Place all required files in a directory on the host. Then run Patapsco:
docker run -v /home/user/my_data:/data --rm -i hltcoe/patapsco /data/eng/eng_basic.yml
It is best if the paths in the YAML configuration file be absolute paths. They should use the container path of the mounted volume.
Example:
run:
name: English basic run
path: /data/runs/eng-basic-run
text:
normalize:
lowercase: true
tokenize: whitespace
stem: porter
documents:
input:
format: json
lang: eng
encoding: utf8
path: /data/eng/eng_mini_docs.jsonl
process:
inherit: text
database:
name: sqlite
index:
name: lucene
topics:
input:
format: json
lang: eng
encoding: utf8
path: /data/eng/eng_mini_topics.jsonl
fields: title
queries:
process:
inherit: text
retrieve:
name: bm25
number: 5
score:
input:
path: /data/eng/eng_mini_qrels
This image has the spacy models for English and the multi-language model used by moses for sentence segmentation. If other languages are desired, it is best to inherit from this image and install those models in the new image.
Content type
Image
Digest
Size
7.2 GB
Last updated
almost 5 years ago
docker pull hltcoe/patapsco