Running a job keeps appending fbk15k-237 to my entity file ids even if I try to manually override it

[JothamWong]

Hi,

Here is my config.yaml file

dataset:
name: wn18
num_entities: 40943
num_relations: 18
files:
train:
filename: train.del
type: triples
valid:
filename: valid.del
type: triples
entity_ids:
filename: entity_ids.del
type: map
relation_ids:
filename: relation_ids.del
type: map
entity_strings:
filename: entity_strings.del
type: map
relation_strings:
filename: relation_ids.del
entity_ranking:
metrics_per:
relation_type: true
eval:
batch_size: 256
trace_level: example
import:

• rescal
  lookup_embedder:
  dim: 256
  initialize: xavier_uniform_
  initialize_args:
  normal_:
  mean: 0.0
  std: 0.0027452953967162064
  uniform_:
  a: -0.7461025531688481
  xavier_normal_:
  gain: 1.0
  xavier_uniform_:
  gain: 1.0
  model: rescal
  negative_sampling:
  implementation: batch
  num_samples:
  p: -1
  rescal:
  entity_embedder:
  dropout: 0.46531374032890227
  regularize_weight: 5.116940665766654e-09
  relation_embedder:
  dropout: 0.38594622673484935
  regularize_weight: 2.5263177120255504e-08
  train:
  auto_correct: true
  batch_size: 256
  loss_arg: 1.0
  lr_scheduler: ReduceLROnPlateau
  lr_scheduler_args:
  factor: 0.95
  mode: max
  patience: 6
  threshold: 0.0001
  max_epochs: 400
  optimizer_args:
  lr: 0.164019723382278
  type: 1vsAll
  valid:
  early_stopping:
  min_threshold:
  epochs: 50
  metric_value: 0.05
  patience: 10

Even though I have manually set the dataset to wn18, the entity_ids keep getting fbk15k-237 appended to them which causes an error. I don't understand why this is happening.

Here is the error code

OSError: File d:\jetbrains\pycharmprojects\knowledgeinjection\data\kge\data\wn18\fbk15k-237-entity_ids.del for key entity_ids could not be found

[JothamWong]

Even running the default argument which uses the toy complex search results in the exact same issue.

[AdrianKs]

Hi, could you please post your dataset.yaml which is localted in the datafolder of wn18, additionally. This will help to debug this problem.

[JothamWong]

Hello, here is the dataset.yaml located in my wn18 folder

dataset:
files.entity_ids.filename: entity_ids.del
files.entity_ids.type: map
files.entity_strings.filename: entity_strings.del
files.entity_strings.type: idmap
files.relation_ids.filename: relation_ids.del
files.relation_ids.type: map
files.test.filename: test.del
files.test.size: 5000
files.test.split_type: test
files.test.type: triples
files.test_without_unseen.filename: test_without_unseen.del
files.test_without_unseen.size: 5000
files.test_without_unseen.split_type: test
files.test_without_unseen.type: triples
files.train.filename: train.del
files.train.size: 141442
files.train.split_type: train
files.train.type: triples
files.train_sample.filename: train_sample.del
files.train_sample.size: 5000
files.train_sample.split_type: train
files.train_sample.type: triples
files.valid.filename: valid.del
files.valid.size: 5000
files.valid.split_type: valid
files.valid.type: triples
files.valid_without_unseen.filename: valid_without_unseen.del
files.valid_without_unseen.size: 5000
files.valid_without_unseen.split_type: valid
files.valid_without_unseen.type: triples
name: wn18
num_entities: 40943
num_relations: 18