img = np.zeros((nx, nz), dtype=int)
img.fill(5)
#air
for ind, (x, z) in enumerate(locations):
i = int((x - 14700) // 40)
k = int((z - 2200) // 20)
img[i, k] = 4
for c, cluster in enumerate([kmeans, pca, fcew, sfcew, fc, sfc]):
fig = plt.figure(figsize=(20, 10))
ax = fig.add_subplot(111, aspect='equal')
#ore
new_clusters = adjust_clusters(cluster, equivalences[c])
for ind, (x, z) in enumerate(locations_ore):
i = int((x - 14700) // 40)
k = int((z - 2200) // 20)
img[i, k] = new_clusters[ind]
new_img = img[20:80, 10:]
nx2, nz2 = new_img.shape
ax.set_xlim(0, nx2 * sx)
ax.set_ylim(0, nz2 * sz)
for i in range(nx2):
for k in range(nz2):
equivalences[1] = {0:3, 1:1, 2:2, 3:0, 4:4}
equivalences[2] = {0:2, 1:0, 2:1, 3:3, 4:4}
equivalences[3] = {0:2, 1:0, 2:1, 3:3, 4:4}
equivalences[4] = {0:1, 1:0, 2:2, 3:3, 4:4}
equivalences[5] = {0:1, 1:0, 2:2, 3:3, 4:4}
clay_color = ['r','b','g']
i = 5
cluster = sfc
if i in equivalences:
adjust_cluster = adjust_clusters(cluster,equivalences[i])
else:
adjust_cluster = cluster
fignames = ['fig7a','fig7b','fig7c','fig7d']
print("ND",ND,attributes)
for var in range(ND-1):
fig, ax = plt.subplots(figsize=(6, 6))
#ax.set_ylim([mins[v],maxs[v]])
if var_types[var] == 3:
counter = Counter(data[:,var])
mostc = counter.most_common()
cats = len(mostc)
for j in range(NC):
if j in equivalence:
c = equivalence[j]
else:
c = j
cluster = np.where(clusters[true_c] == c)[0]
cm[i, j] = len(cluster)
return cm
classes = ['C1', 'C2', 'C3', 'C4']
#KMeans
i = 1
if i in equivalences:
adjust_cluster = adjust_clusters(kmeans_clusters, equivalences[i])
else:
adjust_cluster = kmeans_clusters
cm_kmeans = build_cm(NC, true_clusters, adjust_cluster)
fig, ax = plt.subplots()
plot_confusion_matrix(cm_kmeans,
classes,
normalize=True,
title='Confusion matrix K-Means',
cmap=plt.cm.Blues,
xlabel="Predicted",
ylabel="True")
plt.savefig("../figures/confusion-matrix-kmeans-ds4")