angelolab / segmentation

61%

master: 81%

LAST BUILD ON BRANCH synthetic_data_expr_mat

branch: synthetic_data_expr_mat

CHANGE BRANCH

x

Reset

• synthetic_data_expr_mat
• master
• pep8
• plotting_tweaks
• watershed_cleanup
• Spatial_Analysis
• watershed_bugfix
• merge_conflict
• io_utils
• NewTestEdits
• working
• CellLabelSpatial
• phenotype_analysis
• extract_single_cell_bug
• simplified_saving
• refactor_spatial
• synthetic_data
• tif-label-masks
• segmentation_test
• NeighborhoodAnslysis
• plotting_functions
• docker_build_update
• subcellular_segmentation_update
• synthetic_data_p2
• subcellular_update_2
• synthetic_data_p1_integrate
• bad-path-revision
• deepcell-format-update
• coveralls_troubleshooting
• travis_export
• filename-fix
• readme-clarification
• dimensionality_reduction
• jupyter_abstract
• code_cleanup
• io-utils-update
• preprocessing
• signal_extract_v1
• issue_template
• closenum-speedup
• visualization_example
• code_cleanup_2
• repo_rename

Build # 646

Build Type

push

travis-ci-com

Committed by web-flow

Commit Message

Integrating synthetic data generation with spatial testing functions (#126)

* Remove unnecessary comment, mostly for creating a new PR

* Reorganize functions so generate_random_cell_shapes can easily integrate with existing code

* Add code to generate ellipses and visualize them in a format compatible with Ellipse in matplotlib.patches

* PEP8 fix in visualize

* matplotlib fixes

* Add a TODO to describe process

* PEP8 fix and add diagonal check to ensure proper distance matrix

* Fix Python style to include tuple arguments instead of dict arguments

* Fix distr_AB to include a lower mean

* Fix - error, needed to be 0

* Add first pass of integration

* Commit minor fixes

* Add random seed for randomized spatial operations, correct distribution parameters for synthetic data generation

* Update parameters of direct_init_dist_matrix to pass test

* Add cases for point and direct dist mat initialization and fix PEP8 error

* Both direct and point-based random matrix initialization passes for testing purposes of spatial analysis

* Those darn PEP8 checks

* Final PEP8 checks

* Fix error in generate_random_centroids to handle regeneration in the case of duplicate and if the seed is not set

* Make changes as requested by Noah and Adam: remove pointwise dist matrix tests, change parameter passing, and get rid of while loop regeneration of centroid points

* Remove all random seed initialization and non-related code, as well as update docs

* Remove random seeding from spatial_analysis_utils

* More and more leftover random seed removal

* Remove another instance of a random seed in a test function

* Update seed doc in direct_init_dist_matrix

* Change distribution of dist_mat_neg and calculate_enrichment_stats to reflect strict inferiority of A-B to A-C, also fix make_distance_matrix to be more robust to different dist_lims

* Fix parameter passing issues in spatial_analysis_test

* Fix more di... (continued)

Run Details

1119 of 1846 relevant lines covered (60.62%)

0.61 hits per line