def get_chipscores(hsA, hsB, count2rr_AB):
'''
Input: Two database handles and queries from A to B
Output: Matrix of chip scores. As well as
'''
num_results = -1 # ensure there are N top results
num_queries = len(count2rr_AB)
# Get top scores of result
chipscore_list = [[] for _ in xrange(num_queries)]
# Indicator as to if chipscore was a true positive
isscoreTP_list = [[] for _ in xrange(num_queries)]
for count in xrange(num_queries):
rr = count2rr_AB[count]
res = QueryResult(hsB, rr, hsA)
res.force_num_top(-1)
top_scores = res.top_scores()
gtpos_full = res.get_groundtruth_ranks()
gtpos_full = [] if gtpos_full is None else gtpos_full
gtpos_list = []
for gt_pos in gtpos_full:
if num_results == -1 or gt_pos < num_results:
gtpos_list.append(gt_pos)
isscoreTP_list[count] = gtpos_list
chipscore_list[count] = top_scores
num_results = len(chipscore_list[0])
assert np.all(num_results == np.array([len(_) for _ in chipscore_list])), \
'There must be the same number of results'
chipscore_data = np.array(chipscore_list)
ischipscore_TP = np.zeros((num_queries, num_results), dtype=np.bool)
for count in xrange(num_queries):
gtpos_list = [isscoreTP_list[count]]
ischipscore_TP[count][gtpos_list] = 1
return chipscore_data, ischipscore_TP
def viz_threshold_matchings(hsA, hsB, count2rr_AB, thresh_out_dir):
'returns database, cx, database cx'
import numpy as np
valid_cxsB = hsB.cm.get_valid_cxs()
num_matching = 0
MATCH_THRESHOLD = __method_2_matchthresh__.get(__METHOD__, 10)
results_name = get_results_name(hsA, hsB)
print(' * Visualizing threshold matchings ' + results_name +
' give it some time to plot...')
threshdb_out_dir = join_mkdir(thresh_out_dir, results_name)
# For each query run in hsA vs hsB
for count in xrange(len(count2rr_AB)):
rr = count2rr_AB[count]
qcx = rr.qcx
res = QueryResult(hsB, rr, hsA)
qname = res.qhs.cm.cx2_name(qcx)
# Set matching threshold
res.top_thresh = MATCH_THRESHOLD
res.num_top_min = 0
res.num_top_max = 5
res.num_extra_return = 0
# Check to see if any matched over the threshold
top_cxs = res.top_cx()
top_names = res.hs.cm.cx2_name(top_cxs)
top_scores = res.scores()[top_cxs]
if len(top_cxs) > 0:
# Visualize the result
# Stupid segfaults #if qcx == 113: #41: #import IPython #IPython.embed() # Get the scores
num_matching += 1
res.visualize()
# Create a filename showing dataset, score, cx, match names
matchname_set = \
set([name.replace('Lionfish','') for name in (top_names+[qname])])
matchnames = '-'.join(list(matchname_set))
scorestr = str(int(round(top_scores[0])))
fig_fname = '-'.join([
results_name, 'score' + scorestr, 'cx' + str(qcx),
'MATCHES' + matchnames
]) + '.jpg'
fig = myfigure(0)
fig_fpath = join(threshdb_out_dir, fig_fname)
sys.stdout.write('.')
safe_savefig(fig, fig_fpath)
print(' * Visualized %d above thresh: %f from expt: %s ' %
(num_matching, MATCH_THRESHOLD, results_name))
def write_all_chipscore_results(hsA, hsB, count2rr_AB, txt_output_fpath):
txt_header = '''#<COMMENTED_XML_HEADER>
# <filename>
# ''' + txt_output_fpath + '''
# </filename>
#
# <description>
# Sorted Query Results. The results are stored in the format:
# The first line gives the query image name, the next lines are
# a sorted list of result image names and scores. Queries are
# separated by two new lines.
# </description>
#
# <data_format>
# queryA
# resultA1 scoreA1
# resultA2 scoreA2
# resultA3 scoreA3
# ...
#
# queryB-gname
# resultB1 scoreB1
# resultB2 scoreB2
# resultB3 scoreB3
# ...
#
# ...
# </data_format>
#</COMMENTED_XML_HEADER>
'''
txt_body = ''
for count in range(len(count2rr_AB)):
rr = count2rr_AB[count]
res = QueryResult(hsB, rr, hsA)
txt_body += hsA.cm.cx2_gname(rr.qcx) + '\n'
top_cxs = res.cx_sort()
scores = res.scores()
output_tup_list = []
for cx in top_cxs:
output_tup_list.append((hsB.cm.cx2_gname(cx), scores[cx]))
output_tup_list.sort(key=lambda tup: tup[1])
output_tup_list = output_tup_list[::-1]
for tup in output_tup_list:
txt_body += tup[0] + ' ' + str(tup[1]) + '\n'
txt_body += '\n'
with open(txt_output_fpath, 'w') as file:
file.write(txt_body)
def write_all_chipscore_results(hsA, hsB, count2rr_AB, txt_output_fpath):
txt_header = '''#<COMMENTED_XML_HEADER>
# <filename>
# '''+txt_output_fpath+'''
# </filename>
#
# <description>
# Sorted Query Results. The results are stored in the format:
# The first line gives the query image name, the next lines are
# a sorted list of result image names and scores. Queries are
# separated by two new lines.
# </description>
#
# <data_format>
# queryA
# resultA1 scoreA1
# resultA2 scoreA2
# resultA3 scoreA3
# ...
#
# queryB-gname
# resultB1 scoreB1
# resultB2 scoreB2
# resultB3 scoreB3
# ...
#
# ...
# </data_format>
#</COMMENTED_XML_HEADER>
'''
txt_body = ''
for count in range(len(count2rr_AB)):
rr = count2rr_AB[count]
res = QueryResult(hsB, rr, hsA)
txt_body += hsA.cm.cx2_gname(rr.qcx)+'\n'
top_cxs = res.cx_sort()
scores = res.scores()
output_tup_list = []
for cx in top_cxs:
output_tup_list.append((hsB.cm.cx2_gname(cx), scores[cx]))
output_tup_list.sort(key=lambda tup: tup[1])
output_tup_list = output_tup_list[::-1]
for tup in output_tup_list:
txt_body += tup[0]+' '+str(tup[1])+'\n'
txt_body+='\n'
with open(txt_output_fpath,'w') as file:
file.write(txt_body)
def run_all_queries(em):
cm, vm = em.hs.get_managers('cm', 'vm')
all_cxs = cm.get_valid_cxs()
cx2_rr = em.batch_query(force_recomp=em.recompute_bit,
test_cxs=all_cxs)
em.cx2_res = np.array([ [] if rr == [] else\
QueryResult(em.hs,rr) for rr in cx2_rr])
pass
def query(hs, qcid, qhs=None):
'''
Runs query against this database with cid.
If hsother is specified uses the chip from that database
'''
qhs = hs if qhs is None else qhs
qcx = qhs.cm.cx(qcid)
hs.ensure_model()
rawres = hs.qm.cx2_rr(qcx, qhs)
return QueryResult(hs, rawres, qhs)
def viz_threshold_matchings(hsA, hsB, count2rr_AB, thresh_out_dir):
'returns database, cx, database cx'
import numpy as np
valid_cxsB = hsB.cm.get_valid_cxs()
num_matching = 0
MATCH_THRESHOLD = __method_2_matchthresh__.get(__METHOD__, 10)
results_name = get_results_name(hsA, hsB)
print(' * Visualizing threshold matchings '+results_name+' give it some time to plot...')
threshdb_out_dir = join_mkdir(thresh_out_dir, results_name)
# For each query run in hsA vs hsB
for count in xrange(len(count2rr_AB)):
rr = count2rr_AB[count]
qcx = rr.qcx
res = QueryResult(hsB, rr, hsA)
qname = res.qhs.cm.cx2_name(qcx)
# Set matching threshold
res.top_thresh = MATCH_THRESHOLD
res.num_top_min = 0
res.num_top_max = 5
res.num_extra_return = 0
# Check to see if any matched over the threshold
top_cxs = res.top_cx()
top_names = res.hs.cm.cx2_name(top_cxs)
top_scores = res.scores()[top_cxs]
if len(top_cxs) > 0:
# Visualize the result
# Stupid segfaults #if qcx == 113: #41: #import IPython #IPython.embed() # Get the scores
num_matching += 1
res.visualize()
# Create a filename showing dataset, score, cx, match names
matchname_set = \
set([name.replace('Lionfish','') for name in (top_names+[qname])])
matchnames = '-'.join(list(matchname_set))
scorestr = str(int(round(top_scores[0])))
fig_fname = '-'.join([results_name,
'score'+scorestr,
'cx'+str(qcx),
'MATCHES'+matchnames])+'.jpg'
fig = myfigure(0)
fig_fpath = join(threshdb_out_dir, fig_fname)
sys.stdout.write('.')
safe_savefig(fig, fig_fpath)
print(' * Visualized %d above thresh: %f from expt: %s ' % (num_matching,
MATCH_THRESHOLD,
results_name))
def get_chipscores(hsA, hsB, count2rr_AB):
'''
Input: Two database handles and queries from A to B
Output: Matrix of chip scores. As well as
'''
num_results = -1 # ensure there are N top results
num_queries = len(count2rr_AB)
# Get top scores of result
chipscore_list = [[] for _ in xrange(num_queries)]
# Indicator as to if chipscore was a true positive
isscoreTP_list = [[] for _ in xrange(num_queries)]
for count in xrange(num_queries):
rr = count2rr_AB[count]
res = QueryResult(hsB, rr, hsA)
res.force_num_top(-1)
top_scores = res.top_scores()
gtpos_full = res.get_groundtruth_ranks()
gtpos_full = [] if gtpos_full is None else gtpos_full
gtpos_list = []
for gt_pos in gtpos_full:
if num_results == -1 or gt_pos < num_results:
gtpos_list.append(gt_pos)
isscoreTP_list[count] = gtpos_list
chipscore_list[count] = top_scores
num_results = len(chipscore_list[0])
assert np.all(num_results == np.array([len(_) for _ in chipscore_list])), \
'There must be the same number of results'
chipscore_data = np.array(chipscore_list)
ischipscore_TP = np.zeros((num_queries, num_results), dtype=np.bool)
for count in xrange(num_queries):
gtpos_list = [isscoreTP_list[count]]
ischipscore_TP[count][gtpos_list] = 1
return chipscore_data, ischipscore_TP
def run_singleton_queries(em):
'''Quick experiment:
Query each chip with a duplicate against whole database
Do not remove anyone from ANN matching'''
cm, vm = em.hs.get_managers('cm', 'vm')
valid_cx = cm.get_valid_cxs()
cx2_num_other = cm.cx2_num_other_chips(valid_cx)
singleton_cx = valid_cx[cx2_num_other == 1] # find singletons
duplicate_cx = valid_cx[cx2_num_other > 1] # find matchables
cx2_rr = em.batch_query(force_recomp=em.recompute_bit,
test_cxs=duplicate_cx)
em.cx2_res = np.array([ [] if rr == [] else\
QueryResult(em.hs,rr) for rr in cx2_rr])
def run_matching_experiment(em,
expt_name='MatchingExperiment',
with_images=True):
'''Quick experiment:
Query each chip with a duplicate against whole database
Do not remove anyone from ANN matching'''
import os
logmsg('Running Quick Experiment: ' + expt_name)
hs = em.hs
am, cm, vm, qm, iom, dm = hs.get_managers('am', 'cm', 'vm', 'qm',
'iom', 'dm')
with_ellipses = True
with_points = False
with_full_results = False
with_ellipse_and_points = False
# Create an Experiment Directory in .hs_internals/computed
timestamp = iom.get_timestamp()
expt_dpath = iom.ensure_directory(
iom.get_user_fpath('_'.join(['expt', expt_name, timestamp])))
expt_img_dpath = iom.ensure_directory(
os.path.join(expt_dpath, 'result_images'))
expt_rr_dpath = iom.ensure_directory(
os.path.join(expt_dpath, 'raw_results'))
# Write Algorithm Settings to the file
algo_prefs_text = am.get_algo_name('all')
algo_prefs_fpath = os.path.join(expt_dpath, 'algo_prefs.txt')
iom.write(algo_prefs_fpath, algo_prefs_text)
vm.build_model() # Defaults to building model of all
prev_ell = dm.draw_prefs.ellipse_bit
prev_pts = dm.draw_prefs.points_bit
dm.fignum = 1
# Create Matches File To Append to
with open(os.path.join(expt_dpath, 'matches.txt'), 'a') as file:
for cx in iter(cm.get_valid_cxs()):
# Preform Query
res = QueryResult(qm.hs, qm.cx2_rr(cx))
# Get Query Info
qcid, gname = cm.cx2_(res.rr.qcx, 'cid', 'gname')
# Get Result Info
(tcid, tgname, tscore) = res.tcid2_('cid', 'gname', 'score')
# Print Query Info
logmsg('---QUERY---')
outstr = 'QUERY, gname=%s, cid=%4d' % (gname, qcid)
print outstr
file.write(outstr + '\n')
# Print Result Info
if len(tscore) == 0:
maxsim = 0 # Best Score
if len(tscore) > 0:
maxsim = tscore[0] # Best Score
for (rank, tup) in enumerate(
zip(*[x.tolist() for x in (tgname, tcid, tscore)])):
outstr = ' rank=%d, gname=%s, cid=%4d, score=%7.2f' % tuple(
[rank + 1] + list(tup))
print outstr
file.write(outstr + '\n')
print ''
file.write('\n\n')
# Output Images
query_name = 'sim=%07.2f-qcid=%d.png' % (maxsim, qcid)
if with_full_results:
#import shelve
#shelve.open(os.path.join(expt_rr_dpath, 'rr_'+query_name+'.shelf')
#import cPickle
#rr = res.rr
#cPickle.dump(, rr)
pass
if with_images:
dm.draw_prefs.ellipse_bit = False
dm.draw_prefs.points_bit = False
dm.draw_prefs.bbox_bit = False
dm.show_query(res)
dm.save_fig(
os.path.join(expt_img_dpath, 'img_' + query_name))
if with_ellipses:
dm.draw_prefs.ellipse_bit = True
dm.draw_prefs.points_bit = False
dm.show_query(res)
dm.save_fig(
os.path.join(expt_img_dpath,
'ellipse_' + query_name))
if with_points:
dm.draw_prefs.ellipse_bit = False
dm.draw_prefs.points_bit = True
dm.show_query(res)
dm.save_fig(
os.path.join(expt_img_dpath,
'point_' + query_name))
if with_ellipse_and_points:
dm.draw_prefs.ellipse_bit = True
dm.draw_prefs.points_bit = True
dm.show_query(res)
dm.save_fig(
os.path.join(expt_img_dpath, 'both_' + query_name))
logmsg('Finished Matching Experiment: ' + expt_name)
timestamp = iom.get_timestamp()
iom.write(os.path.join(expt_dpath, 'Finished_' + str(timestamp)),
timestamp + '\n' + em.hs.get_database_stat_str())
prev_ell = dm.draw_prefs.ellipse_bit
prev_pts = dm.draw_prefs.points_bit
def run_matching_experiment(em, expt_name='MatchingExperiment',
with_images=True):
'''Quick experiment:
Query each chip with a duplicate against whole database
Do not remove anyone from ANN matching'''
import os
logmsg('Running Quick Experiment: '+expt_name)
hs = em.hs
am, cm, vm, qm, iom, dm = hs.get_managers('am', 'cm','vm', 'qm', 'iom', 'dm')
with_ellipses = True
with_points = False
with_full_results = False
with_ellipse_and_points = False
# Create an Experiment Directory in .hs_internals/computed
timestamp = iom.get_timestamp()
expt_dpath = iom.ensure_directory(iom.get_user_fpath('_'.join(['expt',expt_name,timestamp])))
expt_img_dpath = iom.ensure_directory(os.path.join(expt_dpath, 'result_images'))
expt_rr_dpath = iom.ensure_directory(os.path.join(expt_dpath, 'raw_results'))
# Write Algorithm Settings to the file
algo_prefs_text = am.get_algo_name('all')
algo_prefs_fpath = os.path.join(expt_dpath, 'algo_prefs.txt')
iom.write(algo_prefs_fpath, algo_prefs_text)
vm.build_model() # Defaults to building model of all
prev_ell = dm.draw_prefs.ellipse_bit
prev_pts = dm.draw_prefs.points_bit
dm.fignum = 1
# Create Matches File To Append to
with open(os.path.join(expt_dpath,'matches.txt'), 'a') as file:
for cx in iter(cm.get_valid_cxs()):
# Preform Query
res = QueryResult(qm.hs, qm.cx2_rr(cx))
# Get Query Info
qcid, gname = cm.cx2_(res.rr.qcx, 'cid', 'gname')
# Get Result Info
(tcid , tgname , tscore ) = res.tcid2_('cid','gname','score')
# Print Query Info
logmsg('---QUERY---')
outstr = 'QUERY, gname=%s, cid=%4d' % (gname, qcid)
print outstr
file.write(outstr+'\n')
# Print Result Info
if len(tscore) == 0:
maxsim = 0 # Best Score
if len(tscore) > 0:
maxsim = tscore[0] # Best Score
for (rank, tup) in enumerate(zip(*[x.tolist() for x in (tgname, tcid, tscore )])):
outstr = ' rank=%d, gname=%s, cid=%4d, score=%7.2f' % tuple([rank+1]+list(tup))
print outstr
file.write(outstr+'\n')
print ''
file.write('\n\n')
# Output Images
query_name = 'sim=%07.2f-qcid=%d.png' % (maxsim, qcid)
if with_full_results:
#import shelve
#shelve.open(os.path.join(expt_rr_dpath, 'rr_'+query_name+'.shelf')
#import cPickle
#rr = res.rr
#cPickle.dump(, rr)
pass
if with_images:
dm.draw_prefs.ellipse_bit = False
dm.draw_prefs.points_bit = False
dm.draw_prefs.bbox_bit = False
dm.show_query(res)
dm.save_fig(os.path.join(expt_img_dpath, 'img_'+query_name))
if with_ellipses:
dm.draw_prefs.ellipse_bit = True
dm.draw_prefs.points_bit = False
dm.show_query(res)
dm.save_fig(os.path.join(expt_img_dpath, 'ellipse_'+query_name))
if with_points:
dm.draw_prefs.ellipse_bit = False
dm.draw_prefs.points_bit = True
dm.show_query(res)
dm.save_fig(os.path.join(expt_img_dpath, 'point_'+query_name))
if with_ellipse_and_points:
dm.draw_prefs.ellipse_bit = True
dm.draw_prefs.points_bit = True
dm.show_query(res)
dm.save_fig(os.path.join(expt_img_dpath, 'both_'+query_name))
logmsg('Finished Matching Experiment: '+expt_name)
timestamp = iom.get_timestamp()
iom.write(os.path.join(expt_dpath, 'Finished_'+str(timestamp)), timestamp+'\n'+em.hs.get_database_stat_str())
prev_ell = dm.draw_prefs.ellipse_bit
prev_pts = dm.draw_prefs.points_bit