Commit 8868581a authored by Łukasz Kopociński's avatar Łukasz Kopociński

Generate spert jsons

parent 102ba87b
md5: 61a641eba25cc1edbe294207a2a32af8
cmd: ./spert/scripts/generate_indices.py --dataset-keys ./semrel/data/data/vectors/elmo.rel.keys
--output-path ./spert/data/indices.json
deps:
- md5: 03e061fa821289f78bb75cac93982ad8.dir
path: semrel/data/data
- md5: 9fa4a0aee79e27b64e60367e27bf7f23.dir
path: semrel/data/data/vectors
- md5: f19baa48b871341efdf5baa97020a4f9
path: spert/scripts/generate_indices.py
outs:
- md5: 6dc4f745f4ef13bd71c2062a11fead73
path: spert/data/indices.json
cache: true
metric: false
persist: false
md5: fe081dae76b84b3c09bbc44d4b5031fb
cmd: ./spert/scripts/generate_spert_json.py --input-path ./semrel/data/data/relations/relations.tsv
--indices-file ./spert/data/indices.json --output-dir ./spert/data/dataset
deps:
- md5: 9f143991d477536ee442ceea00338dc3
path: semrel/data/data/relations/relations.tsv
- md5: 6dc4f745f4ef13bd71c2062a11fead73
path: spert/data/indices.json
- md5: f95cef92e3bad810780c5de4a718d6bf
path: spert/scripts/generate_spert_json.py
outs:
- md5: d729f3e4d6bfe5ba6dae8b06a53685a0.dir
path: spert/data/dataset
cache: true
metric: false
persist: false
......@@ -2,7 +2,7 @@ IN_RELATION_LABEL = 'in_relation'
NO_RELATION_LABEL = 'no_relation'
BRAND_NAME_KEY = 'BRAND_NAME'
PRODUCT_NAME_KEY = 'BRAND_NAME'
PRODUCT_NAME_KEY = 'PRODUCT_NAME'
CHANNELS = ((BRAND_NAME_KEY, PRODUCT_NAME_KEY),
(PRODUCT_NAME_KEY, BRAND_NAME_KEY))
......
......@@ -37,7 +37,7 @@ def main(dataset_keys, output_path):
in_domain = params.get(IN_DOMAIN_KEY)
lexical_split = params.get(LEXICAL_SPLIT_KEY, False)
(train, valid, test), keys = get_indices(
train, valid, test, keys = get_indices(
keys_file=Path(dataset_keys),
balanced=True,
lexical_split=lexical_split,
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment