NickCrews/mismo

Issues

• Unable to use 'cluster.connected_components' on a pyspark dataframe

  #53 opened 3 months ago by jstammers
  2

• Add active learning API for predicate-based blocking and matching

  #54 opened 3 months ago by jstammers
  2

• Incorrect join on large tables for add_tfidf

  #50 opened 4 months ago by jstammers
  4

• DuckDb ConversionException when running MinHashLSH example

  #45 opened 4 months ago by lmores
  6

• Testing: Standardized workflow and datasets for speed and performance benchmarking

  #19 opened 4 months ago by OlivierBinette
  3

• Address parsing performance

  #47 opened 4 months ago by lmores
  4

• viz: Plot pair scores using manifold learning

  #3 opened a year ago by NickCrews
  0

• Add ability to sample from blocked pairs when training an FS model

  #44 opened 5 months ago by jstammers
  3

• Poor scaling of add_tfidf to larger datasets

  #42 opened 6 months ago by jstammers
  7

• Inefficient Sampling From Known Labels

  #35 opened 6 months ago by jstammers
  17

• expose liepzig affiliations dataset

  #41 opened 6 months ago by NickCrews
  0

• Incremental clustering

  #36 opened 7 months ago by lmores
  1

• benchmarks for array.filter(x -> x.isin(<column from other relation>))

  #32 opened 8 months ago by NickCrews
  1

• Add more example datasets

  #13 opened a year ago by NickCrews
  4

• Add TF-IDF comparer based on sklearn

  #31 opened 8 months ago by NickCrews
  0

• feat: test on spark using docker

  #27 opened 9 months ago by NickCrews
  1

• joining on arrays is slow

  #29 opened 9 months ago by NickCrews
  3

• explore ipydatagrid for showing data

  #28 opened 9 months ago by NickCrews
  0

• Consider supporting latent-entity based algorithms

  #26 opened 9 months ago by NickCrews
  0

• feat: plot clusters

  #22 opened 10 months ago by NickCrews
  1

• clustering: sillouette, rand index, adjusted rand index, Mutual Information, etc. reference

  #6 opened 10 months ago by NickCrews
  1

• Support set-wise comparison and pooling

  #1 opened 10 months ago by NickCrews
  1

• ROADMAP

  #4 opened a year ago by NickCrews
  0

• better connected_components() API

  #5 opened 10 months ago by NickCrews
  1

• Testing: test_fs is too computationally and memory intensive

  #21 opened 10 months ago by OlivierBinette
  8

• Testing: Refactor some of the tests to facilitate test-driven development and modularity

  #17 opened a year ago by OlivierBinette
  2

• Why deal with left and right tables?

  #20 opened a year ago by OlivierBinette
  4

• Design: Should type aliases be used for Ibis types?

  #16 opened a year ago by OlivierBinette
  2

• FEAT: Implement Pipelines?

  #18 opened a year ago by OlivierBinette
  2

• Add usage note on metrics

  #12 opened a year ago by NickCrews
  1

• Block using KDTree

  #15 opened a year ago by NickCrews
  0

• Eval: look into new cluster eval metric

  #14 opened a year ago by NickCrews
  0

• Assess link quality via comparison of links vs non-links

  #11 opened a year ago by NickCrews
  0

• Assess link quality via sensitivity analysis

  #10 opened a year ago by NickCrews
  0

• viz: plot blocking with https://upset.app/

  #7 opened a year ago by NickCrews
  1

• EM algorithm of FS model for unlabeled data

  #9 opened a year ago by NickCrews
  0

• FEAT: wizard that checks things for you

  #8 opened a year ago by NickCrews
  0

• Consider using DuckDB for SQL operations

  #2 opened a year ago by NickCrews
  1