fix conflict & pull hw5

hw4/hw4_10.py

@@ -33,32 +33,57 @@ def error(gt, pred):
         err = (err+1) if gt[index]!=pred[index] else err
     return err/len(gt)
 
-if __name__ == '__main__':
-    x, y = read_data(FILENAME)
-    x = form(x)
-    prob = problem(y, x)
-    lambda_powers = [-6, -4, -2, 0, 2]
+def transform(features):
+    output_features = []
+    for index, feature in enumerate(features):
+        output_features.append([ 0 for _ in range(84) ])
+        output_features[index][0] = 1
+        
+        d_index = 1   
+        # 1-order
+        for i in feature:
+            output_features[index][d_index] = i
+            d_index += 1
+    
+        # 2-orde
+        for i in range(len(feature)):
+            for j in range(i, len(feature)):
+                output_features[index][d_index] = feature[i]*feature[j]
+                d_index += 1
+        # 3-order
+        for i in range(len(feature)):
+            for j in range(i, len(feature)):
+                for k in range(j, len(feature)):
+                    output_features[index][d_index] = i*j*k
+                    d_index += 1
+    return output_features
 
-    results = []
-    for lambda_power in lambda_powers:
-        lambda_value = 10 ** lambda_power
-        param_C = 1/(2*lambda_value) 
-        param = parameter('-s 0 -c {} -e 0.000001 -q'.format(param_C))
-        model = train(prob, param)
-        p_label, p_acc, p_val = predict(y, x, model)
-        err = error(y, p_label)
-        print("0/1 error: ", err)
-        print()
-        results.append({'lambda': lambda_power, 'error': err})
+x, y = read_data(FILENAME)
+x = transform(x)
+x = form(x)
+prob = problem(y, x)
+lambda_powers = [-6, -4, -2, 0, 2]
 
-    ans, min_err = None, 1
-    for i in results:
-        print(i)
-        if i['error'] <= min_err:
-            min_err = i['error']
-            ans = i
+results = []
+for lambda_power in lambda_powers:
+    lambda_value = 10 ** lambda_power
+    param_C = 1/(2*lambda_value) 
+    param = parameter('-s 0 -c {} -e 0.000001 -q'.format(param_C))
+    model = train(prob, param)
+    p_label, p_acc, p_val = predict(y, x, model)
+    err = error(y, p_label)
+    print("0/1 error: ", err)
+    print()
+    results.append({'lambda': lambda_power, 'error': err})
 
-    print("the largest lambda: {}, log_10(lambda*): {}".format(10**ans['lambda'], ans['lambda']))
+ans, min_err = None, 1
+for i in results:
+    print(i['error'])
+    if i['error'] <= min_err:
+        min_err = i['error']
+        ans = i
+
+print("the largest lambda: {}, log_10(lambda*): {}".format(10**ans['lambda'], ans['lambda']))
 
         
 

hw4/hw4_11.py

@@ -43,10 +43,37 @@ def new_split(x, y):
     train_y, val_y = y[:120], y[120:]
     return (train_x, train_y), (val_x, val_y)
 
+def transform(features):
+    output_features = []
+    for index, feature in enumerate(features):
+        output_features.append([ 0 for _ in range(84) ])
+        output_features[index][0] = 1
+        
+        d_index = 1   
+        # 1-order
+        for i in feature:
+            output_features[index][d_index] = i
+            d_index += 1
+    
+        # 2-orde
+        for i in range(len(feature)):
+            for j in range(i, len(feature)):
+                output_features[index][d_index] = feature[i]*feature[j]
+                d_index += 1
+        # 3-order
+        for i in range(len(feature)):
+            for j in range(i, len(feature)):
+                for k in range(j, len(feature)):
+                    output_features[index][d_index] = i*j*k
+                    d_index += 1
+    return output_features
+
 x, y = read_data(FILENAME)
+x = transform(x)
 x = format(x)
 log_lambda = []
-for i in range(128):
+for index in range(128):
+    random.seed(datetime.datetime.now().timestamp()+index)
     (train_x, train_y), (val_x, val_y) = new_split(x, y)
     random.seed(datetime.datetime.now().timestamp()+i)
     prob = problem(train_y, train_x)

hw4/hw4_12.py

@@ -52,7 +52,33 @@ def new_split(x, y):
 
     return folds
 
+def transform(features):
+    output_features = []
+    for index, feature in enumerate(features):
+        output_features.append([ 0 for _ in range(84) ])
+        output_features[index][0] = 1
+        
+        d_index = 1   
+        # 1-order
+        for i in feature:
+            output_features[index][d_index] = i
+            d_index += 1
+    
+        # 2-orde
+        for i in range(len(feature)):
+            for j in range(i, len(feature)):
+                output_features[index][d_index] = feature[i]*feature[j]
+                d_index += 1
+        # 3-order
+        for i in range(len(feature)):
+            for j in range(i, len(feature)):
+                for k in range(j, len(feature)):
+                    output_features[index][d_index] = i*j*k
+                    d_index += 1
+    return output_features
+
 x, y = read_data(FILENAME)
+x = transform(x)
 x = format(x)
 log_lambda = []
 lambda_powers = [-6, -4, -2, 0, 2]

hw5/hw5_10.py

@@ -0,0 +1,48 @@
+import numpy as np
+import datetime
+import random
+from libsvm.svmutil import *
+import matplotlib.pyplot as plt
+
+FILENAME = "satimage.scale"
+TEST_FILENAME = "satimage.scale.t"
+TARGET = 1
+
+def new_label(y, target):
+    ans = []
+    for i in y:
+        if i == target:
+            ans.append(1)
+        else:
+            ans.append(0)
+    return ans
+
+def error(predict, gt):
+    error_count = 0
+    for index in range(len(predict)):
+        if predict[index] != gt[index]:
+            error_count += 1
+    return error_count / len(predict)
+
+if __name__ == '__main__':
+    
+    y, x = svm_read_problem(FILENAME)
+    y = new_label(y, TARGET)
+
+    test_y, test_x = svm_read_problem(TEST_FILENAME)
+    test_y = new_label(test_y, TARGET)
+
+    for c in [0.01, 0.1, 1, 10, 100]:
+        print("C=", c)
+        prob  = svm_problem(y, x)
+        param = svm_parameter('-s 0 -t 2 -g 1 -c {} -q'.format(c))
+        m = svm_train(prob, param)
+
+        p_label, p_acc, p_val = svm_predict(test_y, test_x, m)
+        my_error = error(p_label, test_y)
+        print("p_acc:", p_acc)
+        print("0/1 error:", my_error)
+
+
+        print("="*20)
+

hw5/hw5_9.py

@@ -0,0 +1,30 @@
+import numpy as np
+import datetime
+import random
+from libsvm.svmutil import *
+import matplotlib.pyplot as plt
+
+FILENAME = "satimage.scale"
+TEST_FILENAME = "satimage.scale.t"
+
+def new_label(y, target):
+    ans = []
+    for i in y:
+        if i == target:
+            ans.append(1)
+        else:
+            ans.append(0)
+    return ans
+
+if __name__ == '__main__':
+    y, x = svm_read_problem(FILENAME)
+    y = new_label(y, 4)
+
+    for c in [0.1, 1, 10]:
+        for q in [2, 3, 4]:
+            print("(C, Q)=({}, {})".format(c, q))
+            prob  = svm_problem(y, x)
+            param = svm_parameter('-s 0 -t 1 -d {} -c {}'.format(q, c))
+            m = svm_train(prob, param)
+            print("="*20)
+