def biggest_dirs(drive):
print('Biggest Dirs in %r' % (drive,))
dpath_list = drive.dpath_list
fidxs_list = ut.dict_take(drive.dpath_to_fidx, dpath_list)
unflat_dpath_bytes_list = ut.list_unflat_take(drive.fpath_bytes_list, fidxs_list)
dpath_nbytes_list = list(map(sum, unflat_dpath_bytes_list))
sortx = ut.list_argsort(dpath_nbytes_list)[::-1]
sel = sortx[0:10]
biggest_nbytes = ut.take(dpath_nbytes_list, sel)
biggest_dpaths = ut.take(dpath_list, sel)
biginfo_list = list(zip(map(ut.byte_str2, biggest_nbytes), biggest_dpaths))
print(ut.list_str(biginfo_list, strvals=True))
pass
def biggest_dirs(drive):
print('Biggest Dirs in %r' % (drive, ))
dpath_list = drive.dpath_list
fidxs_list = ut.dict_take(drive.dpath_to_fidx, dpath_list)
unflat_dpath_bytes_list = ut.list_unflat_take(drive.fpath_bytes_list,
fidxs_list)
dpath_nbytes_list = list(map(sum, unflat_dpath_bytes_list))
sortx = ut.list_argsort(dpath_nbytes_list)[::-1]
sel = sortx[0:10]
biggest_nbytes = ut.take(dpath_nbytes_list, sel)
biggest_dpaths = ut.take(dpath_list, sel)
biginfo_list = list(
zip(map(ut.byte_str2, biggest_nbytes), biggest_dpaths))
print(ut.repr2(biginfo_list, strvals=True))
pass
def make_name_model(num_annots, num_names=None, verbose=True, mode=1):
"""
Defines the general name model
CommandLine:
python -m wbia.algo.hots.bayes --exec-make_name_model --show
Example:
>>> # DISABLE_DOCTEST
>>> from wbia.algo.hots.bayes import * # NOQA
>>> defaults = dict(num_annots=2, num_names=2, verbose=True, mode=2)
>>> kw = ut.argparse_funckw(make_name_model, defaults)
>>> model = make_name_model(**kw)
>>> ut.quit_if_noshow()
>>> show_model(model, show_prior=True)
>>> ut.show_if_requested()
"""
# annots = ut.chr_range(num_annots, base='a')
mode = ut.get_argval('--mode', default=mode)
annots = ut.chr_range(num_annots,
base=ut.get_argval('--base', default='a'))
# The indexes of match CPDs will not change if another annotation is added
upper_diag_idxs = ut.colwise_diag_idxs(num_annots, 2)
if num_names is None:
num_names = num_annots
# -- Define CPD Templates
def match_pmf(match_type, n1, n2):
if n1 == n2:
val = 1.0 if match_type == 'same' else 0.0
# val = .999 if match_type == 'same' else 0.001
elif n1 != n2:
# val = 0.01 if match_type == 'same' else .99
val = 0.0 if match_type == 'same' else 1.0
return val
def score_pmf(score_type, match_type):
score_lookup = {
'same': {
'low': 0.1,
'high': 0.9,
'veryhigh': 0.9
},
'diff': {
'low': 0.9,
'high': 0.09,
'veryhigh': 0.01
}
#'same': {'low': .1, 'high': .9},
#'diff': {'low': .9, 'high': .1}
}
val = score_lookup[match_type][score_type]
return val
def score_pmf3(score_type, match_type, isdup='False'):
score_lookup = {
'False': {
'same': {
'low': 0.1,
'high': 0.5,
'veryhigh': 0.4
},
'diff': {
'low': 0.9,
'high': 0.09,
'veryhigh': 0.01
},
},
'True': {
'same': {
'low': 0.01,
'high': 0.2,
'veryhigh': 0.79
},
'diff': {
'low': 0.4,
'high': 0.4,
'veryhigh': 0.2
},
},
}
val = score_lookup[isdup][match_type][score_type]
return val
def score_pmf2(score_type, n1, n2):
score_lookup = {
True: {
'low': 0.1,
'high': 0.4,
'veryhigh': 0.5
},
False: {
'low': 0.9,
'high': 0.09,
'veryhigh': 0.01
},
}
val = score_lookup[n1 == n2][score_type]
return val
def dup_pmf(dupstate, match_type):
lookup = {
'same': {
'True': 0.5,
'False': 0.5
},
'diff': {
'True': 0.0,
'False': 1.0
},
}
return lookup[match_type][dupstate]
def check_pmf(n0, n1, match_type):
pass
def trimatch_pmf(match_ab, match_bc, match_ca):
lookup = {
'same': {
'same': {
'same': 1,
'diff': 0
},
'diff': {
'same': 0,
'diff': 1
},
},
'diff': {
'same': {
'same': 0,
'diff': 1
},
'diff': {
'same': 0.5,
'diff': 0.5
},
},
}
return lookup[match_ca][match_bc][match_ab]
name_cpd_t = pgm_ext.TemplateCPD('name', ('n', num_names),
varpref='N',
special_basis_pool=SPECIAL_BASIS_POOL)
if mode == 1 or mode == 5:
match_cpd_t = pgm_ext.TemplateCPD(
'match',
['diff', 'same'],
varpref='M',
evidence_ttypes=[name_cpd_t, name_cpd_t],
pmf_func=match_pmf,
)
if mode == 5:
trimatch_cpd_t = pgm_ext.TemplateCPD(
'tri_match',
['diff', 'same'],
varpref='T',
# evidence_ttypes=[match_cpd_t, match_cpd_t, match_cpd_t],
evidence_ttypes=[match_cpd_t, match_cpd_t],
pmf_func=trimatch_pmf,
)
score_cpd_t = pgm_ext.TemplateCPD(
#'score', ['low', 'high', 'veryhigh'],
'score',
['low', 'high'],
varpref='S',
evidence_ttypes=[match_cpd_t],
pmf_func=score_pmf,
)
else:
score_cpd_t = pgm_ext.TemplateCPD(
#'score', ['low', 'high', 'veryhigh'],
'score',
['low', 'high'],
varpref='S',
evidence_ttypes=[match_cpd_t],
pmf_func=score_pmf,
)
elif mode == 2:
name_cpd_t = pgm_ext.TemplateCPD('name', ('n', num_names),
varpref='N',
special_basis_pool=SPECIAL_BASIS_POOL)
score_cpd_t = pgm_ext.TemplateCPD(
#'score', ['low', 'high', 'veryhigh'],
'score',
['low', 'high'],
varpref='S',
evidence_ttypes=[name_cpd_t, name_cpd_t],
pmf_func=score_pmf2,
)
elif mode == 3 or mode == 4:
match_cpd_t = pgm_ext.TemplateCPD(
'match',
['diff', 'same'],
varpref='M',
evidence_ttypes=[name_cpd_t, name_cpd_t],
pmf_func=match_pmf,
)
if mode == 3:
dup_cpd_t = pgm_ext.TemplateCPD('dup', ['False', 'True'],
varpref='D')
else:
dup_cpd_t = pgm_ext.TemplateCPD(
'dup',
['False', 'True'],
varpref='D',
evidence_ttypes=[match_cpd_t],
pmf_func=dup_pmf,
)
score_cpd_t = pgm_ext.TemplateCPD(
'score',
['low', 'high', 'veryhigh'],
varpref='S',
evidence_ttypes=[match_cpd_t, dup_cpd_t],
pmf_func=score_pmf3,
)
# Instanciate templates
if mode == 1 or mode == 5:
name_cpds = [name_cpd_t.new_cpd(parents=aid) for aid in annots]
namepair_cpds = ut.list_unflat_take(name_cpds, upper_diag_idxs)
match_cpds = [
match_cpd_t.new_cpd(parents=cpds) for cpds in namepair_cpds
]
score_cpds = [
score_cpd_t.new_cpd(parents=cpds) for cpds in zip(match_cpds)
]
if mode == 5:
# triple_idxs = ut.colwise_diag_idxs(num_annots, 3)
tid2_match = {cpd._template_id: cpd for cpd in match_cpds}
trimatch_cpds = []
# such hack
for cpd in match_cpds:
parents = []
this_ = list(cpd._template_id)
for aid in annots:
if aid in this_:
continue
for aid2 in this_:
key = aid2 + aid
if key not in tid2_match:
key = aid + aid2
parents += [tid2_match[key]]
trimatch_cpds += [trimatch_cpd_t.new_cpd(parents=parents)]
# score_cpds = [score_cpd_t.new_cpd(parents=cpds)
# for cpds in zip(trimatch_cpds)]
cpd_list = name_cpds + score_cpds + match_cpds + trimatch_cpds
else:
cpd_list = name_cpds + score_cpds + match_cpds
elif mode == 2:
name_cpds = [name_cpd_t.new_cpd(parents=aid) for aid in annots]
namepair_cpds = ut.list_unflat_take(name_cpds, upper_diag_idxs)
score_cpds = [
score_cpd_t.new_cpd(parents=cpds) for cpds in namepair_cpds
]
cpd_list = name_cpds + score_cpds
elif mode == 3 or mode == 4:
name_cpds = [name_cpd_t.new_cpd(parents=aid) for aid in annots]
namepair_cpds = ut.list_unflat_take(name_cpds, upper_diag_idxs)
match_cpds = [
match_cpd_t.new_cpd(parents=cpds) for cpds in namepair_cpds
]
if mode == 3:
dup_cpds = [
dup_cpd_t.new_cpd(parents=''.join(map(str, aids)))
for aids in ut.list_unflat_take(annots, upper_diag_idxs)
]
else:
dup_cpds = [
dup_cpd_t.new_cpd(parents=[mcpds]) for mcpds in match_cpds
]
score_cpds = [
score_cpd_t.new_cpd(parents=([mcpds] + [dcpd]))
for mcpds, dcpd in zip(match_cpds, dup_cpds)
]
cpd_list = name_cpds + score_cpds + match_cpds + dup_cpds
# logger.info('upper_diag_idxs = %r' % (upper_diag_idxs,))
logger.info('score_cpds = %r' %
(ut.list_getattr(score_cpds, 'variable'), ))
# import sys
# sys.exit(1)
# Make Model
model = pgm_ext.define_model(cpd_list)
model.num_names = num_names
if verbose:
model.print_templates()
# ut.colorprint('\n --- CPD Templates ---', 'blue')
# for temp_cpd in templates:
# ut.colorprint(temp_cpd._cpdstr('psql'), 'cyan')
# print_ascii_graph(model)
return model
def name_model_mode1(num_annots, num_names=None, verbose=True):
r"""
spaghettii
CommandLine:
python -m wbia.algo.hots.bayes --exec-name_model_mode1 --show
python -m wbia.algo.hots.bayes --exec-name_model_mode1
python -m wbia.algo.hots.bayes --exec-name_model_mode1 --num-annots=3
Example:
>>> # DISABLE_DOCTEST
>>> from wbia.algo.hots.bayes import * # NOQA
>>> defaults = dict(num_annots=2, num_names=2, verbose=True)
>>> kw = ut.argparse_funckw(name_model_mode1, defaults)
>>> model = name_model_mode1(**kw)
>>> ut.quit_if_noshow()
>>> show_model(model, show_prior=False, show_title=False)
>>> ut.show_if_requested()
Ignore:
import nx2tikz
logger.info(nx2tikz.dumps_tikz(model, layout='layered', use_label=True))
"""
annots = ut.chr_range(num_annots,
base=ut.get_argval('--base', default='a'))
# The indexes of match CPDs will not change if another annotation is added
upper_diag_idxs = ut.colwise_diag_idxs(num_annots, 2)
if num_names is None:
num_names = num_annots
# +--- Define CPD Templates ---
# +-- Name Factor ---
name_cpd_t = pgm_ext.TemplateCPD('name', ('n', num_names),
varpref='N',
special_basis_pool=SPECIAL_BASIS_POOL)
name_cpds = [name_cpd_t.new_cpd(parents=aid) for aid in annots]
# +-- Match Factor ---
def match_pmf(match_type, n1, n2):
return {
True: {
'same': 1.0,
'diff': 0.0
},
False: {
'same': 0.0,
'diff': 1.0
},
}[n1 == n2][match_type]
match_cpd_t = pgm_ext.TemplateCPD(
'match',
['diff', 'same'],
varpref='M',
evidence_ttypes=[name_cpd_t, name_cpd_t],
pmf_func=match_pmf,
)
namepair_cpds = ut.list_unflat_take(name_cpds, upper_diag_idxs)
match_cpds = [match_cpd_t.new_cpd(parents=cpds) for cpds in namepair_cpds]
# +-- Score Factor ---
def score_pmf(score_type, match_type):
score_lookup = {
'same': {
'low': 0.1,
'high': 0.9,
'veryhigh': 0.9
},
'diff': {
'low': 0.9,
'high': 0.09,
'veryhigh': 0.01
},
}
val = score_lookup[match_type][score_type]
return val
score_cpd_t = pgm_ext.TemplateCPD(
'score',
['low', 'high'],
varpref='S',
evidence_ttypes=[match_cpd_t],
pmf_func=score_pmf,
)
score_cpds = [
score_cpd_t.new_cpd(parents=cpds) for cpds in zip(match_cpds)
]
# L___ End CPD Definitions ___
cpd_list = name_cpds + score_cpds + match_cpds
logger.info('score_cpds = %r' %
(ut.list_getattr(score_cpds, 'variable'), ))
# Make Model
model = pgm_ext.define_model(cpd_list)
model.num_names = num_names
if verbose:
model.print_templates()
return model
def name_model_mode5(num_annots, num_names=None, verbose=True, mode=1):
mode = ut.get_argval('--mode', default=mode)
annots = ut.chr_range(num_annots,
base=ut.get_argval('--base', default='a'))
# The indexes of match CPDs will not change if another annotation is added
upper_diag_idxs = ut.colwise_diag_idxs(num_annots, 2)
if num_names is None:
num_names = num_annots
# -- Define CPD Templates
name_cpd_t = pgm_ext.TemplateCPD('name', ('n', num_names),
varpref='N',
special_basis_pool=SPECIAL_BASIS_POOL)
name_cpds = [name_cpd_t.new_cpd(parents=aid) for aid in annots]
def match_pmf(match_type, n1, n2):
return {
True: {
'same': 1.0,
'diff': 0.0
},
False: {
'same': 0.0,
'diff': 1.0
},
}[n1 == n2][match_type]
match_cpd_t = pgm_ext.TemplateCPD(
'match',
['diff', 'same'],
varpref='M',
evidence_ttypes=[name_cpd_t, name_cpd_t],
pmf_func=match_pmf,
)
namepair_cpds = ut.list_unflat_take(name_cpds, upper_diag_idxs)
match_cpds = [match_cpd_t.new_cpd(parents=cpds) for cpds in namepair_cpds]
def trimatch_pmf(match_ab, match_bc, match_ca):
lookup = {
'same': {
'same': {
'same': 1,
'diff': 0
},
'diff': {
'same': 0,
'diff': 1
},
},
'diff': {
'same': {
'same': 0,
'diff': 1
},
'diff': {
'same': 0.5,
'diff': 0.5
},
},
}
return lookup[match_ca][match_bc][match_ab]
trimatch_cpd_t = pgm_ext.TemplateCPD(
'tri_match',
['diff', 'same'],
varpref='T',
evidence_ttypes=[match_cpd_t, match_cpd_t],
pmf_func=trimatch_pmf,
)
# triple_idxs = ut.colwise_diag_idxs(num_annots, 3)
tid2_match = {cpd._template_id: cpd for cpd in match_cpds}
trimatch_cpds = []
# such hack
for cpd in match_cpds:
parents = []
this_ = list(cpd._template_id)
for aid in annots:
if aid in this_:
continue
for aid2 in this_:
key = aid2 + aid
if key not in tid2_match:
key = aid + aid2
parents += [tid2_match[key]]
trimatch_cpds += [trimatch_cpd_t.new_cpd(parents=parents)]
def score_pmf(score_type, match_type):
score_lookup = {
'same': {
'low': 0.1,
'high': 0.9,
'veryhigh': 0.9
},
'diff': {
'low': 0.9,
'high': 0.09,
'veryhigh': 0.01
},
}
val = score_lookup[match_type][score_type]
return val
score_cpd_t = pgm_ext.TemplateCPD(
'score',
['low', 'high'],
varpref='S',
evidence_ttypes=[match_cpd_t],
pmf_func=score_pmf,
)
score_cpds = [
score_cpd_t.new_cpd(parents=cpds) for cpds in zip(match_cpds)
]
# score_cpds = [score_cpd_t.new_cpd(parents=cpds)
# for cpds in zip(trimatch_cpds)]
cpd_list = name_cpds + score_cpds + match_cpds + trimatch_cpds
logger.info('score_cpds = %r' %
(ut.list_getattr(score_cpds, 'variable'), ))
# Make Model
model = pgm_ext.define_model(cpd_list)
model.num_names = num_names
if verbose:
model.print_templates()
return model
def make_name_model(num_annots, num_names=None, verbose=True, mode=1,
num_scores=2, p_score_given_same=None,
hack_score_only=False, score_basis=None,
special_names=None):
r"""
CommandLine:
python -m ibeis.algo.hots.bayes --exec-make_name_model --show
python -m ibeis.algo.hots.bayes --exec-make_name_model
python -m ibeis.algo.hots.bayes --exec-make_name_model --num-annots=3
Example:
>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.bayes import * # NOQA
>>> defaults = dict(num_annots=2, num_names=2, verbose=True)
>>> modeltype = ut.get_argval('--modeltype', default='bayes')
>>> kw = ut.argparse_funckw(make_name_model, defaults)
>>> model = make_name_model(**kw)
>>> ut.quit_if_noshow()
>>> model.show_model(show_prior=False, show_title=False, modeltype=modeltype)
>>> ut.show_if_requested()
"""
if special_names is None:
special_names = SPECIAL_BASIS_POOL
assert mode == 1, 'only can do mode 1'
base = ut.get_argval('--base', type_=str, default='a')
annots = ut.chr_range(num_annots, base=base)
# The indexes of match CPDs will not change if another annotation is added
upper_diag_idxs = ut.colwise_diag_idxs(num_annots, 2)
if hack_score_only:
upper_diag_idxs = upper_diag_idxs[-hack_score_only:]
if num_names is None:
num_names = num_annots
# +--- Define CPD Templates and Instantiation ---
cpd_list = []
# Name Factor
name_cpd_t = pgm_ext.TemplateCPD(
'name', ('n', num_names),
special_basis_pool=special_names)
name_cpds = [name_cpd_t.new_cpd(parents=aid) for aid in annots]
#name_cpds = [name_cpd_t.new_cpd(parents=aid, constrain_state=count)
# for count, aid in enumerate(annots, start=1)]
cpd_list.extend(name_cpds)
# Match Factor
def match_pmf(match_type, n1, n2):
return {
True: {'same': 1.0, 'diff': 0.0},
False: {'same': 0.0, 'diff': 1.0},
}[n1 == n2][match_type]
match_states = ['diff', 'same']
match_cpd_t = pgm_ext.TemplateCPD(
'match', match_states,
evidence_ttypes=[name_cpd_t, name_cpd_t], pmf_func=match_pmf)
namepair_cpds = ut.list_unflat_take(name_cpds, upper_diag_idxs)
match_cpds = [match_cpd_t.new_cpd(parents=cpds)
for cpds in namepair_cpds]
cpd_list.extend(match_cpds)
# Score Factor
score_states = list(range(num_scores))
if score_basis is not None:
score_states = ['%.2f' % (s,) for s in score_basis]
if p_score_given_same is None:
tmp = np.arange(num_scores + 1)[1:]
tmp = np.cumsum(tmp)
tmp = (tmp / tmp.sum())
p_score_given_same = tmp
def score_pmf(score_type, match_type):
if isinstance(score_type, six.string_types):
score_type = score_states.index(score_type)
if match_type == 'same':
return p_score_given_same[score_type]
else:
return p_score_given_same[-(score_type + 1)]
score_cpd_t = pgm_ext.TemplateCPD(
'score', score_states,
evidence_ttypes=[match_cpd_t], pmf_func=score_pmf)
score_cpds = [score_cpd_t.new_cpd(parents=cpds)
for cpds in zip(match_cpds)]
cpd_list.extend(score_cpds)
with_humans = False
if with_humans:
human_states = ['diff', 'same']
human_cpd_t = pgm_ext.TemplateCPD(
'human', human_states,
evidence_ttypes=[match_cpd_t], pmf_func=[[.9, .1], [.1, .9]])
human_cpds = [human_cpd_t.new_cpd(parents=cpds)
for cpds in zip(match_cpds)]
cpd_list.extend(human_cpds)
with_rank = False # Rank depends on dependant scores
if with_rank:
rank_states = ['0', '1', '2', '3']
rank_cpd_t = pgm_ext.TemplateCPD(
'rank', rank_states,
evidence_ttypes=[match_cpd_t], pmf_func=None)
rank_cpds = [rank_cpd_t.new_cpd(parents=cpds)
for cpds in zip(match_cpds)]
cpd_list.extend(rank_cpds)
# L___ End CPD Definitions ___
print('score_cpds = %r' % (ut.list_getattr(score_cpds, 'variable'),))
# Make Model
model = pgm_ext.define_model(cpd_list)
model.num_names = num_names
if verbose:
model.print_templates(ignore_ttypes=['match'])
return model
def get_col(table, tbl_rowids, colnames=None):
"""
colnames = ('mask', 'size')
FIXME; unpacking is confusing with sql controller
"""
# print('Get prop of %r, colnames=%r' % (table, colnames))
try:
request_unpack = False
if colnames is None:
colnames = table.data_colnames
#table._internal_data_colnames
else:
if isinstance(colnames, six.text_type):
request_unpack = True
colnames = (colnames,)
# print('* colnames = %r' % (colnames,))
eager = True
nInput = None
total = 0
intern_colnames = []
extern_resolve_colxs = []
nesting_xs = []
for c in colnames:
if c in table.external_to_internal:
intern_colnames.append([table.external_to_internal[c]])
read_func = table.extern_read_funcs[c]
extern_resolve_colxs.append((total, read_func))
nesting_xs.append(total)
total += 1
elif c in table.nested_to_flat:
nest = table.nested_to_flat[c]
nesting_xs.append(list(range(total, total + len(nest))))
intern_colnames.append(nest)
total += len(nest)
else:
nesting_xs.append(total)
intern_colnames.append([c])
total += 1
flat_intern_colnames = tuple(ut.flatten(intern_colnames))
# do sql read
# FIXME: understand unpack_scalars and keepwrap
raw_prop_list = table.get_internal_columns(
tbl_rowids, flat_intern_colnames, eager, nInput,
unpack_scalars=True, keepwrap=True)
# unpack_scalars=not
# request_unpack)
# print('depth(raw_prop_list) = %r' % (ut.depth_profile(raw_prop_list),))
prop_listT = list(zip(*raw_prop_list))
for extern_colx, read_func in extern_resolve_colxs:
data_list = []
for uri in prop_listT[extern_colx]:
try:
# FIXME: only do this for a localpath
uri1 = ut.unixjoin(table.depc.cache_dpath, uri)
data = read_func(uri1)
except Exception as ex:
ut.printex(ex, 'failed to load external data', iswarning=False)
raise
# FIXME
#data = None
data_list.append(data)
prop_listT[extern_colx] = data_list
nested_proplistT = ut.list_unflat_take(prop_listT, nesting_xs)
for tx in ut.where([isinstance(xs, list) for xs in nesting_xs]):
nested_proplistT[tx] = list(zip(*nested_proplistT[tx]))
prop_list = list(zip(*nested_proplistT))
if request_unpack:
prop_list = [None if p is None else p[0] for p in prop_list]
except Exception as ex:
ut.printex(ex, 'failed in get col', keys=[
'table.tablename',
'request_unpack',
'tbl_rowids',
'colnames',
'raw_prop_list',
(ut.depth_profile, 'raw_prop_list'),
'prop_listT',
(ut.depth_profile, 'prop_listT'),
'nesting_xs',
'nested_proplistT',
'prop_list'])
raise
return prop_list
def fix_duplicates(drive):
r"""
for every duplicate file passing a (eg avi) filter, remove the file
that is in the smallest directory. On a tie use the smallest dpath.
This will filter all duplicate files in a folder into a single folder.
but... need to look at non-duplicates in that folder and decide if they
should be moved as well. So, should trigger on folders that have at
least 50% duplicate. Might not want to move curated folders.
Example:
cd ~/local/scripts
>>> from register_files import * # NOQA
>>> dpaths = ut.get_argval('--drives', type_=list, default=['E:/'])#'D:/', 'E:/', 'F:/'])
>>> drives = [Drive(root_dpath) for root_dpath in dpaths]
>>> E = drive = drives[0]
>>> #D, E, F = drives
"""
print('Fixing Duplicates in %r' % (drive,))
list_ = drive.fpath_hashX_list
multiindex_dict_ = build_multindex(list_)
duplicate_hashes = [
key for key, val in six.iteritems(multiindex_dict_)
if len(val) > 1
]
duplicate_idxs = ut.dict_take(multiindex_dict_, duplicate_hashes)
unflat_fpaths = ut.list_unflat_take(drive.fpath_list, duplicate_idxs)
# Check if any dups have been removed
still_exists = ut.unflat_map(exists, unflat_fpaths)
unflat_idxs2 = ut.zipcompress(duplicate_idxs, still_exists)
duplicate_idxs = [idxs for idxs in unflat_idxs2 if len(idxs) > 1]
# Look at duplicate files
unflat_fpaths = ut.list_unflat_take(drive.fpath_list, duplicate_idxs)
unflat_sizes = ut.list_unflat_take(drive.fpath_bytes_list, duplicate_idxs)
# Find highly coupled directories
if True:
coupled_dirs = []
for fpaths in unflat_fpaths:
#basedir = ut.longest_existing_path(commonprefix(fpaths))
dirs = sorted(list(map(dirname, fpaths)))
_list = list(range(len(dirs)))
idxs = ut.upper_diag_self_prodx(_list)
coupled_dirs.extend(list(map(tuple, ut.list_unflat_take(dirs, idxs))))
hist_ = ut.dict_hist(coupled_dirs)
coupled_idxs = ut.list_argsort(hist_.values())[::-1]
most_coupled = ut.take(list(hist_.keys()), coupled_idxs[0:100])
print('Coupled fpaths: ' + ut.list_str(most_coupled, nl=True))
print('%d unique files are duplicated' % (len(unflat_sizes),))
#print('Duplicate sizes: ' + ut.list_str(unflat_sizes[0:10], nl=True))
#print('Duplicate fpaths: ' + ut.list_str(unflat_fpaths[0:10], nl=True))
#print('Duplicate fpaths: ' + ut.list_str(unflat_fpaths[0::5], nl=True))
print('Duplicate fpaths: ' + ut.list_str(unflat_fpaths, nl=True))
# Find duplicate directories
dpath_list = list(drive.dpath_to_fidx.keys())
fidxs_list = ut.dict_take(drive.dpath_to_fidx, drive.dpath_list)
#exists_list = list(map(exists, drive.fpath_list))
#unflat_exists = ut.list_unflat_take(exists_list, fidxs_list)
fname_registry = [basename(fpath) for fpath in drive.fpath_list]
unflat_fnames = ut.list_unflat_take(fname_registry, fidxs_list)
def unsorted_list_hash(list_):
return ut.hashstr27(str(sorted(list_)))
unflat_fname_sets = list(map(unsorted_list_hash, ut.ProgIter(unflat_fnames, freq=10000)))
fname_based_duplicate_dpaths = []
multiindex_dict2_ = build_multindex(unflat_fname_sets)
fname_based_duplicate_hashes = [key for key, val in multiindex_dict2_.items() if len(val) > 1]
print('#fname_based_duplicate_dpaths = %r' % (len(fname_based_duplicate_hashes),))
fname_based_duplicate_didxs = ut.dict_take(multiindex_dict2_, fname_based_duplicate_hashes)
fname_based_duplicate_dpaths = ut.list_unflat_take(dpath_list, fname_based_duplicate_didxs)
print(ut.repr3(fname_based_duplicate_dpaths[0:10]))
def fix_duplicates(drive):
r"""
for every duplicate file passing a (eg avi) filter, remove the file
that is in the smallest directory. On a tie use the smallest dpath.
This will filter all duplicate files in a folder into a single folder.
but... need to look at non-duplicates in that folder and decide if they
should be moved as well. So, should trigger on folders that have at
least 50% duplicate. Might not want to move curated folders.
Example:
cd ~/local/scripts
>>> from register_files import * # NOQA
>>> dpaths = ut.get_argval('--drives', type_=list, default=['E:/'])#'D:/', 'E:/', 'F:/'])
>>> drives = [Drive(root_dpath) for root_dpath in dpaths]
>>> E = drive = drives[0]
>>> #D, E, F = drives
"""
print('Fixing Duplicates in %r' % (drive, ))
list_ = drive.fpath_hashX_list
multiindex_dict_ = build_multindex(list_)
duplicate_hashes = [
key for key, val in six.iteritems(multiindex_dict_) if len(val) > 1
]
duplicate_idxs = ut.dict_take(multiindex_dict_, duplicate_hashes)
unflat_fpaths = ut.list_unflat_take(drive.fpath_list, duplicate_idxs)
# Check if any dups have been removed
still_exists = ut.unflat_map(exists, unflat_fpaths)
unflat_idxs2 = ut.zipcompress(duplicate_idxs, still_exists)
duplicate_idxs = [idxs for idxs in unflat_idxs2 if len(idxs) > 1]
# Look at duplicate files
unflat_fpaths = ut.list_unflat_take(drive.fpath_list, duplicate_idxs)
unflat_sizes = ut.list_unflat_take(drive.fpath_bytes_list,
duplicate_idxs)
# Find highly coupled directories
if True:
coupled_dirs = []
for fpaths in unflat_fpaths:
#basedir = ut.longest_existing_path(commonprefix(fpaths))
dirs = sorted(list(map(dirname, fpaths)))
_list = list(range(len(dirs)))
idxs = ut.upper_diag_self_prodx(_list)
coupled_dirs.extend(
list(map(tuple, ut.list_unflat_take(dirs, idxs))))
hist_ = ut.dict_hist(coupled_dirs)
coupled_idxs = ut.list_argsort(hist_.values())[::-1]
most_coupled = ut.take(list(hist_.keys()), coupled_idxs[0:100])
print('Coupled fpaths: ' + ut.repr2(most_coupled, nl=True))
print('%d unique files are duplicated' % (len(unflat_sizes), ))
#print('Duplicate sizes: ' + ut.repr2(unflat_sizes[0:10], nl=True))
#print('Duplicate fpaths: ' + ut.repr2(unflat_fpaths[0:10], nl=True))
#print('Duplicate fpaths: ' + ut.repr2(unflat_fpaths[0::5], nl=True))
print('Duplicate fpaths: ' + ut.repr2(unflat_fpaths, nl=True))
# Find duplicate directories
dpath_list = list(drive.dpath_to_fidx.keys())
fidxs_list = ut.dict_take(drive.dpath_to_fidx, drive.dpath_list)
#exists_list = list(map(exists, drive.fpath_list))
#unflat_exists = ut.list_unflat_take(exists_list, fidxs_list)
fname_registry = [basename(fpath) for fpath in drive.fpath_list]
unflat_fnames = ut.list_unflat_take(fname_registry, fidxs_list)
def unsorted_list_hash(list_):
return ut.hashstr27(str(sorted(list_)))
unflat_fname_sets = list(
map(unsorted_list_hash, ut.ProgIter(unflat_fnames, freq=10000)))
fname_based_duplicate_dpaths = []
multiindex_dict2_ = build_multindex(unflat_fname_sets)
fname_based_duplicate_hashes = [
key for key, val in multiindex_dict2_.items() if len(val) > 1
]
print('#fname_based_duplicate_dpaths = %r' %
(len(fname_based_duplicate_hashes), ))
fname_based_duplicate_didxs = ut.dict_take(
multiindex_dict2_, fname_based_duplicate_hashes)
fname_based_duplicate_dpaths = ut.list_unflat_take(
dpath_list, fname_based_duplicate_didxs)
print(ut.repr3(fname_based_duplicate_dpaths[0:10]))
def make_name_model(num_annots, num_names=None, verbose=True, mode=1,
num_scores=2, p_score_given_same=None,
hack_score_only=False, score_basis=None,
special_names=None):
r"""
CommandLine:
python -m ibeis.algo.hots.bayes --exec-make_name_model --no-cnn
python -m ibeis.algo.hots.bayes --exec-make_name_model --show --no-cnn
python -m ibeis.algo.hots.bayes --exec-make_name_model --num-annots=3
Example:
>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.bayes import * # NOQA
>>> defaults = dict(num_annots=2, num_names=2, verbose=True)
>>> modeltype = ut.get_argval('--modeltype', default='bayes')
>>> kw = ut.argparse_funckw(make_name_model, defaults)
>>> model = make_name_model(**kw)
>>> ut.quit_if_noshow()
>>> model.show_model(show_prior=False, show_title=False, modeltype=modeltype)
>>> ut.show_if_requested()
"""
if special_names is None:
special_names = SPECIAL_BASIS_POOL
assert mode == 1, 'only can do mode 1'
base = ut.get_argval('--base', type_=str, default='a')
annots = ut.chr_range(num_annots, base=base)
# The indexes of match CPDs will not change if another annotation is added
upper_diag_idxs = ut.colwise_diag_idxs(num_annots, 2)
if hack_score_only:
upper_diag_idxs = upper_diag_idxs[-hack_score_only:]
if num_names is None:
num_names = num_annots
# +--- Define CPD Templates and Instantiation ---
cpd_list = []
# Name Factor
name_cpd_t = pgm_ext.TemplateCPD(
NAME_TTYPE, ('n', num_names),
special_basis_pool=special_names)
name_cpds = [name_cpd_t.new_cpd(parents=aid) for aid in annots]
#name_cpds = [name_cpd_t.new_cpd(parents=aid, constrain_state=count)
# for count, aid in enumerate(annots, start=1)]
cpd_list.extend(name_cpds)
# Match Factor
def match_pmf(match_type, n1, n2):
return {
True: {'same': 1.0, 'diff': 0.0},
False: {'same': 0.0, 'diff': 1.0},
}[n1 == n2][match_type]
match_states = ['diff', 'same']
match_cpd_t = pgm_ext.TemplateCPD(
MATCH_TTYPE, match_states,
evidence_ttypes=[name_cpd_t, name_cpd_t], pmf_func=match_pmf)
#match_cpd_t.varpref = 'S'
namepair_cpds = ut.list_unflat_take(name_cpds, upper_diag_idxs)
match_cpds = [match_cpd_t.new_cpd(parents=cpds)
for cpds in namepair_cpds]
cpd_list.extend(match_cpds)
# Score Factor
score_states = list(range(num_scores))
if score_basis is not None:
score_states = ['%.2f' % (s,) for s in score_basis]
if p_score_given_same is None:
tmp = np.arange(num_scores + 1)[1:]
tmp = np.cumsum(tmp)
tmp = (tmp / tmp.sum())
p_score_given_same = tmp
def score_pmf(score_type, match_type):
if isinstance(score_type, six.string_types):
score_type = score_states.index(score_type)
if match_type == 'same':
return p_score_given_same[score_type]
else:
return p_score_given_same[-(score_type + 1)]
score_cpd_t = pgm_ext.TemplateCPD(
SCORE_TTYPE, score_states,
evidence_ttypes=[match_cpd_t], pmf_func=score_pmf)
#match_cpd_t.varpref = 'P'
score_cpds = [score_cpd_t.new_cpd(parents=cpds)
for cpds in zip(match_cpds)]
cpd_list.extend(score_cpds)
with_humans = False
if with_humans:
human_states = ['diff', 'same']
human_cpd_t = pgm_ext.TemplateCPD(
'human', human_states,
evidence_ttypes=[match_cpd_t], pmf_func=[[.9, .1], [.1, .9]])
human_cpds = [human_cpd_t.new_cpd(parents=cpds)
for cpds in zip(match_cpds)]
cpd_list.extend(human_cpds)
with_rank = False # Rank depends on dependant scores
if with_rank:
rank_states = ['0', '1', '2', '3']
rank_cpd_t = pgm_ext.TemplateCPD(
'rank', rank_states,
evidence_ttypes=[match_cpd_t], pmf_func=None)
rank_cpds = [rank_cpd_t.new_cpd(parents=cpds)
for cpds in zip(match_cpds)]
cpd_list.extend(rank_cpds)
# L___ End CPD Definitions ___
print('score_cpds = %r' % (ut.list_getattr(score_cpds, 'variable'),))
# Make Model
model = pgm_ext.define_model(cpd_list)
model.num_names = num_names
if verbose:
model.print_templates(ignore_ttypes=[MATCH_TTYPE])
return model
