def test():
test_data, test_labels = load_data(test_data_path)
train_data, train_labels = load_data(train_data_path)
if args.alg == 'pca':
saved_data = np.load(
os.path.join(args.output, args.alg + str(args.dim) + '.npz'))
mean_vecs = saved_data['mean_vecs']
eig_vecs = saved_data['eig_vecs']
reduction_test_data = pca_test(np.array(test_data), mean_vecs,
eig_vecs)
reduction_train_data = pca_test(np.array(train_data), mean_vecs,
eig_vecs)
elif args.alg == 'svd':
saved_data = np.load(
os.path.join(args.output, args.alg + str(args.dim) + '.npz'))
v = saved_data['v']
mean_vecs = saved_data['mean_vecs']
reduction_test_data = svd_test(np.array(test_data), mean_vecs, v)
reduction_train_data = svd_test(np.array(train_data), mean_vecs, v)
else:
reduction_test_data = isomap(np.array(test_data), n=args.dim)
reduction_train_data = isomap(np.array(train_data), n=args.dim)
acc = eval(reduction_test_data, test_labels, reduction_train_data,
train_labels)
print('ACC = ' + str(acc))
def callback(iters, current, env):
print 'Callback: ', iters
# get all traces
env.updategui = False
traces = env.evaluate_policy(current)
env.updategui = True
pickle.dump(traces, open('traces{0}.pck'.format(iters), 'w'),
pickle.HIGHEST_PROTOCOL)
# measure task performance
avg_reward = 0
for t in traces:
avg_reward += sum([i[2] for i in t])
avg_reward = avg_reward / float(len(traces))
print 'Avg reward: ', avg_reward
# find current embedding
ematrix = np.zeros((512, 512))
for (i, t) in enumerate(traces):
for (j, s) in enumerate(traces):
ematrix[i,
j] = edit_distance_vc([e[1] for e in t], [l[1] for l in s],
(1.0, 1.0, 1.2))
pickle.dump(ematrix, open('ematrix{0}.pck'.format(iters), 'w'),
pickle.HIGHEST_PROTOCOL)
y, s, adj = isomap(ematrix)
if len(y) < 512:
# fallback to mds if more than 1 connected component
print "More than 1 CC - Falling back to MDS"
y, s = mds(ematrix)
adj = None
# plot stuff later because of pylab / pygame incompat on mac
# save embedding image - multiple formats?
#scatter(y[:,0],y[:,1], filename='scatter{0}.pdf'.format(iters))
# save scree plot
#plot(s[:10], filename='scree{0}.pdf'.format(iters))
# procrustes error
gt = env.coords_array()
err = procrustes(gt, y)
print "Procrustes ", err
pickle.dump((iters, err, gt, avg_reward, current, y, s, adj),
open('misc{0}.pck'.format(iters), 'w'),
pickle.HIGHEST_PROTOCOL)
env.save('iter{0}.png'.format(iters))
pickle.dump((iters, err, gt, avg_reward, current, y, s, adj),
open('misc{0}.pck'.format(iters), 'w'),
pickle.HIGHEST_PROTOCOL)
env.save('iter{0}.png'.format(iters))
if __name__ == '__main__':
workspace = "{0}/wrk/lspi/bin".format(os.environ['HOME'])
if False:
# fix isomap issue on bad matches
ematrix = pickle.load(open('ematrix0.pck'))
y, s, adj = isomap(ematrix)
if True:
endstates = [272] # [16]
ogw = RBFObserverGridworldGui(
"/Users/stober/wrk/lspi/bin/16/20comp.npy",
"/Users/stober/wrk/lspi/bin/16/states.npy",
endstates=endstates,
walls=None,
nrbf=80)
# ogw = RBFObserverGridworld("/Users/stober/wrk/lspi/bin/16/20comp.npy", "/Users/stober/wrk/lspi/bin/16/states.npy", endstates = endstates, walls=None, nrbf=80)
t = pickle.load(open(workspace + "/traces/complete_trace.pck"))
# -*- coding: utf-8 -*-
from sklearn.manifold import Isomap
import isomap as isa
import sklearn.datasets as db
def isomap(data, d, k):
'''
input:data(ndarray):待降维数据
d(int):降维后数据维度
k(int):最近的k个样本
output:Z(ndarray):降维后数据
'''
# ********* Begin *********#
iso = Isomap(n_neighbors=k, n_components=d)
return iso.fit_transform(data)
if __name__ == '__main__':
ir = db.load_boston()
X1 = isa.isomap(ir.data[:10], d=2, k=4)
X2 = isomap(ir.data[:10], d=2, k=4)
print(X1)
print(X2)
#import lle
#print shape(iris)
#a,b,newData = lle.lle(iris,2,12)
#print shape(newData)
#print newData[w0,:]
#print "---"
#print newData[w1,:]
#print "---"
#print newData[w2,:]
#
#plot(newData[w0,0],newData[w0,1],'ok')
#plot(newData[w1,0],newData[w1,1],'^k')
#plot(newData[w2,0],newData[w2,1],'vk')
#axis('off')
import isomap
print labels
newData,newLabels = isomap.isomap(iris,2,100)
print shape(newData)
print newLabels
w0 = where(newLabels==0)
w1 = where(newLabels==1)
w2 = where(newLabels==2)
plot(newData[w0,0],newData[w0,1],'ok')
plot(newData[w1,0],newData[w1,1],'^k')
plot(newData[w2,0],newData[w2,1],'vk')
axis('off')
show()
print np.shape(iris)
a,b,newData = lle.lle(iris,2,12)
print np.shape(newData)
print newData[w0,:]
print "---"
print newData[w1,:]
print "---"
print newData[w2,:]
pl.plot(newData[w0,0],newData[w0,1],'ok')
pl.plot(newData[w1,0],newData[w1,1],'^k')
pl.plot(newData[w2,0],newData[w2,1],'vk')
pl.axis('off')
import isomap
print labels
newData,newLabels = isomap.isomap(iris,2,100)
print np.shape(newData)
print newLabels
w0 = np.where(newLabels==0)
w1 = np.where(newLabels==1)
w2 = np.where(newLabels==2)
pl.plot(newData[w0,0],newData[w0,1],'ok')
pl.plot(newData[w1,0],newData[w1,1],'^k')
pl.plot(newData[w2,0],newData[w2,1],'vk')
pl.axis('off')
print "Done"
pl.show()
iy = 10 * j + 1
img[ix:ix + 8, iy:iy + 8] = X[i * n_img_per_row + j].reshape((8, 8))
plt.imshow(img, cmap=plt.cm.binary)
plt.xticks([])
plt.yticks([])
plt.title('A selection from the 64-dimensional digits dataset')
# Random 2D projection using a random unitary matrix
print("Computing random projection")
rp = random_projection.SparseRandomProjection(n_components=2, random_state=42)
X_projected = rp.fit_transform(X)
plot_embedding(X_projected, "Random Projection of the digits")
# Isomap projection of the digits dataset
print("Computing Isomap embedding")
t0 = time()
embedding = isomap(input=X, n_neighbors=n_neighbors, n_components=2, n_jobs=4)
print("Done.")
plot_embedding(embedding,"Isomap projection of the digits (time %.2fs)" %(time() - t0))
embedding_three_dim = isomap(input=X, n_neighbors=n_neighbors, n_components=3, n_jobs=4)
# draw 32 plots
fig = plt.figure()
ax = Axes3D(fig)
ax.scatter(embedding_three_dim[:, 0], embedding_three_dim[:, 1], embedding_three_dim[:, 2], c=y)