Initial commit

8knn_fail.py

+import numpy as np
+import matplotlib.pyplot as plt
+from knn_7 import KNN
+
+def get_data():
+	width = 8
+	height = 8
+	N = width * height
+	X = np.zeros((N, 2))
+	Y = np.zeros(N)
+	n = 0
+	start_t = 0
+	for i in range(width):
+		t = start_t
+		for j in range(height):
+			X[n] = [i, j]
+			Y[n] = t
+			n += 1
+			t = (t + 1) % 2
+		start_t = (start_t + 1) % 2
+	return X, Y
+
+if __name__ == "__main__":
+	X, Y = get_data()
+	plt.scatter(X[:,0], X[:,1], s=100, c=Y, alpha=0.5)
+	plt.show()
+
+	model = KNN(3)
+	model.fit(X, Y)
+	print("Train accuracy:", model.score(X, Y))
\ No newline at end of file

knn7.py

+import numpy as np
+from sortedcontainers import SortedList
+from util import get_data
+from datetime import datetime
+
+class KNN(object):
+	def __init__(self, k):
+		self.k = k
+
+	def fit(self, x, y):
+		self.x = x
+		self.y = y
+
+	def predict(self, X):
+		y = np.zeros(len(X))
+		for i, x in enumerate(X):
+			sl = SortedList()
+			for j, xt in enumerate(self.x):
+				diff = x - xt
+				d = diff.dot(diff)
+				if len(sl) < self.k:
+					sl.add((d, self.y[j]))
+				else:
+					if d < sl[-1][0]:
+						del sl[-1]
+						sl.add((d, self.y[j]))
+			votes = {}
+			for _, v in sl:
+				votes[v] = votes.get(v, 0) + 1
+			max_votes = 0
+			max_votes_class = -1
+			for v, count in votes.items():
+				if count > max_votes:
+					max_votes = count
+					max_votes_class = v
+			y[i] = max_votes_class
+		return y
+
+	def score(self, X, Y):
+		P = self.predict(X)
+		return np.mean(P == Y)
+
+if __name__ == "__main__":
+	X, Y = get_data(2000)
+	Ntrain = 1000
+	Xtrain, Ytrain = X[:Ntrain], Y[:Ntrain]
+	Xtest, Ytest = X[Ntrain:], Y[Ntrain:]
+	for k in (1,2,3,4,5):
+		knn = KNN(k)
+		print("k =", k)
+		t0 = datetime.now()
+		knn.fit(Xtrain, Ytrain)
+		print("Training time:", (datetime.now() - t0))
+
+		t0 = datetime.now()
+		print("Train accuracy:", (knn.score(Xtrain, Ytrain)))
+		print("Time to compute train accuracy:", (datetime.now() - t0))
+
+		t0 = datetime.now()
+		print("Test accuracy:", (knn.score(Xtest, Ytest)))
+		print("Time to compute test accuracy:", (datetime.now() - t0))
\ No newline at end of file

knn_donut_10.py

+from util import get_donut
+from knn_7 import KNN
+import matplotlib.pyplot as plt
+
+if __name__ == "__main__":
+	X, Y = get_donut()
+	plt.scatter(X[:,0], X[:,1], c=Y)
+	plt.show()
+
+	model = KNN(3)
+	model.fit(X, Y)
+	print("Accuracy:", model.score(X, Y))

knn_xor_9.py

+from knn_7 import KNN
+from util import get_xor
+import matplotlib.pyplot as plt
+
+if __name__ == "__main__":
+	X, Y = get_xor()
+	plt.scatter(X[:,0], X[:,1], c=Y)
+	plt.show()
+
+	model = KNN(3)
+	model.fit(X, Y)
+	print("Accuracy:", model.score(X, Y))
\ No newline at end of file

naive_bayes_14.py

+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+
+df = pd.read_csv("../files/mnist_train.csv", header=0, nrows=2000)
+Y = df["label"].values
+X = df.loc[:, df.columns != 'label'].values
+Xmeans = X.mean(axis=0)
+
+print(np.arange(Xmeans))
+
+# plt.scatter(np.linspace(Xmeans), c=Y)
+# plt.show()
\ No newline at end of file

util.py

+import pandas as pd
+import numpy as np
+
+def get_data(limit=None):
+    print("Reading in and transforming data...")
+    df = pd.read_csv('../files/mnist_train.csv')
+    data = df.values
+    np.random.shuffle(data)
+    X = data[:, 1:] / 255.0 # data is from 0..255
+    Y = data[:, 0]
+    if limit is not None:
+        X, Y = X[:limit], Y[:limit]
+    return X, Y
+
+def get_xor():
+	X = np.zeros((200, 2))
+	X[:50] = np.random.random((50, 2)) / 2 + 0.5
+	X[50:100] = np.random.random((50, 2)) / 2
+	X[100:150] = np.random.random((50, 2)) / 2 + np.array([[0, 0.5]])
+	X[150:] = np.random.random((50, 2)) / 2 + np.array([[0.5, 0]])
+	Y = np.array([0]*100 + [1]*100)
+	return X, Y
+
+def get_donut():
+	N = 200
+	R_inner = 5
+	R_outer = 10
+
+	R1 = np.random.randn(N//2) + R_inner
+	theta = 2*np.pi*np.random.random(N//2)
+	X_inner = np.concatenate([[R1 * np.cos(theta)], [R1 * np.sin(theta)]]).T
+
+	R2 = np.random.randn(N//2) + R_outer
+	theta = 2*np.pi*np.random.random(N//2)
+	X_outer = np.concatenate([[R2 * np.cos(theta)], [R2 * np.sin(theta)]]).T
+
+	X = np.concatenate([X_inner, X_outer])
+	Y = np.array([0]*(N//2) + [1]*(N//2))
+	return X, Y
\ No newline at end of file