plt.scatter(X[:, f1], X[:, f2])
plt.xlabel("$x_{%d}$" % f1)
plt.ylabel("$x_{%d}$" % f2)
for i in range(n):
plt.annotate(animals[i], (X[i, f1], X[i, f2]))
utils.savefig("two_random_features.png")
elif question == "1.3":
dataset = load_dataset("animals.pkl")
X = dataset["X"].astype(float)
animals = dataset["animals"]
n, d = X.shape
model = MDS(n_components=2)
Z = model.compress(X)
fig, ax = plt.subplots()
ax.scatter(Z[:, 0], Z[:, 1])
plt.ylabel("z2")
plt.xlabel("z1")
plt.title("MDS")
for i in range(n):
ax.annotate(animals[i], (Z[i, 0], Z[i, 1]))
utils.savefig("MDS_animals.png")
elif question == "1.4":
dataset = load_dataset("animals.pkl")
X = dataset["X"].astype(float)
animals = dataset["animals"]
ax[1, 1].imshow(Xhat_pca[i].reshape(h, w).T, cmap='gray')
ax[1, 2].set_title('$|x_i-\hat{x_i}|$>threshold (L1)')
ax[1, 2].imshow(
(np.abs(X[i] - Xhat_pca[i]) < threshold).reshape(h, w).T,
cmap='gray')
utils.savefig('highway_{:03d}.jpg'.format(i))
elif question == '4':
dataset = load_dataset('animals.pkl')
X = dataset['X'].astype(float)
animals = dataset['animals']
n, d = X.shape
model = MDS(n_components=2)
Z = model.compress(X)
fig, ax = plt.subplots()
ax.scatter(Z[:, 0], Z[:, 1])
plt.ylabel('z2')
plt.xlabel('z1')
plt.title('MDS')
for i in range(n):
ax.annotate(animals[i], (Z[i, 0], Z[i, 1]))
utils.savefig('MDS_animals.png')
elif question == '4.1':
dataset = load_dataset('animals.pkl')
X = dataset['X'].astype(float)
animals = dataset['animals']
plt.scatter(X[:, f1], X[:, f2])
plt.xlabel("$x_{%d}$" % f1)
plt.ylabel("$x_{%d}$" % f2)
for i in range(n):
plt.annotate(animals[i], (X[i, f1], X[i, f2]))
utils.savefig('two_random_features.png')
elif question == '1.3':
dataset = load_dataset('animals.pkl')
X = dataset['X'].astype(float)
animals = dataset['animals']
n, d = X.shape
model = MDS(n_components=2)
Z = model.compress(X)
print(
utils.euclidean_dist_squared(Z, Z).shape,
utils.euclidean_dist_squared(X, X).shape)
print(
np.sqrt(utils.euclidean_dist_squared(Z, Z)) -
np.sqrt(utils.euclidean_dist_squared(X, X)))
# diff = np.sum(np.square(np.sqrt(utils.euclidean_dist_squared(Z, Z)) - np.sqrt(utils.euclidean_dist_squared(X, X))))
# print(diff)
fig, ax = plt.subplots()
ax.scatter(Z[:, 0], Z[:, 1])
plt.ylabel('z2')
plt.xlabel('z1')
plt.title('MDS')