt.ylabel("chi", fontsize=18)
+plt.xlabel("epoch", fontsize=18)
+plt.suptitle("Susceptibility vs number of epochs", fontsize=20)
+#plt.show()
+plt.savefig("/home/s1792848/Documents/RBM/rbm_ising/figs/chi_1.8.png")
+plt.close()
+
+
+plt.figure(figsize=(15, 5))
+plt.scatter(epochs, cv)
+plt.errorbar(epochs, cv, yerr = cv_err, elinewidth = 0.8)
+plt.plot(epochs, cv_)
+plt.ylabel("Cv", fontsize=18)
+plt.xlabel("epoch", fontsize=18)
+plt.suptitle("Heat capacity vs number of epochs", fontsize=20)
+#plt.show()
+plt.savefig("/home/s1792848/Documents/RBM/rbm_ising/figs/cv_1.8.png")
+plt.close()
+
+"""
+plt.figure(figsize=(15, 5))
+plt.scatter(epochs, log_likelihood_mean)
+plt.suptitle("-Log-Likelihood vs number of epochs", fontsize=20)
+plt.ylabel("-LL", fontsize=18)
+plt.xlabel("epoch", fontsize=18)
+plt.savefig("./analysis_1.8_L8/LL_1.8.png")
+plt.close()
+"""
+
+plt.figure(figsize=(15, 5))
+plt.scatter(epochs, log_likelihood_mean)
+plt.errorbar(epochs, log_likelihood_mean, yerr = [ll_up, ll_down], elinewidth = 0.8)
+plt.suptitle("-Log-Likelihood vs number of epochs", fontsize=20)
+plt.ylabel("-LL", fontsize=18)
+plt.xlabel("epoch", fontsize=18)
+#plt.show()
+plt.savefig("/home/s1792848/Documents/RBM/rbm_ising/figs/LL_1.8.png")
+plt.close()
+
+
+