def std(self, axis=None, dtype=None, out=None, ddof=1): """ Compute unbiased standard deviation of non-null values """ nona = remove_na(self.values) if len(nona) < 2: return NaN return ndarray.std(nona, axis, dtype, out, ddof)
def std(self, axis=None, dtype=None, out=None, ddof=1): """ Unbiased standard deviation of non-null values """ nona = remove_na(self.values) if len(nona) < 2: return NaN return ndarray.std(nona, axis, dtype, out, ddof)
def compute_stats(samples_healthy, samples_unhealthy, samples_s, samples): samples_healthy = array(samples_healthy) samples_unhealthy = array(samples_unhealthy) samples_s = array(samples_s) healthy_mean = ndarray.mean(samples_healthy,0) unhealthy_mean = ndarray.mean(samples_unhealthy, 0) s_mean = ndarray.mean(samples_s, 0) healthy_std = ndarray.std(samples_healthy, 0, dtype=float64) unhealthy_std = ndarray.std(samples_unhealthy, 0, dtype=float64) s_std = ndarray.std(samples_s, 0, dtype=float64) print "Healthy means: \t", healthy_mean print "Unhealthy means: \t", unhealthy_mean print "S means: \t\t", s_mean print "Healthy standard dev: \t", healthy_std print "Unhealthy standard dev: \t", unhealthy_std print "S standard dev: \t", s_std x1 = range(len(samples_healthy)) x2 = range(len(samples_unhealthy)) x3 = range(len(samples_s)) dur1 = [sample[0] for sample in samples_healthy] mean1 = [sample[1] for sample in samples_healthy] var1 = [sample[2] for sample in samples_healthy] dur2 = [sample[0] for sample in samples_unhealthy] mean2 = [sample[1] for sample in samples_unhealthy] var2 = [sample[2] for sample in samples_unhealthy] dur3 = [sample[0] for sample in samples_s] mean3 = [sample[1] for sample in samples_s] var3 = [sample[2] for sample in samples_s] plt.figure() plt.hist(mean1, normed=True, bins=100, color='red', range=(0, 50)) plt.figure() plt.hist(mean2, normed=True, bins=100, color='green', range=(0, 50)) plt.figure() plt.hist(mean3, normed=True, bins=100, color='blue', range=(0, 50)) plt.show()