Consistent use of plt in ex2

exercises/02450Toolbox_Python/Scripts/ex2_1_1.py

@@ -5,7 +5,7 @@ x = np.array([-0.68, -2.11, 2.39, 0.26, 1.46, 1.33, 1.03, -0.41, -0.33, 0.47])
 
 # Compute values
 mean_x = x.mean()
-std_x = x.std(ddof=1)
+std_x = x.std(ddof=1) # ddof: Delta Degrees of freedom
 median_x = np.median(x)
 range_x = x.max() - x.min()
 

exercises/02450Toolbox_Python/Scripts/ex2_2_1.py

@@ -12,7 +12,7 @@ filename = importlib_resources.files("dtuimldmtools").joinpath("data/digits.mat"
 i = 1
 
 # Similarity: 'SMC', 'Jaccard', 'ExtendedJaccard', 'Cosine', 'Correlation'
-similarity_measure = "SMC"
+similarity_measure = "Jaccard"
 
 # Load the digits
 # Load Matlab data file to python dict structure
@@ -32,7 +32,6 @@ sim = sim.tolist()[0]
 # Tuples of sorted similarities and their indices
 sim_to_index = sorted(zip(sim, noti))
 
-
 # Visualize query image and 5 most/least similar images
 plt.figure(figsize=(12, 8))
 plt.subplot(3, 1, 1)

exercises/02450Toolbox_Python/Scripts/ex2_2_2.py

 # exercise 2.2.2
-
 import numpy as np
 
 from dtuimldmtools import similarity

exercises/02450Toolbox_Python/Scripts/ex2_3_1.py

 # exercise 2.3.1
-
 import importlib_resources
 import numpy as np
 import xlrd
@@ -30,4 +29,4 @@ N = len(y)
 M = len(attributeNames)
 C = len(classNames)
 
-print("Ran Exercise 2.3.1")
+print("Ran Exercise 2.3.1 - loading the Iris data")
\ No newline at end of file

exercises/02450Toolbox_Python/Scripts/ex2_3_2.py

@@ -2,17 +2,17 @@
 import numpy as np
 # (requires data from exercise 2.3.1 so will run that script first)
 from ex2_3_1 import *
-from matplotlib.pyplot import figure, hist, show, subplot, xlabel, ylim
+import matplotlib.pyplot as plt
 
-figure(figsize=(8, 7))
+plt.figure(figsize=(8, 7))
 u = np.floor(np.sqrt(M))
 v = np.ceil(float(M) / u)
 for i in range(M):
-    subplot(int(u), int(v), i + 1)
-    hist(X[:, i], color=(0.2, 0.8 - i * 0.2, 0.4))
-    xlabel(attributeNames[i])
-    ylim(0, N / 2)
+    plt.subplot(int(u), int(v), i + 1)
+    plt.hist(X[:, i], color=(0.2, 0.8 - i * 0.2, 0.4))
+    plt.xlabel(attributeNames[i])
+    plt.ylim(0, N / 2)
 
-show()
+plt.show()
 
 print("Ran Exercise 2.3.2")

exercises/02450Toolbox_Python/Scripts/ex2_3_3.py

 # Exercise 2.3.3
 # (requires data from exercise 2.3.1)
 from ex2_3_1 import *
-from matplotlib.pyplot import boxplot, show, title, xticks, ylabel
+import matplotlib.pyplot as plt
 
-boxplot(X)
-xticks(range(1, 5), attributeNames)
-ylabel("cm")
-title("Fisher's Iris data set - boxplot")
-show()
+plt.figure()
+plt.boxplot(X)
+plt.xticks(range(1, 5), attributeNames)
+plt.ylabel("cm")
+plt.title("Fisher's Iris data set - boxplot")
+plt.show()
 
 print("Ran Exercise 2.3.3")

exercises/02450Toolbox_Python/Scripts/ex2_3_4.py

 # Exercise 2.3.4
 # requires data from exercise 4.1.1
 from ex2_3_1 import *
-from matplotlib.pyplot import boxplot, figure, show, subplot, title, xticks, ylim
+import matplotlib.pyplot as plt
 
-figure(figsize=(14, 7))
+plt.figure(figsize=(14, 7))
 for c in range(C):
-    subplot(1, C, c + 1)
+    plt.subplot(1, C, c + 1)
     class_mask = y == c  # binary mask to extract elements of class c
     # or: class_mask = nonzero(y==c)[0].tolist()[0] # indices of class c
 
-    boxplot(X[class_mask, :])
+    plt.boxplot(X[class_mask, :])
     # title('Class: {0}'.format(classNames[c]))
-    title("Class: " + classNames[c])
-    xticks(
+    plt.title("Class: " + classNames[c])
+    plt.xticks(
         range(1, len(attributeNames) + 1), [a[:7] for a in attributeNames], rotation=45
     )
     y_up = X.max() + (X.max() - X.min()) * 0.1
     y_down = X.min() - (X.max() - X.min()) * 0.1
-    ylim(y_down, y_up)
+    plt.ylim(y_down, y_up)
 
-show()
+plt.show()
 
 print("Ran Exercise 2.3.4")

exercises/02450Toolbox_Python/Scripts/ex2_3_5.py

 # Exercise 2.3.5
 # (requires data from exercise 2.3.1)
 from ex2_3_1 import *
-from matplotlib.pyplot import (
-    figure,
-    legend,
-    plot,
-    show,
-    subplot,
-    xlabel,
-    xticks,
-    ylabel,
-    yticks,
-)
+import matplotlib.pyplot as plt
 
-figure(figsize=(12, 10))
+plt.figure(figsize=(12, 10))
 for m1 in range(M):
     for m2 in range(M):
-        subplot(M, M, m1 * M + m2 + 1)
+        plt.subplot(M, M, m1 * M + m2 + 1)
         for c in range(C):
             class_mask = y == c
-            plot(np.array(X[class_mask, m2]), np.array(X[class_mask, m1]), ".")
+            plt.plot(np.array(X[class_mask, m2]), np.array(X[class_mask, m1]), ".")
             if m1 == M - 1:
-                xlabel(attributeNames[m2])
+                plt.xlabel(attributeNames[m2])
             else:
-                xticks([])
+                plt.xticks([])
             if m2 == 0:
-                ylabel(attributeNames[m1])
+                plt.ylabel(attributeNames[m1])
             else:
-                yticks([])
+                plt.yticks([])
                             
-legend(classNames)
+plt.legend(classNames)
 
-show()
+plt.show()
 
 print("Ran Exercise 2.3.5")

exercises/02450Toolbox_Python/Scripts/ex2_3_6.py

 # Exercise 2.3.6
-
-# requires data from exercise 2.3.1
+# (requires data from exercise 2.3.1)
 from ex2_3_1 import *
-from matplotlib.pyplot import figure, show
+import matplotlib.pyplot as plt
 from mpl_toolkits.mplot3d import Axes3D
 
 # Indices of the variables to plot
 ind = [0, 1, 2]
 colors = ["blue", "green", "red"]
 
-f = figure()
+f = plt.figure()
 ax = f.add_subplot(111, projection="3d")  # Here the mpl_toolkits is used
 for c in range(C):
     class_mask = y == c
@@ -22,6 +21,6 @@ ax.set_xlabel(attributeNames[ind[0]])
 ax.set_ylabel(attributeNames[ind[1]])
 ax.set_zlabel(attributeNames[ind[2]])
 
-show()
+plt.show()
 
 print("Ran Exercise 2.3.6")

exercises/02450Toolbox_Python/Scripts/ex2_3_7.py

 # Exercise 2.3.7
-
-# requires data from exercise 2.3.7
+# (requires data from exercise 2.3.1)
 from ex2_3_1 import *
-from matplotlib.pyplot import (
-    cm,
-    colorbar,
-    figure,
-    imshow,
-    show,
-    title,
-    xlabel,
-    xticks,
-    ylabel,
-)
+import matplotlib.pyplot as plt
 from scipy.stats import zscore
 
 X_standarized = zscore(X, ddof=1)
 
-figure(figsize=(12, 6))
-imshow(X_standarized, interpolation="none", aspect=(4.0 / N), cmap=cm.gray)
-xticks(range(4), attributeNames)
-xlabel("Attributes")
-ylabel("Data objects")
-title("Fisher's Iris data matrix")
-colorbar()
+plt.figure(figsize=(12, 6))
+plt.imshow(X_standarized, interpolation="none", aspect=(4.0 / N), cmap=plt.cm.gray) 
+plt.xticks(range(4), attributeNames)
+plt.xlabel("Attributes")
+plt.ylabel("Data objects")
+plt.title("Fisher's Iris data matrix")
+plt.colorbar()
 
-show()
+plt.show()
 
 print("Ran Exercise 2.3.7")
\ No newline at end of file

exercises/02450Toolbox_Python/Scripts/ex2_4_1.py

 # exercise 2.4.1
-
+"""
+Note: This is a long script. You may want to use breakpoint 
+"""
 import importlib_resources
 import numpy as np
-from matplotlib.pyplot import (
-    boxplot,
-    figure,
-    hist,
-    show,
-    subplot,
-    title,
-    xlabel,
-    xticks,
-    ylim,
-    yticks,
-)
+import matplotlib.pyplot as plt
 from scipy.io import loadmat
 from scipy.stats import zscore
 
@@ -29,56 +20,61 @@ N = mat_data["N"][0, 0]
 attributeNames = [name[0][0] for name in mat_data["attributeNames"]]
 classNames = [cls[0][0] for cls in mat_data["classNames"]]
 
+print("Data loaded")
+
 # We start with a box plot of each attribute
-figure()
-title("Wine: Boxplot")
-boxplot(X)
-xticks(range(1, M + 1), attributeNames, rotation=45)
+plt.figure()
+plt.title("Wine: Boxplot")
+plt.boxplot(X)
+plt.xticks(range(1, M + 1), attributeNames, rotation=45)
 
 # From this it is clear that there are some outliers in the Alcohol
 # attribute (10x10^14 is clearly not a proper value for alcohol content)
 # However, it is impossible to see the distribution of the data, because
 # the axis is dominated by these extreme outliers. To avoid this, we plot a
 # box plot of standardized data (using the zscore function).
-figure(figsize=(12, 6))
-title("Wine: Boxplot (standarized)")
-boxplot(zscore(X, ddof=1), attributeNames)
-xticks(range(1, M + 1), attributeNames, rotation=45)
+plt.figure(figsize=(12, 6))
+plt.title("Wine: Boxplot (standarized)")
+plt.boxplot(zscore(X, ddof=1), attributeNames)
+plt.xticks(range(1, M + 1), attributeNames, rotation=45)
 
 # This plot reveals that there are clearly some outliers in the Volatile
 # acidity, Density, and Alcohol attributes, i.e. attribute number 2, 8,
 # and 11.
+plt.show()
 
 # Next, we plot histograms of all attributes.
-figure(figsize=(14, 9))
+plt.figure(figsize=(14, 9))
 u = np.floor(np.sqrt(M))
 v = np.ceil(float(M) / u)
 for i in range(M):
-    subplot(int(u), int(v), i + 1)
-    hist(X[:, i])
-    xlabel(attributeNames[i])
-    ylim(0, N)  # Make the y-axes equal for improved readability
+    plt.subplot(int(u), int(v), i + 1)
+    plt.hist(X[:, i])
+    plt.xlabel(attributeNames[i])
+    plt.ylim(0, N)  # Make the y-axes equal for improved readability
     if i % v != 0:
-        yticks([])
+        plt.yticks([])
     if i == 0:
-        title("Wine: Histogram")
+        plt.title("Wine: Histogram")
 
+plt.show()
 
 # This confirms our belief about outliers in attributes 2, 8, and 11.
 # To take a closer look at this, we next plot histograms of the
 # attributes we suspect contains outliers
-figure(figsize=(14, 9))
+plt.figure(figsize=(14, 9))
 m = [1, 7, 10]
 for i in range(len(m)):
-    subplot(1, len(m), i + 1)
-    hist(X[:, m[i]], 50)
-    xlabel(attributeNames[m[i]])
-    ylim(0, N)  # Make the y-axes equal for improved readability
+    plt.subplot(1, len(m), i + 1)
+    plt.hist(X[:, m[i]], 50)
+    plt.xlabel(attributeNames[m[i]])
+    plt.ylim(0, N)  # Make the y-axes equal for improved readability
     if i > 0:
-        yticks([])
+        plt.yticks([])
     if i == 0:
-        title("Wine: Histogram (selected attributes)")
+        plt.title("Wine: Histogram (selected attributes)")
 
+plt.show()
 
 # The histograms show that there are a few very extreme values in these
 # three attributes. To identify these values as outliers, we must use our
@@ -95,25 +91,24 @@ X = X[valid_mask, :]
 y = y[valid_mask]
 N = len(y)
 
-
 # Now, we can repeat the process to see if there are any more outliers
 # present in the data. We take a look at a histogram of all attributes:
-figure(figsize=(14, 9))
+plt.figure(figsize=(14, 9))
 u = np.floor(np.sqrt(M))
 v = np.ceil(float(M) / u)
 for i in range(M):
-    subplot(int(u), int(v), i + 1)
-    hist(X[:, i])
-    xlabel(attributeNames[i])
-    ylim(0, N)  # Make the y-axes equal for improved readability
+    plt.subplot(int(u), int(v), i + 1)
+    plt.hist(X[:, i])
+    plt.xlabel(attributeNames[i])
+    plt.ylim(0, N)  # Make the y-axes equal for improved readability
     if i % v != 0:
-        yticks([])
+        plt.yticks([])
     if i == 0:
-        title("Wine: Histogram (after outlier detection)")
+        plt.title("Wine: Histogram (after outlier detection)")
 
 # This reveals no further outliers, and we conclude that all outliers have
 # been detected and removed.
 
-show()
+plt.show()
 
 print("Ran Exercise 2.4.1")

exercises/02450Toolbox_Python/Scripts/ex2_4_2.py

@@ -2,19 +2,10 @@
 
 import importlib_resources
 import numpy as np
-from matplotlib.pyplot import (
-    figure,
-    legend,
-    plot,
-    show,
-    subplot,
-    xlabel,
-    xticks,
-    ylabel,
-    yticks,
-)
+import matplotlib.pyplot as plt
 from scipy.io import loadmat
 from scipy.stats import zscore
+from dtuimldmtools import similarity
 
 filename = importlib_resources.files("dtuimldmtools").joinpath("data/wine.mat")
 
@@ -49,24 +40,24 @@ Xnorm = zscore(X, ddof=1)
 Attributes = [1, 4, 5, 6]
 NumAtr = len(Attributes)
 
-figure(figsize=(12, 12))
+plt.figure(figsize=(12, 12))
 for m1 in range(NumAtr):
     for m2 in range(NumAtr):
-        subplot(NumAtr, NumAtr, m1 * NumAtr + m2 + 1)
+        plt.subplot(NumAtr, NumAtr, m1 * NumAtr + m2 + 1)
         for c in range(C):
             class_mask = y == c
-            plot(X[class_mask, Attributes[m2]], X[class_mask, Attributes[m1]], ".")
+            plt.plot(X[class_mask, Attributes[m2]], X[class_mask, Attributes[m1]], ".")
             if m1 == NumAtr - 1:
-                xlabel(attributeNames[Attributes[m2]])
+                plt.xlabel(attributeNames[Attributes[m2]])
             else:
-                xticks([])
+                plt.xticks([])
             if m2 == 0:
-                ylabel(attributeNames[Attributes[m1]])
+                plt.ylabel(attributeNames[Attributes[m1]])
             else:
-                yticks([])
+                plt.yticks([])
             # ylim(0,X.max()*1.1)
             # xlim(0,X.max()*1.1)
-legend(classNames)
-show()
+plt.legend(classNames)
+plt.show()
 
 print("Ran Exercise 2.4.2")
\ No newline at end of file