df1.describe()

df1.boxplot() df1.plot.scatter('age','fat') plt.show() plt.scatter(df1['age'],df1['fat']) plt.plot([0,70],[0,70],color='red') q1 = df1.quantile(0.25) median = df1.median() q3 = df1.quantile(0.75) plt.plot(q1[0],q1[1],'o',color='gold',markersize=10) plt.plot(median[0],median[1],'o',color='gold',markersize=10) plt.plot(q3[0],q3[1],'o',color='gold',markersize=10) plt.annotate('$Q_{1}$',xy=(q1[0],q1[1]),xytext=(-5,-25),textcoords='offset points',fontsize=16) plt.annotate('$median$',xy=(median[0],median[1]),xytext=(-33,+18),textcoords='offset points',fontsize=16) plt.annotate('$Q_{3}$',xy=(q3[0],q3[1]),xytext=(+12,-15),textcoords='offset points',fontsize=16) plt.show()

df2 = pd.DataFrame({ 'A1':pd.Series([1.5, 2, 1.6, 1.2, 1.5]), 'A2':pd.Series([1.7, 1.9, 1.8, 1.5, 1]) }) x = np.array([1.4,1.6]) euclid = df2.apply(lambda row:np.linalg.norm(row-x,2),1).sort_values() manhattan = df2.apply(lambda row:np.linalg.norm(row-x,1),1).sort_values() inf = df2.apply(lambda row:np.linalg.norm(row-x,np.inf),1).sort_values() cos = df2.apply(lambda row:np.dot(row,x)/(np.linalg.norm(row)*np.linalg.norm(x)),1).sort_values(ascending=False) normalized = df2.apply(lambda row:np.linalg.norm(row/np.linalg.norm(row)-x),1).sort_values() rank = pd.DataFrame({ 'euclid':euclid.index, 'manhattan':manhattan.index, 'inf':inf.index, 'cos':cos.index, 'normalized':normalized.index }) rank

age = np.array([13, 15, 16, 16, 19, 20, 20, 21, 22, 22, 25, 25, 25, 25, 30, 33, 33, 35, 35, 35, 35, 36, 40, 45, 46, 52, 70]) bin_smooth = np.sort(age).reshape(int(age.size/3),3) bin_smooth = np.ones((3,9))*bin_smooth.mean(axis=1) bin_smooth.T

minmax = (35-age.min())/(age.max()-age.min())*(1.0-0.0)+0.0 z_score = (35-age.mean())/age.std() decimal_scaling = 35/10**np.ceil(np.log10(age.max())) print("minmax:",minmax,"z_score:",z_score,"decimal_scaling:",decimal_scaling)