ibeis.algo.hots package¶

Subpackages¶

Submodules¶

ibeis.algo.hots._grave_bayes module¶

Arc reversal http://www.cs.toronto.edu/~cebly/Papers/simulation.pdf

TODO:

Need to find faster more mature libraries http://dlib.net/bayes.html http://www.cs.waikato.ac.nz/ml/weka/ http://www.cs.waikato.ac.nz/~remco/weka.bn.pdf https://code.google.com/p/pebl-project/ https://github.com/abhik/pebl http://www.cs.ubc.ca/~murphyk/Software/bnsoft.html

Demo case where we think we know the labels of others. Only one unknown name. Need to classify it as one of the other known names.

References

https://en.wikipedia.org/wiki/Bayesian_network https://class.coursera.org/pgm-003/lecture/17 http://www.cs.ubc.ca/~murphyk/Bayes/bnintro.html http://www3.cs.stonybrook.edu/~sael/teaching/cse537/Slides/chapter14d_BP.pdf http://www.cse.unsw.edu.au/~cs9417ml/Bayes/Pages/PearlPropagation.html https://github.com/pgmpy/pgmpy.git http://pgmpy.readthedocs.org/en/latest/ http://nipy.bic.berkeley.edu:5000/download/11 http://pgmpy.readthedocs.org/en/latest/wiki.html#add-feature-to-accept-and-output-state-names-for-models http://www.csse.monash.edu.au/bai/book/BAI_Chapter2.pdf

Clustering with CRF:

http://srl.informatik.uni-freiburg.de/publicationsdir/tipaldiIROS09.pdf http://www.dis.uniroma1.it/~dottoratoii/media/students/documents/thesis_tipaldi.pdf An Unsupervised Conditional Random Fields Approach for Clustering Gene Expression Time Series http://bioinformatics.oxfordjournals.org/content/24/21/2467.full

CRFs:

http://homepages.inf.ed.ac.uk/csutton/publications/crftutv2.pdf

AlphaBeta Swap:

https://github.com/amueller/gco_python https://github.com/pmneila/PyMaxflow http://www.cs.cornell.edu/rdz/papers/bvz-iccv99.pdf

http://arxiv.org/pdf/1411.6340.pdf Iteratively Reweighted Graph Cut for Multi-label MRFs with Non-convex Priors

Fusion Moves:

http://www.robots.ox.ac.uk/~vilem/fusion.pdf http://hci.iwr.uni-heidelberg.de/publications/mip/techrep/beier_15_fusion.pdf

Consensus Clustering

Explaining Away

Course Notes:

Tie breaking for MAP assignment. https://class.coursera.org/pgm-003/lecture/60 * random perdibiation

Correspondence Problem is discussed in https://class.coursera.org/pgm-003/lecture/68

Sparse Pattern Factors

Collective Inference: Plate Models / Aggragator CPD is used to define dependencies.

ibeis.algo.hots._grave_bayes.flow()[source]¶

http://pmneila.github.io/PyMaxflow/maxflow.html#maxflow-fastmin

pip install PyMaxFlow pip install pystruct pip install hdbscan

ibeis.algo.hots._grave_bayes.make_name_model(num_annots, num_names=None, verbose=True, mode=1)[source]¶

Defines the general name model

CommandLine:

python -m ibeis.algo.hots.bayes --exec-make_name_model --show

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.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()

ibeis.algo.hots._grave_bayes.name_model_mode1(num_annots, num_names=None, verbose=True)[source]¶

spaghettii

CommandLine:

python -m ibeis.algo.hots.bayes --exec-name_model_mode1 --show
python -m ibeis.algo.hots.bayes --exec-name_model_mode1
python -m ibeis.algo.hots.bayes --exec-name_model_mode1 --num-annots=3

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.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()

ibeis.algo.hots._grave_bayes.name_model_mode5(num_annots, num_names=None, verbose=True, mode=1)[source]¶

ibeis.algo.hots._grave_bayes.show_model(model, evidence={}, soft_evidence={}, **kwargs)[source]¶

References

http://stackoverflow.com/questions/22207802/pygraphviz-networkx-set-node-level-or-layer

ibeis.algo.hots._grave_bayes.try_query(model, infr, evidence, interest_ttypes=[], verbose=True)[source]¶

CommandLine:

python -m ibeis.algo.hots.bayes --exec-try_query --show

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.bayes import *  # NOQA
>>> verbose = True
>>> other_evidence = {}
>>> name_evidence = [1, None, 0, None]
>>> score_evidence = ['high', 'low', 'low']
>>> query_vars = None
>>> model = make_name_model(num_annots=4, num_names=4, verbose=True, mode=1)
>>> model, evidence, soft_evidence = update_model_evidence(model, name_evidence, score_evidence, other_evidence)
>>> interest_ttypes = ['name']
>>> infr = pgmpy.inference.BeliefPropagation(model)
>>> evidence = infr._ensure_internal_evidence(evidence, model)
>>> query_results = try_query(model, infr, evidence, interest_ttypes, verbose)
>>> result = ('query_results = %s' % (str(query_results),))
>>> ut.quit_if_noshow()
>>> show_model(model, show_prior=True, **query_results)
>>> ut.show_if_requested()

ibeis.algo.hots._grave_scorenorm module¶

ibeis.algo.hots._neighbor_experiment module¶

ibeis.algo.hots._neighbor_experiment.augment_nnindexer_experiment()[source]¶

References

http://answers.opencv.org/question/44592/flann-index-training-fails-with-segfault/

CommandLine:

utprof.py -m ibeis.algo.hots._neighbor_experiment --test-augment_nnindexer_experiment
python -m ibeis.algo.hots._neighbor_experiment --test-augment_nnindexer_experiment

python -m ibeis.algo.hots._neighbor_experiment --test-augment_nnindexer_experiment --db PZ_MTEST --diskshow --adjust=.1 --save "augment_experiment_{db}.png" --dpath='.' --dpi=180 --figsize=9,6
python -m ibeis.algo.hots._neighbor_experiment --test-augment_nnindexer_experiment --db PZ_Master0 --diskshow --adjust=.1 --save "augment_experiment_{db}.png" --dpath='.' --dpi=180 --figsize=9,6 --nosave-flann --show
python -m ibeis.algo.hots._neighbor_experiment --test-augment_nnindexer_experiment --db PZ_Master0 --diskshow --adjust=.1 --save "augment_experiment_{db}.png" --dpath='.' --dpi=180 --figsize=9,6 --nosave-flann --show


python -m ibeis.algo.hots._neighbor_experiment --test-augment_nnindexer_experiment --db PZ_Master0 --diskshow --adjust=.1 --save "augment_experiment_{db}.png" --dpath='.' --dpi=180 --figsize=9,6 --nosave-flann --no-api-cache --nocache-uuids

python -m ibeis.algo.hots._neighbor_experiment --test-augment_nnindexer_experiment --db PZ_MTEST --show
python -m ibeis.algo.hots._neighbor_experiment --test-augment_nnindexer_experiment --db PZ_Master0 --show

# RUNS THE SEGFAULTING CASE
python -m ibeis.algo.hots._neighbor_experiment --test-augment_nnindexer_experiment --db PZ_Master0 --show
# Debug it
gdb python
run -m ibeis.algo.hots._neighbor_experiment --test-augment_nnindexer_experiment --db PZ_Master0 --show
gdb python
run -m ibeis.algo.hots._neighbor_experiment --test-augment_nnindexer_experiment --db PZ_Master0 --diskshow --adjust=.1 --save "augment_experiment_{db}.png" --dpath='.' --dpi=180 --figsize=9,6

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots._neighbor_experiment import *  # NOQA
>>> # execute function
>>> augment_nnindexer_experiment()
>>> # verify results
>>> ut.show_if_requested()

ibeis.algo.hots._neighbor_experiment.flann_add_time_experiment()[source]¶

builds plot of number of annotations vs indexer build time.

TODO: time experiment

CommandLine:

python -m ibeis.algo.hots._neighbor_experiment --test-flann_add_time_experiment --db PZ_MTEST --show
python -m ibeis.algo.hots._neighbor_experiment --test-flann_add_time_experiment --db PZ_Master0 --show
utprof.py -m ibeis.algo.hots._neighbor_experiment --test-flann_add_time_experiment --show

valgrind --tool=memcheck --suppressions=valgrind-python.supp python -m ibeis.algo.hots._neighbor_experiment --test-flann_add_time_experiment --db PZ_MTEST --no-with-reindex

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots._neighbor_experiment import *  # NOQA
>>> import ibeis
>>> #ibs = ibeis.opendb('PZ_MTEST')
>>> result = flann_add_time_experiment()
>>> # verify results
>>> print(result)
>>> ut.show_if_requested()

ibeis.algo.hots._neighbor_experiment.pyflann_remove_and_save()[source]¶

References

# Logic goes here ~/code/flann/src/cpp/flann/algorithms/kdtree_index.h

~/code/flann/src/cpp/flann/util/serialization.h ~/code/flann/src/cpp/flann/util/dynamic_bitset.h

# Bindings go here ~/code/flann/src/cpp/flann/flann.cpp ~/code/flann/src/cpp/flann/flann.h

# Contains stuff for the flann namespace like flann::log_level # Also has Index with # Matrix<ElementType> features; SEEMS USEFUL ~/code/flann/src/cpp/flann/flann.hpp

# Wrappers go here ~/code/flann/src/python/pyflann/flann_ctypes.py ~/code/flann/src/python/pyflann/index.py

~/local/build_scripts/flannscripts/autogen_bindings.py

Greping:

cd ~/code/flann/src grep -ER cleanRemovedPoints * grep -ER removed_points_ *

CommandLine:

python -m ibeis.algo.hots._neighbor_experiment --exec-pyflann_remove_and_save

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots._neighbor_experiment import *  # NOQA
>>> pyflann_remove_and_save()

ibeis.algo.hots._neighbor_experiment.pyflann_test_remove_add()[source]¶

CommandLine:

python -m ibeis.algo.hots._neighbor_experiment --exec-pyflann_test_remove_add

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots._neighbor_experiment import *  # NOQA
>>> pyflann_test_remove_add()

ibeis.algo.hots._neighbor_experiment.pyflann_test_remove_add2()[source]¶

CommandLine:

python -m ibeis.algo.hots._neighbor_experiment --exec-pyflann_test_remove_add2

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots._neighbor_experiment import *  # NOQA
>>> pyflann_test_remove_add2()

ibeis.algo.hots._neighbor_experiment.subindexer_time_experiment()[source]¶

builds plot of number of annotations vs indexer build time.

TODO: time experiment

ibeis.algo.hots._neighbor_experiment.test_incremental_add(ibs)[source]¶

Parameters:	ibs (IBEISController) –

CommandLine:

python -m ibeis.algo.hots._neighbor_experiment --test-test_incremental_add

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.neighbor_index_cache import *  # NOQA
>>> import ibeis
>>> ibs = ibeis.opendb('PZ_MTEST')
>>> result = test_incremental_add(ibs)
>>> print(result)

ibeis.algo.hots._neighbor_experiment.test_multiple_add_removes()[source]¶

CommandLine:

python -m ibeis.algo.hots._neighbor_experiment --exec-test_multiple_add_removes

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots._neighbor_experiment import *  # NOQA
>>> result = test_multiple_add_removes()
>>> print(result)

ibeis.algo.hots._pipeline_helpers module¶

ibeis.algo.hots._pipeline_helpers.print_nearest_neighbor_assignments(qvecs_list, nns_list)[source]¶

ibeis.algo.hots._pipeline_helpers.testdata_matching(*args, **kwargs)[source]¶

>>> from ibeis.algo.hots._pipeline_helpers import *  # NOQA

ibeis.algo.hots._pipeline_helpers.testdata_post_sver(defaultdb=u'PZ_MTEST', qaid_list=None, daid_list=None, codename=u'vsmany', cfgdict=None)[source]¶

>>> from ibeis.algo.hots._pipeline_helpers import *  # NOQA

ibeis.algo.hots._pipeline_helpers.testdata_pre(stopnode, defaultdb=u'testdb1', p=[u'default'], a=[u'default:qindex=0:1,dindex=0:5'], **kwargs)[source]¶

New (1-1-2016) generic pipeline node testdata getter

Parameters:	stopnode (str) – defaultdb (str) – (default = u’testdb1’) p (list) – (default = [u’default:’]) a (list) – (default = [u’default:qsize=1,dsize=4’])
Returns:	(ibs, qreq_, args)
Return type:	tuple

CommandLine:

python -m ibeis.algo.hots._pipeline_helpers --exec-testdata_pre --show

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots._pipeline_helpers import *  # NOQA
>>> stopnode = 'build_chipmatches'
>>> defaultdb = 'testdb1'
>>> p = ['default:']
>>> a = ['default:qindex=0:1,dindex=0:5']
>>> qreq_, args = testdata_pre(stopnode, defaultdb, p, a)

ibeis.algo.hots._pipeline_helpers.testdata_pre_baselinefilter(defaultdb=u'testdb1', qaid_list=None, daid_list=None, codename=u'vsmany')[source]¶

ibeis.algo.hots._pipeline_helpers.testdata_pre_sver(defaultdb=u'PZ_MTEST', qaid_list=None, daid_list=None)[source]¶

>>> from ibeis.algo.hots._pipeline_helpers import *  # NOQA

ibeis.algo.hots._pipeline_helpers.testdata_pre_vsonerr(defaultdb=u'PZ_MTEST', qaid_list=[1], daid_list=u'all')[source]¶

>>> from ibeis.algo.hots._pipeline_helpers import *  # NOQA

ibeis.algo.hots._pipeline_helpers.testdata_pre_weight_neighbors(defaultdb=u'testdb1', qaid_list=[1, 2], daid_list=None, codename=u'vsmany', cfgdict=None)[source]¶

TODO: replace testdata_pre_weight_neighbors with

>>> qreq_, args = plh.testdata_pre('weight_neighbors', defaultdb='testdb1',
>>>                                a=['default:qindex=0:1,dindex=0:5,hackerrors=False'],
>>>                                p=['default:codename=vsmany,bar_l2_on=True,fg_on=False'], verbose=True)

ibeis.algo.hots._pipeline_helpers.testdata_scoring(defaultdb=u'PZ_MTEST', qaid_list=[1], daid_list=u'all')[source]¶

ibeis.algo.hots._pipeline_helpers.testdata_sparse_matchinfo_nonagg(defaultdb=u'testdb1', p=[u'default'])[source]¶

ibeis.algo.hots._pipeline_helpers.testrun_pipeline_upto(qreq_, stop_node=None, verbose=True)[source]¶

Main tester function. Runs the pipeline by mirroring request_ibeis_query_L0, but stops at a requested breakpoint and returns the local variables.

convinience: runs pipeline for tests this should mirror request_ibeis_query_L0

ibeis.algo.hots.automatch_suggestor module¶

Reports decisions and confidences about names (identifications) and exemplars using query results objects.

class ibeis.algo.hots.automatch_suggestor.ChoiceTuple(sorted_nids, sorted_nscore, sorted_rawscore, sorted_aids, sorted_ascores)¶

Bases: tuple

__getnewargs__()¶

Return self as a plain tuple. Used by copy and pickle.

__getstate__()¶

Exclude the OrderedDict from pickling

__repr__()¶

Return a nicely formatted representation string

sorted_aids¶

Alias for field number 3

sorted_ascores¶

Alias for field number 4

sorted_nids¶

Alias for field number 0

sorted_nscore¶

Alias for field number 1

sorted_rawscore¶

Alias for field number 2

class ibeis.algo.hots.automatch_suggestor.ExemplarDecision(new_exemplar_aids, remove_exemplar_aids)¶

Bases: tuple

__getnewargs__()¶

Return self as a plain tuple. Used by copy and pickle.

__getstate__()¶

Exclude the OrderedDict from pickling

__repr__()¶

Return a nicely formatted representation string

new_exemplar_aids¶

Alias for field number 0

remove_exemplar_aids¶

Alias for field number 1

ibeis.algo.hots.automatch_suggestor.exemplar_method1_distinctiveness(ibs, qaid, other_exemplars)[source]¶

choose as exemplar if it is distinctive with respect to other exemplars

ibeis.algo.hots.automatch_suggestor.exemplar_method2_randomness(qaid, other_exemplars)[source]¶

CommandLine:

python -m ibeis.algo.hots.automatch_suggestor --test-exemplar_method2_randomness

Example

>>> # ENABLE_DOCTEST
>>> from ibeis.algo.hots.automatch_suggestor import *  # NOQA
>>> # build test data
>>> random.seed(0)
>>> qaid = 4
>>> other_exemplars = [1, 2, 3, 5, 6, 9]
>>> # execute function
>>> exemplar_decision = exemplar_method2_randomness(qaid, other_exemplars)
>>> exemplar_decision_list = [exemplar_method2_randomness(qaid, other_exemplars) for _ in range(1000)]
>>> # verify results
>>> flat_others = ut.flatten(ut.get_list_column(exemplar_decision_list, 1))
>>> result = str(flat_others)
>>> print(result)
[3, 2, 6, 1, 3, 2, 9, 3, 6, 2, 1, 5, 1, 6, 1]

ibeis.algo.hots.automatch_suggestor.get_qres_name_choices(ibs, cm)[source]¶

returns all possible decision a user could make

TODO: Return the possiblity of a merge. TODO: Ensure that the total probability of each possible choice sums to 1. This will define a probability density function that we can take advantage of

Parameters:	ibs (IBEISController) – ibeis controller object cm (QueryResult) – object of feature correspondences and scores
Returns:	choicetup
Return type:	ChoiceTuple

CommandLine:

python -m ibeis.algo.hots.automatch_suggestor --test-get_qres_name_choices:0

Example

>>> # ENABLE_DOCTEST
>>> from ibeis.algo.hots.automatch_suggestor import *  # NOQA
>>> import ibeis  # NOQA
>>> # build test data
>>> cm, qreq_ = ibeis.testdata_cm()
>>> ibs = qreq_.ibs
>>> choicetup = get_qres_name_choices(ibs, cm)
>>> print(choicetup)
>>> result = ut.numpy_str(choicetup.sorted_nids[0:1], force_dtype=False)
>>> print(result)
np.array([1])

ibeis.algo.hots.automatch_suggestor.get_system_exemplar_suggestion(ibs, qaid)[source]¶

hotspotter returns an exemplar suggestion

TODO:

do a vsone query between all of the exemplars to see if this one is good enough to be added.

TODO:

build a complete graph of exemplar scores and only add if this one is lower than any other edge

SeeAlso:

ibsfuncs.set_exemplars_from_quality_and_viewpoint

CommandLine:

python -m ibeis.algo.hots.automatch_suggestor --test-get_system_exemplar_suggestion:1

Example

>>> # ENABLE_DOCTEST
>>> from ibeis.algo.hots.automatch_suggestor import *  # NOQA
>>> import ibeis
>>> # build test data
>>> ibs = ibeis.opendb('testdb1')
>>> qaid = 2
>>> # execute function
>>> (autoexmplr_msg, exemplar_decision, exemplar_confidence) = get_system_exemplar_suggestion(ibs, qaid)
>>> # verify results
>>> result = str((autoexmplr_msg, exemplar_decision, exemplar_confidence))
>>> print(result)

Example

>>> # ENABLE_DOCTEST
>>> from ibeis.algo.hots.automatch_suggestor import *  # NOQA
>>> import ibeis
>>> # build test data
>>> ibs = ibeis.opendb('testdb1')
>>> qaid = 1
>>> # execute function
>>> (autoexmplr_msg, exemplar_decision, exemplar_confidence) = get_system_exemplar_suggestion(ibs, qaid)
>>> # verify results
>>> result = str((autoexmplr_msg, exemplar_decision, exemplar_confidence))
>>> print(result)

ibeis.algo.hots.automatch_suggestor.get_system_name_suggestion(ibs, choicetup)[source]¶

Suggests a decision based on the current choices

Parameters:	ibs (IBEISController) – qaid (int) – query annotation id cm (QueryResult) – object of feature correspondences and scores metatup (None) –
Returns:	(autoname_msg, autoname_func)
Return type:	tuple

CommandLine:

python -m ibeis.algo.hots.automated_matcher --test-test_incremental_queries:0
python -m ibeis.algo.hots.automated_matcher --test-test_incremental_queries:1
python -m ibeis.algo.hots.automated_matcher --test-get_system_name_suggestion

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.automatch_suggestor import *  # NOQA
>>> import ibeis
>>> # build test data
>>> cm, qreq_ = ibeis.testdata_cm()
>>> ibs = qreq_.ibs
>>> choicetup = get_qres_name_choices(ibs, cm)
>>> (autoname_msg, name, name_confidence) = get_system_name_suggestion(ibs, choicetup)
>>> # verify results
>>> result = str((autoname_msg, name, name_confidence))
>>> print(result)

ibeis.algo.hots.automated_helpers module¶

Idea:

what about the probability of a descriptor match being a score like in SIFT. we can learn that too.

Have:

• semantic and visual uuids
• Test that accepts unknown annotations one at a time and for each runs query, makes decision about name, and executes decision.
• As a placeholder for exemplar decisions an exemplar is added if number of exemplars per name is less than threshold.
• vs-one reranking query mode
• test harness but start with larger test set
• vs-one score normalizer ~~/ score normalizer for different values of K * / different params~~ vs-many score normalization doesnt actually matter. We just need the ranking.
• need to add in the multi-indexer code into the pipeline. Need to decide which subindexers to load given a set of daids
• need to use set query as an exemplar if its vs-one reranking scores are below a threshold
• flip the vsone ratio score so its < .8 rather than > 1.2 or whatever
• start from nothing and let the system make the first few decisions correctly
• tell me the correct answer in the automated test
• turn on multi-indexing. (should just work..., probably bugs though. Just need to throw the switch)
• paramater to only add exemplar if post-normlized score is above a threshold
• ensure vsone ratio test is happening correctly
• normalization gets a cfgstr based on the query
• need to allow for scores to be un-invalidatd post spatial verification e.g. when the first match initially is invalidated through spatial verification but the next matches survive.
• keep distinctiveness weights from vsmany for vsone weighting basically involves keeping weights from different filters and not aggregating match weights until the end.
• Put test query mode into the main application and work on the interface for it.
• add matches to multiple animals (merge)
• update normalizer (have setup the datastructure to allow for it need to integrate it seemlessly)
• score normalization update. on add the new support data, reapply bayes

rule, and save to the current cache for a given algorithm configuration.

• spawn background process to reindex chunks of data

TODO:

• Improve vsone scoring.
• test case where there is a 360 view that is linkable from the tests case
• ~~Remember name_confidence of decisions for manual review~~ Defer

Tasks:

Algorithm::

• Incremental query needs to handle

  • test mode and live mode

  • normalizer update

  • use correct distinctivenes score in vsone

  • tested application of distinctiveness, foreground, ratio,

    spatial_verification, vsone verification, and score normalization.

• Mathematically formal description of the space of choices

  • getting the proability of each choice will give us a much better

    confidence measure for our decision. An example of a probability partition might be .2 - merge with rank1. .2 merge with rank 2, .5 merge with rank1 and rank2, .1 others

• Improved automated exemplar decision mechanism

• Improved automated name decision mechanism

SQL::

• New Image Columns

  • image_posix_timedelta

• New Name Columns

  • name_temp_flag
  • name_alias_text
  • name_uuid
  • name_visual_uuid
  • name_member_annot_rowids_evalstr
  • name_member_num_annot_rowids

• New ImageSet Columns

  • imageset_start_time
  • imageset_end_time
  • imageset_lat
  • imageset_lon
  • imageset_processed_flag
  • imageset_shipped_flag

Decision UIs::

• Query versus top N results

  • ability to draw an undirected edge between the query and any number of

    results. ie create a match any of the top results

  • a match to more than one results should by default merge the two names

    (this involves a name enhancement subtask). trigger a split / merge dialog

• Is Exemplar

  • allows for user to set the exemplars for a given name

• Name Progress

  • Shows the current name matching progress

• Split

  • Allows a user to split off some images from a name into a new name or some other name.

• Merge

  • Allows a user to join two names.

GUI::

• NameTree needs to not refresh unless absolutely necessary

• Time Sync

• ImageSet metadata sync from the SMART

• Hide shipped imagesets

  • put flag to turn them on

• Mark processed imagesets

• Gui naturally ensures that all annotations in the query belong

  to the same species

• Garbage collection function that removes all non-exemplar information from imagesets that have been shipped.

• Spawn process that reindexes large chunks of descriptors as the database grows.

LONG TERM TASKS:

Architecture:

• Pipeline needs

  • DEFER: a move from dict based representation to list based
  • DEFER: spatial verification cyth speedup
  • DEFER: nearest neighbor (based on visual uuid caching) caching

Controller:

• LONGTERM: AutogenController

  • register data convertors for verts / other eval columns. Make several convertors standard and we can tag those columns to autogenerate their functions.
  • be able to mark a column as determined by the aggregate of other columns. Then the data is either generated on the fly, or it is cached and the necessary book-keeping functions are autogenerated.

Decision UIs::

• Is Exemplar

  • LONG TERM: it would be cool if they were visualized by using networkx or some gephi like program and clustered by match score.

ibeis.algo.hots.automated_helpers.add_annot_chunk(ibs_gt, ibs2, aids_chunk1, aid1_to_aid2)[source]¶

adds annotations to the tempoarary database and prevents duplicate additions.

aids_chunk1 = aid_list1

Parameters:	ibs_gt (IBEISController) – ibs2 (IBEISController) – aids_chunk1 (list) – aid1_to_aid2 (dict) –
Returns:	aids_chunk2
Return type:	list

ibeis.algo.hots.automated_helpers.annot_testdb_consistency_checks(ibs_gt, ibs2, aid_list1, aid_list2)[source]¶

ibeis.algo.hots.automated_helpers.assert_testdb_annot_consistency(ibs_gt, ibs2, aid_list1, aid_list2)[source]¶

just tests uuids

if anything goes wrong this should fix it:

from ibeis.other import ibsfuncs aid_list1 = ibs_gt.get_valid_aids() ibs_gt.update_annot_visual_uuids(aid_list1) ibs2.update_annot_visual_uuids(aid_list2) ibsfuncs.fix_remove_visual_dupliate_annotations(ibs_gt)

ibeis.algo.hots.automated_helpers.check_results(ibs_gt, ibs2, aid1_to_aid2, aids_list1_, incinfo)[source]¶

reports how well the incremental query ran when the oracle was calling the shots.

ibeis.algo.hots.automated_helpers.ensure_testdb_clean_data(ibs_gt, ibs2, aid_list1, aid_list2)[source]¶

removes previously set names and exemplars

ibeis.algo.hots.automated_helpers.interactive_commandline_prompt(msg, decisiontype)[source]¶

ibeis.algo.hots.automated_helpers.make_incremental_test_database(ibs_gt, aid_list1, reset)[source]¶

Makes test database. adds image and annotations but does not transfer names. if reset is true the new database is gaurenteed to be built from a fresh start.

Parameters:	ibs_gt (IBEISController) – aid_list1 (list) – reset (bool) – if True the test database is completely rebuilt
Returns:	ibs2
Return type:	IBEISController

ibeis.algo.hots.automated_helpers.setup_incremental_test(ibs_gt, clear_names=True, aid_order=u'shuffle')[source]¶

CommandLine:

python -m ibeis.algo.hots.automated_helpers --test-setup_incremental_test:0

python dev.py -t custom --cfg codename:vsone_unnorm --db PZ_MTEST --allgt --vf --va
python dev.py -t custom --cfg codename:vsone_unnorm --db PZ_MTEST --allgt --vf --va --index 0 4 8 --verbose

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.automated_helpers import *  # NOQA
>>> import ibeis # NOQA
>>> ibs_gt = ibeis.opendb('PZ_MTEST')
>>> ibs2, aid_list1, aid1_to_aid2 = setup_incremental_test(ibs_gt)

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.automated_helpers import *  # NOQA
>>> import ibeis  # NOQA
>>> ibs_gt = ibeis.opendb('GZ_ALL')
>>> ibs2, aid_list1, aid1_to_aid2 = setup_incremental_test(ibs_gt)

ibeis.algo.hots.automated_matcher module¶

CommandLine:

python -c "import utool as ut; ut.write_modscript_alias('Tinc.sh', 'ibeis.algo.hots.qt_inc_automatch')"

sh Tinc.sh --test-test_inc_query:0
sh Tinc.sh --test-test_inc_query:1
sh Tinc.sh --test-test_inc_query:2
sh Tinc.sh --test-test_inc_query:3 --num-initial 5000

python -m ibeis.algo.hots.qt_inc_automatch --test-test_inc_query:0

class ibeis.algo.hots.automated_matcher.Metatup(ibs_gt, aid1_to_aid2)¶

Bases: tuple

__getnewargs__()¶

Return self as a plain tuple. Used by copy and pickle.

__getstate__()¶

Exclude the OrderedDict from pickling

__repr__()¶

Return a nicely formatted representation string

aid1_to_aid2¶

Alias for field number 1

ibs_gt¶

Alias for field number 0

ibeis.algo.hots.automated_matcher.exec_exemplar_decision_and_continue(exemplar_decision, ibs, cm, qreq_, incinfo=None)[source]¶

DECISION STEP 4)

The exemplar decision in the previous step is executed. The persistant vsmany query request is updated if needbe and the execution continues. (currently to the end of this iteration)

ibeis.algo.hots.automated_matcher.exec_name_decision_and_continue(chosen_names, ibs, cm, qreq_, incinfo=None)[source]¶

DECISION STEP 2)

The name decision from the previous step is executed and the score normalizer is updated. Then execution continues to the exemplar decision step.

ibeis.algo.hots.automated_matcher.execute_query_batch(ibs, qaid_chunk, qreq_vsmany_, incinfo)[source]¶

TODO: remove special query

ibeis.algo.hots.automated_matcher.generate_incremental_queries(ibs, qaid_list, incinfo=None)[source]¶

qt entry point. generates query results for the qt harness to process.

Parameters:	ibs (IBEISController) – ibeis controller object qaid_list (list) –

CommandLine:

python -m ibeis.algo.hots.automated_matcher --test-generate_incremental_queries

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.automated_matcher import *  # NOQA
>>> ibs, qaid_chunk = testdata_automatch()
>>> generate_incremental_queries(ibs, qaid_list)

ibeis.algo.hots.automated_matcher.generate_subquery_steps(ibs, qaid_chunk, incinfo=None)[source]¶

Generats query results for the qt harness to then send into the next decision steps.

Parameters:	ibs (IBEISController) – ibeis controller object qaid_chunk – incinfo (dict) –

CommandLine:

python -m ibeis.algo.hots.automated_matcher --test-generate_subquery_steps

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.automated_matcher import *  # NOQA
>>> ibs, qaid_chunk = testdata_automatch()
>>> generate_subquery_steps(ibs, qaid_chunk)

ibeis.algo.hots.automated_matcher.get_name_suggestion(ibs, qaid, choicetup, incinfo)[source]¶

ibeis.algo.hots.automated_matcher.initialize_persistant_query_request(ibs, qaid_chunk)[source]¶

ibeis.algo.hots.automated_matcher.load_or_make_qreq(ibs, qreq_vsmany_, qaid_chunk)[source]¶

ibeis.algo.hots.automated_matcher.run_until_exemplar_decision_signal(ibs, cm, qreq_, incinfo=None)[source]¶

DECISION STEP 3)

Either the system or the user decides if the query should be added to the database as an exemplar.

ibeis.algo.hots.automated_matcher.run_until_finish(incinfo=None)[source]¶

DECISION STEP 5)

ibeis.algo.hots.automated_matcher.run_until_name_decision_signal(ibs, cm, qreq_, incinfo=None)[source]¶

DECISION STEP 1)

Either the system or the user makes a decision about the name of the query annotation.

CommandLine:

python -m ibeis.algo.hots.automated_matcher --test-run_until_name_decision_signal

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.automated_matcher import *  # NOQA
>>> ibs, qaid_chunk = testdata_automatch()
>>> exemplar_aids = ibs.get_valid_aids(is_exemplar=True)
>>> incinfo = {}
>>> gen = generate_subquery_steps(ibs, qaid_chunk, incinfo)
>>> item = six.next(gen)
>>> ibs, cm, qreq_, incinfo = item
>>> # verify results
>>> run_until_name_decision_signal(ibs, cm, qreq_, incinfo)

Ignore::

cm.ishow_top(ibs, sidebyside=False, show_query=True)

ibeis.algo.hots.automated_matcher.test_generate_incremental_queries(ibs_gt, ibs, aid_list1, aid1_to_aid2, num_initial=0, incinfo=None)[source]¶

TODO: move this somewhere else Testing function

Adds and queries new annotations one at a time with oracle guidance ibs1 is ibs_gt, ibs2 is ibs

ibeis.algo.hots.automated_matcher.testdata_automatch(dbname=None)[source]¶

ibeis.algo.hots.automated_matcher.update_normalizer(ibs, cm, qreq_, chosen_names)[source]¶

adds new support data to the current normalizer

FIXME: broken

Parameters:	ibs (IBEISController) – ibeis controller object qreq (QueryRequest) – query request object with hyper-parameters choicetup – name –
Returns:	(tp_rawscore, tn_rawscore)
Return type:	tuple

CommandLine:

python -m ibeis.algo.hots.automated_matcher --test-update_normalizer

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.automated_matcher import *  # NOQA
>>> ibs, qaid_chunk = testdata_automatch()
>>> exemplar_aids = ibs.get_valid_aids(is_exemplar=True)
>>> incinfo = {}
>>> gen = generate_subquery_steps(ibs, qaid_chunk, incinfo)
>>> item = six.next(gen)
>>> ibs, cm, qreq_, incinfo = item
>>> qreq_.load_score_normalizer()
>>> # verify results
>>> chosen_names = ['easy']
>>> update_normalizer(ibs, cm, qreq_, chosen_names)

ibeis.algo.hots.automated_oracle module¶

module for making the correct automatic decisions in incremental tests

ibeis.algo.hots.automated_oracle.get_oracle_name_decision(metatup, ibs, qaid, choicetup, oracle_method=1)[source]¶

Find what the correct decision should be ibs is the database we are working with ibs_gt has pristine groundtruth

ibeis.algo.hots.automated_oracle.get_oracle_name_suggestion(ibs, qaid, choicetup, metatup)[source]¶

main entry point for the oracle

ibeis.algo.hots.automated_oracle.oracle_method1(ibs_gt, ibs, qnid1, aid_list2, aid2_to_aid1, sorted_nids, MAX_LOOK)[source]¶

METHOD 1: MAKE BEST DECISION FROM GIVEN INFORMATION

ibeis.algo.hots.automated_oracle.oracle_method2(ibs_gt, qnid1)[source]¶

METHOD 2: MAKE THE ABSOLUTE CORRECT DECISION REGARDLESS OF RESULT

ibeis.algo.hots.automated_params module¶

module that specified how we choose paramaters based on current search database properties

ibeis.algo.hots.automated_params.choose_vsmany_K(num_names, qaids, daids)[source]¶

TODO: Should also scale up the number of checks as well

method for choosing K in the initial vsmany queries

ibeis.algo.hots.bayes module¶

1. Ambiguity / num names
2. independence of annotations
3. continuous
4. exponential case
5. speicifc examples of our prob
6. human in loop

ibeis.algo.hots.bayes.cluster_query(model, query_vars=None, evidence=None, soft_evidence=None, method=None, operation=u'maximize')[source]¶

CommandLine:

python -m ibeis.algo.hots.bayes --exec-cluster_query --show

GridParams:

>>> param_grid = dict(
>>>     #method=['approx', 'bf', 'bp'],
>>>     method=['approx', 'bp'],
>>> )
>>> combos = ut.all_dict_combinations(param_grid)
>>> index = 0
>>> keys = 'method'.split(', ')
>>> method, = ut.dict_take(combos[index], keys)

GridSetup:

>>> from ibeis.algo.hots.bayes import *  # NOQA
>>> verbose = True
>>> other_evidence = {}
>>> name_evidence = [1, None, None, 0]
>>> score_evidence = [2, 0, 2]
>>> special_names = ['fred', 'sue', 'tom', 'paul']
>>> model = make_name_model(
>>>     num_annots=4, num_names=4, num_scores=3, verbose=True, mode=1,
>>>     special_names=special_names)
>>> method = None
>>> model, evidence, soft_evidence = update_model_evidence(
>>>     model, name_evidence, score_evidence, other_evidence)
>>> evidence = model._ensure_internal_evidence(evidence)
>>> query_vars = ut.list_getattr(model.ttype2_cpds[NAME_TTYPE], 'variable')

GridExample:

>>> # DISABLE_DOCTEST
>>> query_results = cluster_query(model, query_vars, evidence,
>>>                               method=method)
>>> print(ut.repr2(query_results['top_assignments'], nl=1))
>>> ut.quit_if_noshow()
>>> pgm_viz.show_model(model, evidence=evidence, **query_results)
>>> ut.show_if_requested()

ibeis.algo.hots.bayes.collapse_factor_labels(model, reduced_joint, evidence)[source]¶

ibeis.algo.hots.bayes.collapse_labels(model, evidence, reduced_variables, reduced_row_idxs, reduced_values)[source]¶

ibeis.algo.hots.bayes.compute_reduced_joint(model, query_vars, evidence, method, operation=u'maximize')[source]¶

ibeis.algo.hots.bayes.draw_tree_model(model, **kwargs)[source]¶

ibeis.algo.hots.bayes.get_hacked_pos(netx_graph, name_nodes=None, prog=u'dot')[source]¶

ibeis.algo.hots.bayes.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)[source]¶

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()

ibeis.algo.hots.bayes.make_temp_state(state)[source]¶

ibeis.algo.hots.bayes.reduce_marginalize(phi, query_variables=None, evidence={}, inplace=False)[source]¶

Hack for reduction followed by marginalization

Example

>>> reduced_joint = joint.observe(
>>>     query_variables, evidence, inplace=False)
>>> new_rows = reduced_joint._row_labels()
>>> new_vals = reduced_joint.values.ravel()
>>> map_vals = new_rows[new_vals.argmax()]
>>> map_assign = dict(zip(reduced_joint.variables, map_vals))

ibeis.algo.hots.bayes.report_partitioning_statistics(new_reduced_joint)[source]¶

ibeis.algo.hots.bayes.show_model(model, evidence={}, soft_evidence={}, **kwargs)[source]¶

References

http://stackoverflow.com/questions/22207802/pygraphviz-networkx-set-node-level-or-layer

CommandLine:

python -m ibeis.algo.hots.bayes --exec-show_model --show

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.bayes import *  # NOQA
>>> model = '?'
>>> evidence = {}
>>> soft_evidence = {}
>>> result = show_model(model, evidence, soft_evidence)
>>> print(result)
>>> ut.quit_if_noshow()
>>> import plottool as pt
>>> ut.show_if_requested()

ibeis.algo.hots.bayes.test_model(num_annots, num_names, score_evidence=[], name_evidence=[], other_evidence={}, noquery=False, verbose=None, **kwargs)[source]¶

ibeis.algo.hots.bayes.update_model_evidence(model, name_evidence, score_evidence, other_evidence)[source]¶

CommandLine:

python -m ibeis.algo.hots.bayes --exec-update_model_evidence

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.bayes import *  # NOQA
>>> verbose = True
>>> other_evidence = {}
>>> name_evidence = [0, 0, 1, 1, None]
>>> score_evidence = ['high', 'low', 'low', 'low', 'low', 'high']
>>> model = make_name_model(num_annots=5, num_names=3, verbose=True,
>>>                         mode=1)
>>> update_model_evidence(model, name_evidence, score_evidence,
>>>                       other_evidence)

ibeis.algo.hots.chip_match module¶

class ibeis.algo.hots.chip_match.AnnotMatch(cm, *args, **kwargs)[source]¶

Bases: ibeis.algo.hots.chip_match.MatchBaseIO, utool.util_dev.NiceRepr

This implements part the match between whole annotations and the other annotaions / names. This does not include algorithm specific feature matches.

algo_annot_scores¶

algo_name_scores¶

argsort(cm)[source]¶

as_dict(cm, *args, **kwargs)[source]¶

as_simple_dict(cm, keys=[])[source]¶

evaluate_dnids(cm, ibs)[source]¶

classmethod from_dict(class_dict, ibs=None)[source]¶

Convert dict of arguments back to ChipMatch object

get_annot_ranks(cm, daids)[source]¶

get_annot_scores(cm, daids, score_method=None)[source]¶

get_chip_shortlist_aids(cm, num_shortlist)[source]¶

Example

>>> # ENABLE_DOCTEST
>>> from ibeis.algo.hots.chip_match import *  # NOQA
>>> ibs, qreq_, cm_list = plh.testdata_pre_sver('PZ_MTEST', qaid_list=[18])
>>> cm = cm_list[0]
>>> cm.score_nsum(qreq_)
>>> top_daids = cm.get_chip_shortlist_aids(5 * 2)
>>> assert cm.qnid in ibs.get_annot_name_rowids(top_daids)

get_groundtruth_daids(cm)[source]¶

get_groundtruth_flags(cm)[source]¶

get_name_ranks(cm, dnids)[source]¶

get_name_shortlist_aids(cm, nNameShortList, nAnnotPerName)[source]¶

Example

>>> # ENABLE_DOCTEST
>>> from ibeis.algo.hots.chip_match import *  # NOQA
>>> ibs, qreq_, cm_list = plh.testdata_pre_sver('PZ_MTEST', qaid_list=[18])
>>> cm = cm_list[0]
>>> cm.score_nsum(qreq_)
>>> top_daids = cm.get_name_shortlist_aids(5, 2)
>>> assert cm.qnid in ibs.get_annot_name_rowids(top_daids)

get_nid_scores(cm, nid_list)[source]¶

get_num_matches_list(cm)[source]¶

get_ranked_nids(cm)[source]¶

get_ranked_nids_and_aids(cm)[source]¶

Hacky func

Returns:	ibeis.algo.hots.name_scoring.NameScoreTup

get_top_aids(cm, ntop=None)[source]¶

get_top_gf_aids(cm, ibs, ntop=None)[source]¶

get_top_gt_aids(cm, ibs, ntop=None)[source]¶

get_top_nids(cm, ntop=None)[source]¶

get_top_scores(cm, ntop=None)[source]¶

get_top_truth_aids(cm, ibs, truth, ntop=None)[source]¶

top scoring aids of a certain truth value

groundtruth_daids¶

initialize(cm, qaid=None, daid_list=None, score_list=None, dnid_list=None, qnid=None, unique_nids=None, name_score_list=None, annot_score_list=None, autoinit=True)[source]¶

qaid and daid_list are not optional. fm_list and fsv_list are strongly encouraged and will probalby break things if they are not there.

ishow_analysis(cm, qreq_, **kwargs)[source]¶

name_argsort(cm)[source]¶

num_daids¶

ranks¶

set_cannonical_annot_score(cm, annot_score_list)[source]¶

set_cannonical_name_score(cm, annot_score_list, name_score_list)[source]¶

show_analysis(cm, qreq_, **kwargs)[source]¶

show_single_namematch(cm, qreq_, dnid, fnum=None, pnum=None, homog=False, **kwargs)[source]¶

HACK FOR ANNOT MATCH

to_dict(cm, ibs=None)[source]¶

unique_name_ranks¶

class ibeis.algo.hots.chip_match.ChipMatch(cm, *args, **kwargs)[source]¶

Bases: ibeis.algo.hots.chip_match._ChipMatchVisualization, ibeis.algo.hots.chip_match.AnnotMatch, ibeis.algo.hots.chip_match._ChipMatchScorers, ibeis.algo.hots.old_chip_match._OldStyleChipMatchSimulator

behaves as as the ChipMatchOldTup named tuple until we completely replace the old structure

append_featscore_column(cm, filtkey, filtweight_list, inplace=True)[source]¶

assert_self(cm, qreq_=None, ibs=None, strict=False, assert_feats=True, verbose=True)[source]¶

classmethod combine_cms(cm_list)[source]¶

Example

>>> # ENABLE_DOCTEST
>>> from ibeis.core_annots import *  # NOQA
>>> ibs, depc, aid_list = testdata_core(size=4)
>>> request = depc.new_request('vsone', [1], [2, 3, 4], {'dim_size': 450})
>>> rawres_list2 = request.execute(postprocess=False)
>>> cm_list = ut.take_column(rawres_list2, 1)
>>> cls = ChipMatch
>>> out = ChipMatch.combine_cms(cm_list)
>>> out.score_nsum(request)
>>> ut.quit_if_noshow()
>>> out.ishow_analysis(request)
>>> ut.show_if_requested()

compress_annots(cm, flags, inplace=False, keepscores=True)[source]¶

compress_top_feature_matches(cm, num=10, rng=<module 'numpy.random' from '/usr/local/lib/python2.7/dist-packages/numpy/random/__init__.pyc'>, use_random=True)[source]¶

DO NOT USE

FIXME: Use boolean lists

Removes all but the best feature matches for testing purposes rng = np.random.RandomState(0)

dfxs_list¶

extend_results(cm, qreq_, other_aids=None)[source]¶

Return a new ChipMatch containing empty data for an extended set of aids

Parameters:	qreq (ibeis.QueryRequest) – query request object with hyper-parameters other_aids (None) – (default = None)
Returns:	out
Return type:	ibeis.ChipMatch

CommandLine:

python -m ibeis.algo.hots.chip_match --exec-extend_results --show

Example

>>> # ENABLE_DOCTEST
>>> from ibeis.algo.hots.chip_match import *  # NOQA
>>> import ibeis
>>> import ibeis
>>> cm, qreq_ = ibeis.testdata_cm('PZ_MTEST',
>>>                               a='default:dindex=0:10,qindex=0:1',
>>>                               t='best:SV=False')
>>> assert len(cm.daid_list) == 9
>>> cm.assert_self(qreq_)
>>> other_aids = qreq_.ibs.get_valid_aids()
>>> out = cm.extend_results(qreq_, other_aids)
>>> assert len(out.daid_list) == 118
>>> out.assert_self(qreq_)

classmethod from_dict(class_dict, ibs=None)[source]¶

Convert dict of arguments back to ChipMatch object

classmethod from_json(json_str)[source]¶

Convert json string back to ChipMatch object

CommandLine:

# FIXME; util_test is broken with classmethods
python -m ibeis.algo.hots.chip_match --test-from_json --show

Example

>>> # ENABLE_DOCTEST
>>> from ibeis.algo.hots.chip_match import *  # NOQA
>>> import ibeis
>>> cls = ChipMatch
>>> cm1, qreq_ = ibeis.testdata_cm()
>>> json_str = cm1.to_json()
>>> cm = ChipMatch.from_json(json_str)
>>> ut.quit_if_noshow()
>>> cm.score_nsum(qreq_)
>>> cm.show_single_namematch(qreq_, 1)
>>> ut.show_if_requested()

classmethod from_unscored(prior_cm, fm_list, fs_list, H_list=None, fsv_col_lbls=None)[source]¶

classmethod from_vsmany_match_tup(vmt, qaid=None, fsv_col_lbls=None)[source]¶

Parameters:	vmt (ibeis.algo.hots.pipeline.ValidMatchTup_) – valid_match_tup qaid (int) – query annotation id fsv_col_lbls (None) –
Returns:	cm
Return type:	ibeis.ChipMatch

classmethod from_vsone_match_tup(vmt_list, daid_list=None, qaid=None, fsv_col_lbls=None)[source]¶

Parameters:	vmt_list (list of ValidMatchTup_) – list of valid_match_tups qaid (int) – query annotation id fsv_col_lbls (None) –
Returns:	cm
Return type:	ibeis.ChipMatch

get_annot_fm(cm, daid)[source]¶

get_cvs_str(cm, numtop=6, ibs=None, sort=True)[source]¶

Parameters:	numtop (int) – (default = 6) ibs (IBEISController) – ibeis controller object(default = None) sort (bool) – (default = True)
Returns:	csv_str
Return type:	str

Notes

Very weird that it got a score qaid 6 vs 41 has

[72, 79, 0, 17, 6, 60, 15, 36, 63] [72, 79, 0, 17, 6, 60, 15, 36, 63] [72, 79, 0, 17, 6, 60, 15, 36, 63] [0.060, 0.053, 0.0497, 0.040, 0.016, 0, 0, 0, 0] [7, 40, 41, 86, 103, 88, 8, 101, 35]

makes very little sense

CommandLine:

python -m ibeis.algo.hots.chip_match --test-get_cvs_str --force-serial

Example

>>> # ENABLE_DOCTEST
>>> from ibeis.algo.hots.chip_match import *  # NOQA
>>> ibs, qreq_, cm_list = plh.testdata_post_sver()
>>> cm = cm_list[0]
>>> numtop = 6
>>> ibs = None
>>> sort = True
>>> csv_str = cm.get_cvs_str(numtop, ibs, sort)
>>> result = ('csv_str = \n%s' % (str(csv_str),))
>>> print(result)

get_flat_fm_info(cm, flags=None)[source]¶

Returns:	info_
Return type:	dict

CommandLine:

python -m ibeis.algo.hots.chip_match --exec-get_flat_fm_info --show

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.chip_match import *  # NOQA
>>> ibs, qreq_, cm_list = plh.testdata_pre_sver(
>>>     defaultdb='PZ_MTEST', qaid_list=[18])
>>> cm = cm_list[0]
>>> info_ = cm.get_flat_fm_info()
>>> ut.assert_all_eq(ut.lmap(len, info_.values()))
>>> result = ('info_ = %s' % (ut.repr3(info_, precision=2),))
>>> print(result)

get_fpath(cm, qreq_)[source]¶

get_fs(cm, idx=None, colx=None, daid=None, col=None)[source]¶

get_fs_list(cm, colx=None, col=None)[source]¶

get_fsv_prod_list(cm)[source]¶

get_inspect_str(cm, qreq_)[source]¶

Parameters:	qreq (QueryRequest) – query request object with hyper-parameters
Returns:	varinfo
Return type:	str

CommandLine:

python -m ibeis.algo.hots.chip_match --exec-get_inspect_str

Example

>>> # ENABLE_DOCTEST
>>> from ibeis.algo.hots.chip_match import *  # NOQA
>>> import ibeis
>>> cm, qreq_ = ibeis.testdata_cm('PZ_MTEST', a='default:dindex=0:10,qindex=0:1', t='best:SV=False')
>>> varinfo = cm.get_inspect_str(qreq_)
>>> result = ('varinfo = %s' % (str(varinfo),))
>>> print(result)

get_num_feat_score_cols(cm)[source]¶

get_rawinfostr(cm, colored=None)[source]¶

Returns:	varinfo
Return type:	str

CommandLine:

python -m ibeis.algo.hots.chip_match --exec-get_rawinfostr --show --cex

Example

>>> # ENABLE_DOCTEST
>>> from ibeis.algo.hots.chip_match import *  # NOQA
>>> import ibeis
>>> cm, qreq_ = ibeis.testdata_cm('PZ_MTEST', a='default:dindex=0:10,qindex=0:1', t='best:SV=False')
>>> varinfo = cm.get_rawinfostr()
>>> result = ('varinfo = %s' % (varinfo,))
>>> print(result)

initialize(cm, qaid=None, daid_list=None, fm_list=None, fsv_list=None, fk_list=None, score_list=None, H_list=None, fsv_col_lbls=None, dnid_list=None, qnid=None, unique_nids=None, name_score_list=None, annot_score_list=None, autoinit=True, filtnorm_aids=None, filtnorm_fxs=None)[source]¶

qaid and daid_list are not optional. fm_list and fsv_list are strongly encouraged and will probalby break things if they are not there.

inspect_difference(cm, other)[source]¶

classmethod load_from_fpath(fpath, verbose=None)[source]¶

naids_list¶

nfxs_list¶

print_csv(cm, *args, **kwargs)[source]¶

print_inspect_str(cm, qreq_)[source]¶

print_rawinfostr(cm)[source]¶

qfxs_list¶

rrr(verbose=True)¶

special class reloading function

save(cm, qreq_, verbose=None)[source]¶

shortlist_subset(cm, top_aids)[source]¶

returns a new cmtup_old with only the requested daids TODO: rectify with take_feature_matches

sortself(cm)[source]¶

reorders the internal data using cm.score_list

take_annots(cm, idx_list, inplace=False, keepscores=True)[source]¶

Example

>>> # ENABLE_DOCTEST
>>> from ibeis.algo.hots.chip_match import *  # NOQA
>>> import ibeis
>>> cm, qreq_ = ibeis.testdata_cm('PZ_MTEST',
>>>                               a='default:dindex=0:10,qindex=0:1',
>>>                               t='best:sv=false')
>>> idx_list = list(range(cm.num_daids))
>>> inplace = False
>>> keepscores = True
>>> other = out = cm.take_annots(idx_list, inplace, keepscores)
>>> result = ('out = %s' % (ut.repr2(out),))
>>> assert cm.inspect_difference(out), 'should be no difference'
>>> print(result)

Example

>>> # ENABLE_DOCTEST
>>> from ibeis.algo.hots.chip_match import *  # NOQA
>>> import ibeis
>>> cm, qreq_ = ibeis.testdata_cm('PZ_MTEST',
>>>                               a='default:dindex=0:10,qindex=0:1',
>>>                               t='best:SV=False')
>>> idx_list = [0, 2]
>>> inplace = False
>>> keepscores = True
>>> other = out = cm.take_annots(idx_list, inplace, keepscores)
>>> result = ('out = %s' % (ut.repr2(out),))
>>> print(result)

take_feature_matches(cm, indicies_list, inplace=False, keepscores=True)[source]¶

Removes outlier feature matches TODO: rectify with shortlist_subset

Parameters:	indicies_list (list) – list of lists of indicies to keep. if an item is None, the match to the corresponding daid is removed. inplace (bool) – (default = False)
Returns:	out
Return type:	ibeis.ChipMatch

CommandLine:

python -m ibeis.algo.hots.chip_match --exec-take_feature_matches --show

Example

>>> # ENABLE_DOCTEST
>>> from ibeis.algo.hots.chip_match import *  # NOQA
>>> import ibeis
>>> cm, qreq_ = ibeis.testdata_cm('PZ_MTEST', a='default:dindex=0:10,qindex=0:1', t='best:SV=False')
>>> indicies_list = [list(range(i + 1)) for i in range(cm.num_daids)]
>>> inplace = False
>>> keepscores = True
>>> out = cm.take_feature_matches(indicies_list, inplace, keepscores)
>>> assert not cm.inspect_difference(out), 'should be different'
>>> result = ('out = %s' % (ut.repr2(out),))
>>> print(result)

to_json(cm)[source]¶

Serialize ChipMatch object as JSON string

CommandLine:

python -m ibeis.algo.hots.chip_match --test-ChipMatch.to_json:0
python -m ibeis.algo.hots.chip_match --test-ChipMatch.to_json
python -m ibeis.algo.hots.chip_match --test-ChipMatch.to_json:1 --show

Example

>>> # ENABLE_DOCTEST
>>> # Simple doctest demonstrating the json format
>>> from ibeis.algo.hots.chip_match import *  # NOQA
>>> import ibeis
>>> ibs = ibeis.opendb(defaultdb='testdb1')
>>> cm, qreq_ = ibs.query_chips(1, [2, 3, 4, 5],
>>>                             return_request=True)
>>> cm.compress_top_feature_matches(num=4, rng=np.random.RandomState(0))
>>> # Serialize
>>> print('\n\nRaw ChipMatch JSON:\n')
>>> json_str = cm.to_json()
>>> print(json_str)
>>> print('\n\nPretty ChipMatch JSON:\n')
>>> # Pretty String Formatting
>>> dictrep = ut.from_json(json_str)
>>> dictrep = ut.delete_dict_keys(dictrep, [key for key, val in dictrep.items() if val is None])
>>> result  = ut.dict_str(dictrep, nl=2, precision=2, hack_liststr=True, key_order_metric='strlen')
>>> result = result.replace('u\'', '"').replace('\'', '"')
>>> print(result)

Example

>>> # ENABLE_DOCTEST
>>> # test to convert back and forth from json
>>> from ibeis.algo.hots.chip_match import *  # NOQA
>>> import ibeis
>>> cm, qreq_ = ibeis.testdata_cm()
>>> cm1 = cm
>>> # Serialize
>>> json_str = cm.to_json()
>>> print(repr(json_str))
>>> # Unserialize
>>> cm = ChipMatch.from_json(json_str)
>>> # Show if it works
>>> ut.quit_if_noshow()
>>> cm.score_nsum(qreq_)
>>> cm.show_single_namematch(qreq_, 1)
>>> ut.show_if_requested()
>>> # result = ('json_str = \n%s' % (str(json_str),))
>>> # print(result)

class ibeis.algo.hots.chip_match.MatchBaseIO[source]¶

Bases: object

copy()[source]¶

classmethod load_from_fpath(fpath, verbose=False)[source]¶

save_to_fpath(cm, fpath, verbose=False)[source]¶

CommandLine:

python ibeis --tf MatchBaseIO.save_to_fpath --verbtest --show

Example

>>> # ENABLE_DOCTEST
>>> from ibeis.algo.hots.chip_match import *  # NOQA
>>> qaid = 18
>>> ibs, qreq_, cm_list = plh.testdata_pre_sver('PZ_MTEST', qaid_list=[qaid])
>>> cm = cm_list[0]
>>> cm.score_nsum(qreq_)
>>> dpath = ut.get_app_resource_dir('ibeis')
>>> fpath = join(dpath, 'tmp_chipmatch.cPkl')
>>> ut.delete(fpath)
>>> cm.save_to_fpath(fpath)
>>> cm2 = ChipMatch.load_from_fpath(fpath)
>>> assert cm == cm2
>>> ut.quit_if_noshow()
>>> cm.ishow_analysis(qreq_)
>>> ut.show_if_requested()

exception ibeis.algo.hots.chip_match.NeedRecomputeError[source]¶

Bases: exceptions.Exception

class ibeis.algo.hots.chip_match.TestLogger(testlog, verbose=True)[source]¶

Bases: object

context(testlog, name)[source]¶

end_test(testlog)[source]¶

log_failed(testlog, msg)[source]¶

log_passed(testlog, msg)[source]¶

log_skipped(testlog, msg)[source]¶

skip_test(testlog)[source]¶

start_test(testlog, name)[source]¶

ibeis.algo.hots.chip_match.check_arrs_eq(arr1, arr2)[source]¶

ibeis.algo.hots.chip_match.extend_nplists(x_list, num, shape, dtype)[source]¶

ibeis.algo.hots.chip_match.extend_nplists_(x_list, num, shape, dtype)[source]¶

ibeis.algo.hots.chip_match.extend_pylist(x_list, num, val)[source]¶

ibeis.algo.hots.chip_match.extend_pylist_(x_list, num, val)[source]¶

ibeis.algo.hots.chip_match.extend_scores(vals, num)[source]¶

ibeis.algo.hots.chip_match.filtnorm_op(filtnorm_, op_, *args, **kwargs)[source]¶

ibeis.algo.hots.chip_match.get_chipmatch_fname(qaid, qreq_, qauuid=None, cfgstr=None, TRUNCATE_UUIDS=False, MAX_FNAME_LEN=200)[source]¶

CommandLine:

python -m ibeis.algo.hots.chip_match --test-get_chipmatch_fname

Example

>>> # ENABLE_DOCTEST
>>> from ibeis.algo.hots.chip_match import *  # NOQA
>>> qreq_, args = plh.testdata_pre('spatial_verification',
>>>                                defaultdb='PZ_MTEST', qaid_override=[18],
>>>                                p='default:sqrd_dist_on=True')
>>> cm_list = args.cm_list_FILT
>>> cm = cm_list[0]
>>> fname = get_chipmatch_fname(cm.qaid, qreq_, qauuid=None,
>>>                             TRUNCATE_UUIDS=False, MAX_FNAME_LEN=200)
>>> result = fname
>>> print(result)

qaid=18_cm_cvgrsbnffsgifyom_quuid=a126d459-b730-573e-7a21-92894b016565.cPkl

ibeis.algo.hots.chip_match.prepare_dict_uuids(class_dict, ibs)[source]¶

Hacks to ensure proper uuid conversion

ibeis.algo.hots.chip_match.safe_check_lens_eq(None, 1)[source]¶

safe_check_lens_eq([3], [2, 4])

ibeis.algo.hots.chip_match.safe_check_nested_lens_eq(None, 1)[source]¶

safe_check_nested_lens_eq([[3, 4]], [[2, 4]]) safe_check_nested_lens_eq([[1, 2, 3], [1, 2]], [[1, 2, 3], [1, 2]]) safe_check_nested_lens_eq([[1, 2, 3], [1, 2]], [[1, 2, 3], [1]])

ibeis.algo.hots.chip_match.safeop(op_, xs, *args, **kwargs)[source]¶

ibeis.algo.hots.chip_match.testdata_cm()[source]¶

ibeis.algo.hots.crf module¶

ibeis.algo.hots.crf.chain_crf()[source]¶

ibeis.algo.hots.crf.crftest()[source]¶

pip install pyqpbo pip install pystruct

http://taku910.github.io/crfpp/#install

cd ~/tmp #wget https://drive.google.com/folderview?id=0B4y35FiV1wh7fngteFhHQUN2Y1B5eUJBNHZUemJYQV9VWlBUb3JlX0xBdWVZTWtSbVBneU0&usp=drive_web#list 7z x CRF++-0.58.tar.gz 7z x CRF++-0.58.tar cd CRF++-0.58 chmod +x configure ./configure make

ibeis.algo.hots.demobayes module¶

ibeis.algo.hots.demobayes.classify_k(cfg={})[source]¶

CommandLine:

python -m ibeis.algo.hots.demobayes --exec-classify_k --show --ev :nA=3
python -m ibeis.algo.hots.demobayes --exec-classify_k --show --ev :nA=3,k=1
python -m ibeis.algo.hots.demobayes --exec-classify_k --show --ev :nA=3,k=0 --method=approx
python -m ibeis.algo.hots.demobayes --exec-classify_k --show --ev :nA=10,k=1 --method=approx

Example

>>> from ibeis.algo.hots.demobayes import *  # NOQA
>>> cfg_list = testdata_demo_cfgs()
>>> classify_k(cfg_list[0])
>>> ut.show_if_requested()

ibeis.algo.hots.demobayes.classify_one_new_unknown()[source]¶

Make a model that knows who the previous annots are and tries to classify a new annot

CommandLine:

python -m ibeis.algo.hots.demobayes --exec-classify_one_new_unknown --verbose
python -m ibeis.algo.hots.demobayes --exec-classify_one_new_unknown --show --verbose --present
python3 -m ibeis.algo.hots.demobayes --exec-classify_one_new_unknown --verbose
python3 -m ibeis.algo.hots.demobayes --exec-classify_one_new_unknown --verbose --diskshow --verbose --present --save demo5.png --dpath . --figsize=20,10 --dpi=128 --clipwhite

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.demobayes import *  # NOQA
>>> result = classify_one_new_unknown()
>>> ut.show_if_requested()

ibeis.algo.hots.demobayes.demo_ambiguity()[source]¶

Test what happens when an annotation need to choose between one of two names

CommandLine:

python -m ibeis.algo.hots.demobayes --exec-demo_ambiguity --show --verbose --present

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.demobayes import *  # NOQA
>>> result = demo_ambiguity()
>>> ut.show_if_requested()

ibeis.algo.hots.demobayes.demo_annot_idependence_overlap()[source]¶

Given:

• an unknown annotation d
• three annots with the same name (Fred) a, b, and c
• a and b are near duplicates
• (a and c) / (b and c) are novel views

Goal:

• If d matches to a and b the probably that d is Fred should not be much more than if d matched only a or only b.

• The probability that d is Fred given it matches to any of the 3 annots

  alone should be equal

  P(d is Fred | Mad=1) = P(d is Fred | Mbd=1) = P(d is Fred | Mcd=1)

• The probability that d is fred given two matches to any of those two annots should be greater than the probability given only one.

  P(d is Fred | Mad=1, Mbd=1) > P(d is Fred | Mad=1) P(d is Fred | Mad=1, Mcd=1) > P(d is Fred | Mad=1)

• The probability that d is fred given matches to two near duplicate matches should be less than if d matches two non-duplicate matches.

  P(d is Fred | Mad=1, Mcd=1) > P(d is Fred | Mad=1, Mbd=1)

• The probability that d is fred given two near duplicates should be only epsilon greater than a match to either one individually.

  P(d is Fred | Mad=1, Mbd=1) = P(d is Fred | Mad=1) + epsilon

Method:

We need to model the fact that there are other causes that create the effect of a high score. Namely, near duplicates. This can be done by adding an extra conditional that score depends on if they match as well as if they are near duplicates.

P(S_ij | Mij) –> P(S_ij | Mij, Dij)

where

Dij is a random variable indicating if the image is a near duplicate.

We can model this as an independant variable

P(Dij) = {True: .5, False: .5}

or as depending on if the names match.

P(Dij | Mij) = {‘same’: {True: .5, False: .5} diff: {True: 0, False 1}}

CommandLine:

python -m ibeis.algo.hots.demobayes --exec-demo_annot_idependence_overlap --verbose --present --show

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.demobayes import *  # NOQA
>>> result = demo_annot_idependence_overlap()
>>> ut.show_if_requested()

ibeis.algo.hots.demobayes.demo_bayesnet(cfg={})[source]¶

Make a model that knows who the previous annots are and tries to classify a new annot

CommandLine:

python -m ibeis --tf demo_bayesnet --diskshow --verbose --save demo4.png --dpath . --figsize=20,10 --dpi=128 --clipwhite

python -m ibeis --tf demo_bayesnet --ev :nA=3,Sab=0,Sac=0,Sbc=1
python -m ibeis --tf demo_bayesnet --ev :nA=4,Sab=0,Sac=0,Sbc=1,Sbd=1 --show
python -m ibeis --tf demo_bayesnet --ev :nA=4,Sab=0,Sac=0,Sbc=1,Scd=1 --show
python -m ibeis --tf demo_bayesnet --ev :nA=4,Sab=0,Sac=0,Sbc=1,Sbd=1,Scd=1 --show

python -m ibeis --tf demo_bayesnet --ev :nA=3,Sab=0,Sac=0,Sbc=1
python -m ibeis --tf demo_bayesnet --ev :nA=5,rand_scores=True --show

python -m ibeis --tf demo_bayesnet --ev :nA=4,nS=3,rand_scores=True --show --verbose
python -m ibeis --tf demo_bayesnet --ev :nA=5,nS=2,Na=fred,rand_scores=True --show --verbose
python -m ibeis --tf demo_bayesnet --ev :nA=5,nS=5,Na=fred,rand_scores=True --show --verbose
python -m ibeis --tf demo_bayesnet --ev :nA=4,nS=2,Na=fred,rand_scores=True --show --verbose

python -m ibeis.algo.hots.demobayes --exec-demo_bayesnet \
        --ev =:nA=4,Sab=0,Sac=0,Sbc=1 \
        :Sbd=1 :Scd=1 :Sbd=1,Scd=1 :Sbd=1,Scd=1,Sad=0 \
        --show --present

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.demobayes import *  # NOQA
>>> cfg_list = testdata_demo_cfgs()
>>> print('cfg_list = %r' % (cfg_list,))
>>> for cfg in cfg_list:
>>>     demo_bayesnet(cfg)
>>> ut.show_if_requested()

ibeis.algo.hots.demobayes.demo_conflicting_evidence()[source]¶

Notice that the number of annotations in the graph does not affect the probability of names.

ibeis.algo.hots.demobayes.demo_model_idependencies()[source]¶

Independences of the 3 annot 3 name model

CommandLine:

python -m ibeis.algo.hots.demobayes --exec-demo_model_idependencies --mode=1 --num-names=2 --show
python -m ibeis.algo.hots.demobayes --exec-demo_model_idependencies --mode=2

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.demobayes import *  # NOQA
>>> result = demo_model_idependencies()
>>> print(result)
>>> ut.show_if_requested()

ibeis.algo.hots.demobayes.demo_modes()[source]¶

Look at the last result of the different names demo under differet modes

ibeis.algo.hots.demobayes.demo_name_annot_complexity()[source]¶

This demo is meant to show the structure of the graph as more annotations and names are added.

CommandLine:

python -m ibeis.algo.hots.demobayes --exec-demo_name_annot_complexity --show

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.demobayes import *  # NOQA
>>> demo_name_annot_complexity()
>>> ut.show_if_requested()

ibeis.algo.hots.demobayes.demo_single_add()[source]¶

This demo shows how a name is assigned to a new annotation.

CommandLine:

python -m ibeis.algo.hots.demobayes --exec-demo_single_add --show --present --mode=1

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.demobayes import *  # NOQA
>>> demo_single_add()
>>> ut.show_if_requested()

ibeis.algo.hots.demobayes.demo_structure()[source]¶

CommandLine:

python -m ibeis.algo.hots.demobayes --exec-demo_structure --show

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.demobayes import *  # NOQA
>>> result = demo_structure()
>>> ut.show_if_requested()

ibeis.algo.hots.demobayes.get_toy_annots(num_annots, num_names=None, initial_aids=None, initial_nids=None, nid_sequence=None, seed=None)[source]¶

Parameters:	num_annots (int) – num_names (int) – (default = None) initial_aids (None) – (default = None) initial_nids (None) – (default = None) nid_sequence (None) – (default = None) seed (None) – (default = None)
Returns:	(aids, nids, aids1, nids1, all_aids, all_nids)
Return type:	tuple

CommandLine:

python -m ibeis.algo.hots.demobayes --exec-get_toy_annots

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.demobayes import *  # NOQA
>>> num_annots = 1
>>> num_names = 5
>>> initial_aids = np.array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9], dtype=np.int64)
>>> initial_nids = np.array([0, 0, 1, 2, 2, 1, 1, 1, 2, 3], dtype=np.int64)
>>> nid_sequence = np.array([0, 0, 1, 2, 2, 1, 1], dtype=np.int64)
>>> seed = 0
>>> (aids, nids, aids1, nids1, all_aids, all_nids) = get_toy_annots(num_annots, num_names, initial_aids, initial_nids, nid_sequence, seed)
>>> result = ('(aids, nids, aids1, nids1, all_aids, all_nids) = %s' % (ut.repr2((aids, nids, aids1, nids1, all_aids, all_nids), nl=1),))
>>> print(result)

ibeis.algo.hots.demobayes.get_toy_data_1v1(num_annots=5, num_names=None, **kwargs)[source]¶

CommandLine:

python -m ibeis.algo.hots.demobayes --exec-get_toy_data_1v1 --show

Example

>>> # ENABLE_DOCTEST
>>> from ibeis.algo.hots.demobayes import *  # NOQA
>>> toy_data = get_toy_data_1v1()
>>> ut.quit_if_noshow()
>>> import plottool as pt
>>> show_toy_distributions(toy_data['toy_params'])
>>> ut.show_if_requested()

Example1:

>>> # ENABLE_DOCTEST
>>> from ibeis.algo.hots.demobayes import *  # NOQA
>>> toy_data = get_toy_data_1v1()
>>> kwargs = {}
>>> initial_aids = toy_data['aids']
>>> initial_nids = toy_data['nids']
>>> num_annots = 1
>>> num_names = 6
>>> toy_data2 = get_toy_data_1v1(num_annots, num_names, initial_aids=initial_aids, initial_nids=initial_nids)
>>> ut.quit_if_noshow()
>>> import plottool as pt
>>> show_toy_distributions(toy_data['toy_params'])
>>> ut.show_if_requested()

ibeis.algo.hots.demobayes.get_toy_data_1vM(num_annots, num_names=None, **kwargs)[source]¶

Parameters:	num_annots (int) – num_names (int) – (default = None)

Kwargs:

initial_aids, initial_nids, nid_sequence, seed

Returns:	(pair_list, feat_list)
Return type:	tuple

CommandLine:

python -m ibeis.algo.hots.demobayes --exec-get_toy_data_1vM --show

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.demobayes import *  # NOQA
>>> num_annots = 1000
>>> num_names = 40
>>> get_toy_data_1vM(num_annots, num_names)
>>> ut.quit_if_noshow()
>>> import plottool as pt
>>> ut.show_if_requested()

ibeis.algo.hots.demobayes.learn_prob_score(num_scores=5, pad=55, ret_enc=False, use_cache=None)[source]¶

Parameters:	num_scores (int) – (default = 5)
Returns:	(discr_domain, discr_p_same)
Return type:	tuple

CommandLine:

python -m ibeis.algo.hots.demobayes --exec-learn_prob_score --show

Example

>>> # ENABLE_DOCTEST
>>> from ibeis.algo.hots.demobayes import *  # NOQA
>>> num_scores = 2
>>> (discr_domain, discr_p_same, encoder) = learn_prob_score(num_scores, ret_enc=True, use_cache=False)
>>> print('discr_p_same = %r' % (discr_p_same,))
>>> ut.quit_if_noshow()
>>> import plottool as pt
>>> encoder.visualize()
>>> ut.show_if_requested()

ibeis.algo.hots.demobayes.make_bayes_notebook()[source]¶

CommandLine:

python -m ibeis.algo.hots.demobayes --exec-make_bayes_notebook

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.demobayes import *  # NOQA
>>> result = make_bayes_notebook()
>>> print(result)

ibeis.algo.hots.demobayes.show_model_templates()[source]¶

CommandLine:

python -m ibeis.algo.hots.demobayes --exec-show_model_templates

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.demobayes import *  # NOQA
>>> result = show_model_templates()
>>> ut.show_if_requested()

ibeis.algo.hots.demobayes.show_toy_distributions(toy_params)[source]¶

ibeis.algo.hots.demobayes.test_triangle_property()[source]¶

CommandLine:

python -m ibeis.algo.hots.demobayes --exec-test_triangle_property --show

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.demobayes import *  # NOQA
>>> result = test_triangle_property()
>>> ut.show_if_requested()

ibeis.algo.hots.demobayes.testdata_demo_cfgs()[source]¶

ibeis.algo.hots.devcases module¶

development module storing my “development state”

TODO:

• figure out what packages I use have lisencing issues.

  • Reimplement them or work around them.

Excplitict Negative Matches between chips

ibeis.algo.hots.devcases.find_close_incorrect_match(ibs, qaids)[source]¶

ibeis.algo.hots.devcases.fix_pz_master()[source]¶

CommandLine:

python -m ibeis.algo.hots.devcases --test-fix_pz_master --show

Example

>>> # DISABLE_DOCTEST
>>> from ibeis.algo.hots.devcases import *  # NOQA
>>> # build test data
>>> # execute function
>>> result = fix_pz_master()
>>> # verify results
>>> print(result)

ibeis.algo.hots.devcases.get_dev_test_fpaths(index)[source]¶

ibeis.algo.hots.devcases.get_gzall_small_test()[source]¶

ibs.get_annot_visual_uuids([qaid, aid])

ibeis.algo.hots.devcases.get_pz_master_testcase()[source]¶