def demodata_infr2(defaultdb='PZ_MTEST'):
defaultdb = 'PZ_MTEST'
import ibeis
ibs = ibeis.opendb(defaultdb=defaultdb)
annots = ibs.annots()
names = list(annots.group_items(annots.nids).values())[0:20]
def dummy_phi(c, n):
x = np.arange(n)
phi = c * x / (c * x + 1)
phi = phi / phi.sum()
phi = np.diff(phi)
return phi
phis = {c: dummy_phi(c, 30) for c in range(1, 4)}
aids = ut.flatten(names)
infr = ibeis.AnnotInference(ibs, aids, autoinit=True)
infr.init_termination_criteria(phis)
infr.init_refresh_criteria()
# Partially review
n1, n2, n3, n4 = names[0:4]
for name in names[4:]:
for a, b in ut.itertwo(name.aids):
infr.add_feedback((a, b), POSTV)
for name1, name2 in it.combinations(names[4:], 2):
infr.add_feedback((name1.aids[0], name2.aids[0]), NEGTV)
return infr
def demodata_mtest_infr(state='empty'):
import ibeis
ibs = ibeis.opendb(db='PZ_MTEST')
annots = ibs.annots()
names = list(annots.group_items(annots.nids).values())
ut.shuffle(names, rng=321)
test_aids = ut.flatten(names[1::2])
infr = ibeis.AnnotInference(ibs, test_aids, autoinit=True)
infr.reset(state=state)
return infr
def make_dummy_infr(annots_per_name):
import ibeis
nids = [
val for val, num in enumerate(annots_per_name, start=1)
for _ in range(num)
]
aids = range(len(nids))
infr = ibeis.AnnotInference(None,
aids,
nids=nids,
autoinit=True,
verbose=1)
return infr
def start(actor, dbdir, aids='all', config={},
**kwargs):
import ibeis
assert dbdir is not None, 'must specify dbdir'
assert actor.infr is None, ('AnnotInference already running')
ibs = ibeis.opendb(dbdir=dbdir, use_cache=False, web=False,
force_serial=True)
# Create the AnnotInference
print('starting via actor with ibs = %r' % (ibs, ))
actor.infr = ibeis.AnnotInference(ibs=ibs, aids=aids, autoinit=True)
actor.infr.print('started via actor')
actor.infr.print('config = {}'.format(ut.repr3(config)))
# Configure query_annot_infr
for key in config:
actor.infr.params[key] = config[key]
# Initialize
# TODO: Initialize state from staging reviews after annotmatch
# timestamps (in case of crash)
actor.infr.print('Initializing infr tables')
table = kwargs.get('init', 'staging')
actor.infr.reset_feedback(table, apply=True)
actor.infr.ensure_mst()
actor.infr.apply_nondynamic_update()
actor.infr.print('infr.status() = {}'.format(ut.repr4(actor.infr.status())))
# Load random forests (TODO: should this be config specifiable?)
actor.infr.print('loading published models')
try:
actor.infr.load_published()
except:
pass
# Start actor.infr Main Loop
actor.infr.print('start id review')
actor.infr.start_id_review()
return 'initialized'
def reset_mtest_graph():
"""
Resets the annotmatch and stating table
CommandLine:
python -m ibeis reset_mtest_graph
Example:
>>> # SCRIPT
>>> from ibeis.init.sysres import * # NOQA
>>> reset_mtest_graph()
"""
if True:
# Delete the graph databases to and set them up for tests
import ibeis
ibs = ibeis.opendb('PZ_MTEST')
annotmatch = ibs.db['annotmatch']
staging = ibs.staging['reviews']
annotmatch.clear()
staging.clear()
# Make this CC connected using positive edges
from ibeis.algo.graph.state import POSTV, NEGTV, INCMP, DIFF, NULL, SAME # NOQA
from ibeis.algo.graph import nx_utils as nxu
import itertools as it
# Add some graph properties to MTEST
infr = ibeis.AnnotInference(ibs, 'all', autoinit=True)
# Connect the names with meta decisions
infr.ensure_mst(meta_decision=SAME)
# big_ccs = [cc for cc in infr.positive_components() if len(cc) > 3]
small_ccs = [
cc for cc in infr.positive_components() if len(cc) <= 3 and len(cc) > 1
]
# single_ccs = [cc for cc in infr.positive_components() if len(cc) == 1]
cc = infr.pos_graph.connected_to(1)
for edge in nxu.edges_between(infr.graph, cc):
infr.add_feedback(edge, POSTV, user_id='user:setup1')
# Make all small PCCs k-negative-redundant
count = 0
for cc1, cc2 in it.combinations(small_ccs, 2):
count += 1
for edge in infr.find_neg_augment_edges(cc1, cc2, k=1):
if count > 10:
# So some with meta
infr.add_feedback(edge,
meta_decision=DIFF,
user_id='user:setup2')
else:
# So some with evidence
infr.add_feedback(edge, NEGTV, user_id='user:setup3')
# Make some small PCCs k-positive-redundant
from ibeis.algo.graph.state import POSTV, NEGTV, INCMP, UNREV, UNKWN # NOQA
cand = list(infr.find_pos_redun_candidate_edges())
for edge in cand[0:2]:
infr.add_feedback(edge, evidence_decision=POSTV, user_id='user:setup4')
assert infr.status()['nInconsistentCCs'] == 0
# Write consistent state to both annotmatch and staging
infr.write_ibeis_staging_feedback()
infr.write_ibeis_annotmatch_feedback()
# Add an 2 inconsistencies to the staging database ONLY
cand = list(infr.find_pos_redun_candidate_edges())
for edge in cand[0:2]:
infr.add_feedback(edge,
evidence_decision=NEGTV,
user_id='user:voldemort')
assert infr.status()['nInconsistentCCs'] == 2
infr.write_ibeis_staging_feedback()
infr.reset_feedback('annotmatch', apply=True)
assert infr.status()['nInconsistentCCs'] == 0
def _precollect(self):
"""
Sets up an ibs object with an aids_pool
Example:
>>> from ibeis.scripts.thesis import *
>>> self = Chap3('humpbacks_fb')
>>> self = Chap3('GZ_Master1')
>>> self = Chap3('GIRM_Master1')
>>> self = Chap3('PZ_MTEST')
>>> self = Chap3('PZ_PB_RF_TRAIN')
>>> self = Chap3('PZ_Master1')
>>> self = Chap3('RotanTurtles')
>>> self._precollect()
>>> from ibeis.scripts.thesis import *
>>> self = Chap4('PZ_Master1')
>>> self._precollect()
"""
import ibeis
from ibeis.init import main_helpers
self.dbdir = ibeis.sysres.lookup_dbdir(self.dbname)
ibs = ibeis.opendb(dbdir=self.dbdir)
if ibs.dbname.startswith('PZ_PB_RF_TRAIN'):
aids = ibs.get_valid_aids()
elif ibs.dbname.startswith('LF_ALL'):
aids = ibs.get_valid_aids()
elif ibs.dbname.startswith('PZ_Master'):
# PZ_Master is too big to run in full. Select a smaller sample.
# Be sure to include photobomb and incomparable cases.
aids = ibs.filter_annots_general(
require_timestamp=True,
species='primary',
is_known=True,
minqual='poor',
)
infr = ibeis.AnnotInference(ibs=ibs, aids=aids)
infr.reset_feedback('staging', apply=True)
minority_ccs = find_minority_class_ccs(infr)
minority_aids = set(ut.flatten(minority_ccs))
# We need to do our best to select a small sample here
flags = [
'left' in text for text in ibs.annots(aids).viewpoint_code
]
left_aids = ut.compress(aids, flags)
majority_aids = set(
ibs.filter_annots_general(left_aids,
require_timestamp=True,
species='primary',
minqual='poor',
require_quality=True,
min_pername=2,
max_pername=15))
# This produces 5720 annotations
aids = sorted(majority_aids.union(minority_aids))
else:
aids = ibs.filter_annots_general(require_timestamp=True,
is_known=True,
species='primary',
minqual='poor')
if ibs.dbname.startswith('MantaMatcher'):
# Remove some of the singletons for this db
annots = ibs.annots(aids)
names = annots.group2(annots.nids)
multis = [aids for aids in names if len(aids) > 1]
singles = [aids for aids in names if len(aids) == 1]
rng = np.random.RandomState(3988708794)
aids = ut.flatten(multis)
aids += ut.shuffle(ut.flatten(singles), rng=rng)[0:358]
# ibs.print_annot_stats(aids, prefix='P')
main_helpers.monkeypatch_encounters(ibs, aids, minutes=30)
print('post monkey patch')
# if False:
# ibs.print_annot_stats(aids, prefix='P')
self.ibs = ibs
self.aids_pool = aids
def remerge_subset():
"""
Assumes ibs1 is an updated subset of ibs2.
Re-merges ibs1 back into ibs2.
TODO: annotmatch table must have non-directional edges for this to work.
I.e. u < v
Ignore:
# Ensure annotmatch and names are up to date with staging
# Load graph
import ibei
ibs = ibeis.opendb('PZ_PB_RF_TRAIN')
infr = ibeis.AnnotInference(aids='all', ibs=ibs, verbose=3)
infr.reset_feedback('staging', apply=True)
infr.relabel_using_reviews()
# Check deltas
infr.ibeis_name_group_delta_info()
infr.ibeis_delta_info()
# Write if it looks good
infr.write_ibeis_annotmatch_feedback()
infr.write_ibeis_name_assignment()
Ignore:
import ibeis
ibs = ibeis.opendb('PZ_Master1')
infr = ibeis.AnnotInference(ibs, 'all')
infr.reset_feedback('annotmatch', apply=True)
CommandLine:
python -m ibeis.dbio.export_subset remerge_subset
"""
import ibeis
ibs1 = ibeis.opendb('PZ_PB_RF_TRAIN')
ibs2 = ibeis.opendb('PZ_Master1')
gids1, gids2 = ibs1.images(), ibs2.images()
idxs1, idxs2 = ut.isect_indices(gids1.uuids, gids2.uuids)
isect_gids1, isect_gids2 = gids1.take(idxs1), gids2.take(idxs2)
assert all(
set.issubset(set(a1), set(a2))
for a1, a2 in zip(isect_gids1.annot_uuids, isect_gids2.annot_uuids)
)
annot_uuids = ut.flatten(isect_gids1.annot_uuids)
# aids1 = ibs1.annots(ibs1.get_annot_aids_from_uuid(annot_uuids), asarray=True)
# aids2 = ibs2.annots(ibs2.get_annot_aids_from_uuid(annot_uuids), asarray=True)
aids1 = ibs1.annots(uuids=annot_uuids, asarray=True)
aids2 = ibs2.annots(uuids=annot_uuids, asarray=True)
import numpy as np
to_aids2 = dict(zip(aids1, aids2))
# to_aids1 = dict(zip(aids2, aids1))
# Step 1) Update individual annot properties
# These annots need updates
# np.where(aids1.visual_uuids != aids2.visual_uuids)
# np.where(aids1.semantic_uuids != aids2.semantic_uuids)
annot_unary_props = [
# 'yaws', 'bboxes', 'thetas', 'qual', 'species', 'unary_tags']
'yaws', 'bboxes', 'thetas', 'qual', 'species', 'case_tags', 'multiple',
'age_months_est_max', 'age_months_est_min', # 'sex_texts'
]
to_change = {}
for key in annot_unary_props:
prop1 = getattr(aids1, key)
prop2 = getattr(aids2, key)
diff_idxs = set(np.where(prop1 != prop2)[0])
if diff_idxs:
diff_prop1 = ut.take(prop1, diff_idxs)
diff_prop2 = ut.take(prop2, diff_idxs)
print('key = %r' % (key,))
print('diff_prop1 = %r' % (diff_prop1,))
print('diff_prop2 = %r' % (diff_prop2,))
to_change[key] = diff_idxs
if to_change:
changed_idxs = ut.unique(ut.flatten(to_change.values()))
print('Found %d annots that need updated properties' % len(changed_idxs))
print('changing unary attributes: %r' % (to_change,))
if False and ut.are_you_sure('apply change'):
for key, idxs in to_change.items():
subaids1 = aids1.take(idxs)
subaids2 = aids2.take(idxs)
prop1 = getattr(subaids1, key)
# prop2 = getattr(subaids2, key)
setattr(subaids2, key, prop1)
else:
print('Annot properties are in sync. Nothing to change')
# Step 2) Update annotmatch - pairwise relationships
infr1 = ibeis.AnnotInference(aids=aids1.aids, ibs=ibs1, verbose=3,
autoinit=False)
# infr2 = ibeis.AnnotInference(aids=ibs2.annots().aids, ibs=ibs2, verbose=3)
aids2 = ibs2.get_valid_aids(is_known=True)
infr2 = ibeis.AnnotInference(aids=aids2, ibs=ibs2, verbose=3)
infr2.reset_feedback('annotmatch', apply=True)
# map feedback from ibs1 onto ibs2 using ibs2 aids.
fb1 = infr1.read_ibeis_annotmatch_feedback()
fb1_t = {(to_aids2[u], to_aids2[v]): val for (u, v), val in fb1.items()}
fb1_df_t = infr2._pandas_feedback_format(fb1_t).drop('am_rowid', axis=1)
# Add transformed feedback into ibs2
infr2.add_feedback_from(fb1_df_t)
# Now ensure that dummy connectivity exists to preserve origninal names
# from ibeis.algo.graph import nx_utils
# for (u, v) in infr2.find_mst_edges('name_label'):
# infr2.draw_aids((u, v))
# cc1 = infr2.pos_graph.connected_to(u)
# cc2 = infr2.pos_graph.connected_to(v)
# print(nx_utils.edges_cross(infr2.graph, cc1, cc2))
# infr2.neg_redundancy(cc1, cc2)
# infr2.pos_redundancy(cc2)
infr2.relabel_using_reviews(rectify=True)
infr2.apply_nondynamic_update()
if False:
infr2.ibeis_delta_info()
infr2.ibeis_name_group_delta_info()
if len(list(infr2.inconsistent_components())) > 0:
raise NotImplementedError('need to fix inconsistencies first')
# Make it so it just loops until inconsistencies are resolved
infr2.prioritize()
infr2.qt_review_loop()
else:
infr2.write_ibeis_staging_feedback()
infr2.write_ibeis_annotmatch_feedback()
infr2.write_ibeis_name_assignment()
# if False:
# # Fix any inconsistency
# infr2.start_qt_interface(loop=False)
# test_nodes = [5344, 5430, 5349, 5334, 5383, 2280, 2265, 2234, 5399,
# 5338, 2654]
# import networkx as nx
# nx.is_connected(infr2.graph.subgraph(test_nodes))
# # infr = ibeis.AnnotInference(aids=test_nodes, ibs=ibs2, verbose=5)
# # randomly sample some new labels to verify
# import guitool_ibeis as gt
# from ibeis.gui import inspect_gui
# gt.ensure_qapp()
# ut.qtensure()
# old_groups = ut.group_items(name_delta.index.tolist(), name_delta['old_name'])
# del old_groups['____']
# new_groups = ut.group_items(name_delta.index.tolist(), name_delta['new_name'])
# from ibeis.algo.hots import simulate
# c = simulate.compare_groups(
# list(new_groups.values()),
# list(old_groups.values()),
# )
# ut.map_vals(len, c)
# for aids in c['pred_splits']:
# old_nids = ibs2.get_annot_nids(aids)
# new_nids = ut.take_column(infr2.gen_node_attrs('name_label', aids), 1)
# split_aids = ut.take_column(ut.group_items(aids, new_nids).values(), 0)
# aid1, aid2 = split_aids[0:2]
# if False:
# inspect_gui.show_vsone_tuner(ibs2, aid1, aid2)
# infr2.start_qt_interface(loop=False)
# if False:
# # import ibeis
# ibs1 = ibeis.opendb('PZ_PB_RF_TRAIN')
# infr1 = ibeis.AnnotInference(aids='all', ibs=ibs1, verbose=3)
# infr1.initialize_graph()
# # infr1.reset_feedback('staging')
# infr1.reset_feedback('annotmatch')
# infr1.apply_feedback_edges()
# infr1.relabel_using_reviews()
# infr1.apply_review_inference()
# infr1.start_qt_interface(loop=False)
# delta = infr2.match_state_delta()
# print('delta = %r' % (delta,))
# infr2.ensure_mst()
# infr2.relabel_using_reviews()
# infr2.apply_review_inference()
# mst_edges = infr2.find_mst_edges()
# set(infr2.graph.edges()).intersection(mst_edges)
return
"""
def fix_annotmatch_pzmaster1():
"""
PZ_Master1 had annotmatch rowids that did not agree with the current name
labeling. Looking at the inconsistencies in the graph interface was too
cumbersome, because over 3000 annots were incorrectly grouped together.
This function deletes any annotmatch rowid that is not consistent with the
current labeling so we can go forward with using the new AnnotInference
object
"""
import ibeis
ibs = ibeis.opendb('PZ_Master1')
infr = ibeis.AnnotInference(ibs=ibs, aids=ibs.get_valid_aids(), verbose=5)
infr.initialize_graph()
annots = ibs.annots()
aid_to_nid = ut.dzip(annots.aids, annots.nids)
if False:
infr.reset_feedback()
infr.ensure_mst()
infr.apply_feedback_edges()
infr.relabel_using_reviews()
infr.start_qt_interface()
# Get annotmatch rowids that agree with current labeling
if False:
annotmatch = ibs.db.get_table_as_pandas('annotmatch')
import pandas as pd
flags1 = pd.isnull(annotmatch['annotmatch_evidence_decision'])
flags2 = annotmatch['annotmatch_tag_text'] == ''
bad_part = annotmatch[flags1 & flags2]
rowids = bad_part.index.tolist()
ibs.delete_annotmatch(rowids)
if False:
# Delete bidirectional annotmatches
annotmatch = ibs.db.get_table_as_pandas('annotmatch')
df = annotmatch.set_index(['annot_rowid1', 'annot_rowid2'])
# Find entires that have both directions
pairs1 = annotmatch[['annot_rowid1', 'annot_rowid2']].values
f_edges = {tuple(p) for p in pairs1}
b_edges = {tuple(p[::-1]) for p in pairs1}
isect_edges = {tuple(sorted(p)) for p in b_edges.intersection(f_edges)}
isect_edges1 = list(isect_edges)
isect_edges2 = [p[::-1] for p in isect_edges]
# cols = ['annotmatch_evidence_decision', 'annotmatch_tag_text']
import pandas as pd
custom_ = {
(559, 4909): (False, ['photobomb']),
(7918, 8041): (False, ['photobomb']),
(6634, 6754): (False, ['photobomb']),
(3707, 3727): (False, ['photobomb']),
(86, 103): (False, ['photobomb']),
}
extra_ = {
}
fixme_edges = []
d1 = df.loc[isect_edges1].reset_index(drop=False)
d2 = df.loc[isect_edges2].reset_index(drop=False)
flags = d1['annotmatch_evidence_decision'] != d2['annotmatch_evidence_decision']
from ibeis.tag_funcs import _parse_tags
for f, r1, r2 in zip(flags, d1.iterrows(), d2.iterrows()):
v1, v2 = r1[1], r2[1]
aid1 = v1['annot_rowid1']
aid2 = v1['annot_rowid2']
truth_real = (ibs.const.EVIDENCE_DECISION.POSITIVE
if aid_to_nid[aid1] == aid_to_nid[aid2] else
ibs.const.EVIDENCE_DECISION.NEGATIVE)
truth1 = v1['annotmatch_evidence_decision']
truth2 = v2['annotmatch_evidence_decision']
t1 = _parse_tags(v1['annotmatch_tag_text'])
t2 = _parse_tags(v2['annotmatch_tag_text'])
newtag = ut.union_ordered(t1, t2)
if (aid1, aid2) in custom_:
continue
fixme_flag = False
if not pd.isnull(truth1):
if truth_real != truth1:
fixme_flag = True
if not pd.isnull(truth2):
if truth_real != truth2:
fixme_flag = True
if fixme_flag:
print('newtag = %r' % (newtag,))
print('truth_real = %r' % (truth_real,))
print('truth1 = %r' % (truth1,))
print('truth2 = %r' % (truth2,))
print('aid1 = %r' % (aid1,))
print('aid2 = %r' % (aid2,))
fixme_edges.append((aid1, aid2))
else:
extra_[(aid1, aid2)] = (truth_real, newtag)
extra_.update(custom_)
new_pairs = extra_.keys()
new_truths = ut.take_column(ut.dict_take(extra_, new_pairs), 0)
new_tags = ut.take_column(ut.dict_take(extra_, new_pairs), 1)
new_tag_texts = [';'.join(t) for t in new_tags]
aids1, aids2 = ut.listT(new_pairs)
# Delete the old
ibs.delete_annotmatch((d1['annotmatch_rowid'].values.tolist() +
d2['annotmatch_rowid'].values.tolist()))
# Add the new
ams = ibs.add_annotmatch_undirected(aids1, aids2)
ibs.set_annotmatch_evidence_decision(ams, new_truths)
ibs.set_annotmatch_tag_text(ams, new_tag_texts)
if False:
import guitool_ibeis as gt
gt.ensure_qapp()
ut.qtensure()
from ibeis.gui import inspect_gui
inspect_gui.show_vsone_tuner(ibs, aid1, aid2)
# pairs2 = pairs1.T[::-1].T
# idx1, idx2 = ut.isect_indices(list(map(tuple, pairs1)),
# list(map(tuple, pairs2)))
# r_edges = list(set(map(tuple, map(sorted, pairs1[idx1]))))
# unique_pairs = list(set(map(tuple, map(sorted, pairs1[idx1]))))
# df = annotmatch.set_index(['annot_rowid1', 'annot_rowid2'])
x = ut.ddict(list)
annotmatch = ibs.db.get_table_as_pandas('annotmatch')
import ubelt as ub
_iter = annotmatch.iterrows()
prog = ub.ProgIter(_iter, length=len(annotmatch))
for k, m in prog:
aid1 = m['annot_rowid1']
aid2 = m['annot_rowid2']
if m['annotmatch_evidence_decision'] == ibs.const.EVIDENCE_DECISION.POSITIVE:
if aid_to_nid[aid1] == aid_to_nid[aid2]:
x['agree1'].append(k)
else:
x['disagree1'].append(k)
elif m['annotmatch_evidence_decision'] == ibs.const.EVIDENCE_DECISION.NEGATIVE:
if aid_to_nid[aid1] == aid_to_nid[aid2]:
x['disagree2'].append(k)
else:
x['agree2'].append(k)
ub.map_vals(len, x)
ut.dict_hist(annotmatch.loc[x['disagree1']]['annotmatch_tag_text'])
disagree1 = annotmatch.loc[x['disagree1']]
pb_disagree1 = disagree1[disagree1['annotmatch_tag_text'] == 'photobomb']
aids1 = pb_disagree1['annot_rowid1'].values.tolist()
aids2 = pb_disagree1['annot_rowid2'].values.tolist()
aid_pairs = list(zip(aids1, aids2))
infr = ibeis.AnnotInference.from_pairs(aid_pairs, ibs=ibs, verbose=5)
if False:
feedback = infr.read_ibeis_annotmatch_feedback(edges=infr.edges())
infr.external_feedback = feedback
infr.apply_feedback_edges()
infr.start_qt_interface(loop=False)
# Delete these values
if False:
nonpb_disagree1 = disagree1[disagree1['annotmatch_tag_text'] != 'photobomb']
disagree2 = annotmatch.loc[x['disagree2']]
ibs.delete_annotmatch(nonpb_disagree1['annotmatch_rowid'])
ibs.delete_annotmatch(disagree2['annotmatch_rowid'])
# ut.dict_hist(disagree1['annotmatch_tag_text'])
import networkx as nx
graph = nx.Graph()
graph.add_edges_from(zip(pb_disagree1['annot_rowid1'], pb_disagree1['annot_rowid2']))
list(nx.connected_components(graph))
set(annotmatch.loc[x['disagree2']]['annotmatch_tag_text'])
def fix_bidirectional_annotmatch(ibs):
import ibeis
infr = ibeis.AnnotInference(ibs=ibs, aids='all', verbose=5)
infr.initialize_graph()
annots = ibs.annots()
aid_to_nid = ut.dzip(annots.aids, annots.nids)
# Delete bidirectional annotmatches
annotmatch = ibs.db.get_table_as_pandas('annotmatch')
df = annotmatch.set_index(['annot_rowid1', 'annot_rowid2'])
# Find entires that have both directions
pairs1 = annotmatch[['annot_rowid1', 'annot_rowid2']].values
f_edges = {tuple(p) for p in pairs1}
b_edges = {tuple(p[::-1]) for p in pairs1}
isect_edges = {tuple(sorted(p)) for p in b_edges.intersection(f_edges)}
print('Found %d bidirectional edges' % len(isect_edges))
isect_edges1 = list(isect_edges)
isect_edges2 = [p[::-1] for p in isect_edges]
import pandas as pd
extra_ = {}
fixme_edges = []
d1 = df.loc[isect_edges1].reset_index(drop=False)
d2 = df.loc[isect_edges2].reset_index(drop=False)
flags = d1['annotmatch_evidence_decision'] != d2['annotmatch_evidence_decision']
from ibeis.tag_funcs import _parse_tags
for f, r1, r2 in zip(flags, d1.iterrows(), d2.iterrows()):
v1, v2 = r1[1], r2[1]
aid1 = v1['annot_rowid1']
aid2 = v1['annot_rowid2']
truth_real = (ibs.const.EVIDENCE_DECISION.POSITIVE
if aid_to_nid[aid1] == aid_to_nid[aid2] else
ibs.const.EVIDENCE_DECISION.NEGATIVE)
truth1 = v1['annotmatch_evidence_decision']
truth2 = v2['annotmatch_evidence_decision']
t1 = _parse_tags(v1['annotmatch_tag_text'])
t2 = _parse_tags(v2['annotmatch_tag_text'])
newtag = ut.union_ordered(t1, t2)
fixme_flag = False
if not pd.isnull(truth1):
if truth_real != truth1:
fixme_flag = True
if not pd.isnull(truth2):
if truth_real != truth2:
fixme_flag = True
if fixme_flag:
print('--')
print('t1, t2 = %r, %r' % (t1, t2))
print('newtag = %r' % (newtag,))
print('truth_real, truth1, truth2 = %r, %r, %r' % (
truth_real, truth1, truth2,))
print('aid1, aid2 = %r, %r' % (aid1, aid2))
fixme_edges.append(tuple(sorted((aid1, aid2))))
else:
extra_[(aid1, aid2)] = (truth_real, newtag)
if len(fixme_edges) > 0:
# need to manually fix these edges
fix_infr = ibeis.AnnotInference.from_pairs(fixme_edges, ibs=ibs, verbose=5)
feedback = fix_infr.read_ibeis_annotmatch_feedback(only_existing_edges=True)
infr = fix_infr
fix_infr.external_feedback = feedback
fix_infr.apply_feedback_edges()
fix_infr.start_qt_interface(loop=False)
# DELETE OLD EDGES TWICE
ams = ibs.get_annotmatch_rowid_from_edges(fixme_edges)
ibs.delete_annotmatch(ams)
ams = ibs.get_annotmatch_rowid_from_edges(fixme_edges)
ibs.delete_annotmatch(ams)
# MANUALLY CALL THIS ONCE FINISHED
# TO ONLY CHANGE ANNOTMATCH EDGES
infr.write_ibeis_staging_feedback()
infr.write_ibeis_annotmatch_feedback()
# extra_.update(custom_)
new_pairs = extra_.keys()
new_truths = ut.take_column(ut.dict_take(extra_, new_pairs), 0)
new_tags = ut.take_column(ut.dict_take(extra_, new_pairs), 1)
new_tag_texts = [';'.join(t) for t in new_tags]
aids1, aids2 = ut.listT(new_pairs)
# Delete the old
ibs.delete_annotmatch((d1['annotmatch_rowid'].values.tolist() +
d2['annotmatch_rowid'].values.tolist()))
# Add the new
ams = ibs.add_annotmatch_undirected(aids1, aids2)
ibs.set_annotmatch_evidence_decision(ams, new_truths)
ibs.set_annotmatch_tag_text(ams, new_tag_texts)
if False:
import guitool_ibeis as gt
gt.ensure_qapp()
ut.qtensure()
from ibeis.gui import inspect_gui
inspect_gui.show_vsone_tuner(ibs, aid1, aid2)