jketreno/ketr.photos
1
0
Fork 0
Code Issues 1 Activity

Improved stats output during clustering

Browse Source 
Signed-off-by: James Ketrenos <james_git@ketrenos.com>
This commit is contained in:
James Ketr 2023-01-16 19:55:13 -08:00
parent e8de846ed0
commit 46b2075515
1 changed files with 22 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
ketrface/ketrface/dbscan.py
View File

@ -1,4 +1,5 @@
from ketrface.util import * from ketrface.util import *
import time
MIN_PTS = 5 MIN_PTS = 5
MAX_DISTANCE = 0.25 MAX_DISTANCE = 0.25
@ -9,26 +10,32 @@ Noise = -2
# Union of two lists of dicts # Union of two lists of dicts
def Union(A, B): def Union(A, B):
C = [] C = A
for key in A + B: for key in B:
if key not in C: if key not in C:
C.append(key) C.append(key)
return C return C
# https://en.wikipedia.org/wiki/DBSCAN # https://en.wikipedia.org/wiki/DBSCAN
#
def DBSCAN(points, eps = MAX_DISTANCE, minPts = MIN_PTS, verbose = True): def DBSCAN(points, eps = MAX_DISTANCE, minPts = MIN_PTS, verbose = True):
clusters = [] # Cluster list clusters = [] # Cluster list
perc = -1 perc = -1
total = len(points) total = len(points)
last = 0
start = time.time()
for i, P in enumerate(points): for i, P in enumerate(points):
if verbose == True: if verbose == True:
new_perc = int(100 * (i+1) / total) new_perc = int(100 * (i+1) / total)
if new_perc != perc: now = time.time()
if new_perc != perc or now - last > 5:
perc = new_perc perc = new_perc
print(f'Clustering points {perc}% ({i}/{total} processed) complete with {len(clusters)} identities.') print(f'Clustering points {perc}% ({i}/{total} processed) complete with {len(clusters)} identities ({now - start}s).')
last = now
if P['cluster'] != Undefined: # Previously processed in inner loop if P['cluster'] != Undefined: # Previously processed in inner loop
continue continue
N = RangeQuery(points, P, eps) # Find neighbors N = RangeQuery(points, P, eps) # Find neighbors
if len(N) < minPts: # Density check if len(N) < minPts: # Density check
P['cluster'] = Noise # Label as Noise P['cluster'] = Noise # Label as Noise
@ -45,7 +52,16 @@ def DBSCAN(points, eps = MAX_DISTANCE, minPts = MIN_PTS, verbose = True):
S = N # Neighbors to expand (exclude P) S = N # Neighbors to expand (exclude P)
S.remove(P) S.remove(P)
for Q in S: # Process every seed point sub_perc = -1
sub_total = len(S)
for j, Q in enumerate(S): # Process every seed point
if verbose == True:
sub_new_perc = int(100 * (j+1) / total)
now = time.time()
if sub_new_perc != perc or now - last > 5:
sub_perc = sub_new_perc
print(f'... points {sub_perc}% ({j}/{sub_total} processed) complete with {len(clusters)} identities ({now - start}s).')
last = now
if Q['cluster'] == Noise: # Change Noise to border point if Q['cluster'] == Noise: # Change Noise to border point
Q['cluster'] = C Q['cluster'] = C
C['faces'].append(Q) C['faces'].append(Q)