numpy.mean

Classification with scikit-learn: telling two populations apart

• acc_ceiling = np.mean(bayes_rule(X_fresh) == y_fresh)
• print(f"Bayes rule, the 500 test events: {np.mean(bayes_rule(X_test) == y_test):.3f}")

Filtering with scipy.signal: mains hum and noise out of an ECG

• return np.sqrt(np.mean(d ** 2))
• print(f"R peak height {100 * np.mean(y[r_true] / ecg[r_true]):.0f} % of the truth")

Random numbers with numpy.random: ten thousand reproducible random walks

• print(f"{name:8s} ⟨x_N²⟩ = {np.mean(final**2):7.1f}")
• return np.mean(x_N**2)
• print(f"combined: {np.mean(results):6.1f}")

scipy.stats from the ground up: is the difference between two samples real?

• return np.mean(x, axis=axis) - np.mean(y, axis=axis)
• above, below = np.mean(null >= diff), np.mean(null <= -diff)
• print(f"runs with p < 0.05: {100 * np.mean(rep.pvalue < 0.05):.1f} %")
• print(f"intervals containing 2.0 mg: {100 * np.mean(covers):.1f} %")
• print(f"26 batches per protocol: p < 0.05 in {100 * np.mean(rerun(26)[2].pvalue < 0.05):.1f} % of runs")
• f"{100 * np.mean(skew_ln > s_high):.0f} % above {s_high:.2f}")

and 1 more line in this tutorial