Knowledge-Graph-Hub/neat-ml

Issues

• Generate embeddings from Monarch graph for use in Exomiser

  #100 opened 2 years ago by justaddcoffee
  0

• Output link predictions as SSSOM

  #90 opened 3 years ago by caufieldjh
  1

• Drop support for Tensorflow and scikit-learn methods

  #98 opened 2 years ago by caufieldjh
  1

• RuntimeError when validation fails isn't very informative

  #94 opened 2 years ago by caufieldjh
  3

• Create docs

  #71 opened 3 years ago by caufieldjh
  0

• Create landing page

  #79 opened 3 years ago by caufieldjh
  0

• Include line number in schema validation errors

  #96 opened 2 years ago by caufieldjh
  0

• Update to use most recent `grape` functions for link prediction

  #97 opened 2 years ago by caufieldjh
  0

• Upload to PyPi

  #69 opened 2 years ago by caufieldjh
  1

• Convert edge prediction methods to grape methods

  #85 opened 2 years ago by caufieldjh
  0

• Set up methods for using Ensmallen/Embiggen native methods

  #77 opened 2 years ago by caufieldjh
  0

• Set up test workflow to use `tox`

  #87 opened 2 years ago by caufieldjh
  1

• `poetry`fying the project

  #83 opened 2 years ago by hrshdhgd
  1

• Missing key in defining metrics in the config

  #89 opened 2 years ago by caufieldjh
  0

• Update for Ensmallen / Embiggen -> grape

  #81 opened 3 years ago by caufieldjh
  0

• Need to get history output for embeddings and classifiers

  #84 opened 3 years ago by caufieldjh
  0

• Use LinkML to parse and validate NEAT YAML files

  #52 opened 3 years ago by justaddcoffee
  3

• In link prediction, filter nodes by prefix or other slots

  #68 opened 3 years ago by caufieldjh
  4

• Example YAMLs should use different value for loading node types

  #54 opened 3 years ago by caufieldjh
  1

• Move `graph_path` argument out of `graph_data` arguments

  #76 opened 3 years ago by caufieldjh
  1

• Scheduled NEAT runs fail due to extra `graph_path` keyword

  #73 opened 3 years ago by caufieldjh
  1

• Incorporate GNNs

  #70 opened 3 years ago by caufieldjh
  0

• Loading from URL looks for wrong filename

  #65 opened 3 years ago by caufieldjh
  0

• OSError: [Errno 9] Bad file descriptor

  #60 opened 3 years ago by caufieldjh
  2

• Specify compressed node/edge file in config yaml

  #43 opened 3 years ago by caufieldjh
  0

• Allow config YAML to not contain graph files

  #64 opened 3 years ago by caufieldjh
  0

• Enable yaml helper to load embeddings from URLs

  #62 opened 3 years ago by caufieldjh
  0

• Add quickstart notebook

  #57 opened 3 years ago by caufieldjh
  1

• Display errors to users at conclusion of NEAT run

  #59 opened 3 years ago by caufieldjh
  1

• Replace NEATs BERT text embedding stuff with Ensmallen's (better) version of this

  #45 opened 3 years ago by justaddcoffee
  1

• Consider incorporating model cards to better describe and benchmark trained ML model

  #58 opened 3 years ago by justaddcoffee
  0

• pre_run_checks should catch `botocore.exceptions.NoCredentialsError`

  #55 opened 3 years ago by caufieldjh
  1

• Create index.html upon uploading new output

  #51 opened 3 years ago by caufieldjh
  2

• support link prediction using classifiers after training

  #19 opened 4 years ago by justaddcoffee
  0

• Fail gracefully when provided node/edge path/URL that isn't node or edge

  #50 opened 3 years ago by caufieldjh
  0

• Use URLs as node and edge list locations

  #36 opened 3 years ago by caufieldjh
  3

• NEAT should check S3 credentials at beginning of run to avoid credential error at the end of a long run

  #26 opened 3 years ago by justaddcoffee
  2

• Requirement management

  #32 opened 3 years ago by caufieldjh
  3

• history.to_json() is None

  #35 opened 3 years ago by caufieldjh
  1

• Input 0 of layer "sequential" is incompatible with the layer

  #33 opened 3 years ago by caufieldjh
  1

• output metrics after classifier training in JSON format

  #22 opened 4 years ago by justaddcoffee
  3

• support pushing results to AWS bucket

  #17 opened 4 years ago by justaddcoffee
  1

• support creating positive/negative test/train splits

  #16 opened 4 years ago by justaddcoffee
  1

• Change YAML format to include input_data_dir and output_data_dir

  #14 opened 4 years ago by justaddcoffee
  2

• error when not specifying validation sets?

  #13 opened 4 years ago by realmarcin
  4

• Graph inputs should be full graph not splits, rename example paths

  #3 opened 4 years ago by realmarcin
  1

• add parameters for NEAT embedding output file naming, format, and ensure that node identifiers part of any embedding output

  #5 opened 4 years ago by realmarcin
  1

• some way to handle singleton nodes in graph input and resulting embeddings

  #4 opened 4 years ago by realmarcin
  1

• add parameters for NEAT predictive model intermediates and outputs

  #6 opened 4 years ago by realmarcin
  1