FintoAI-data-YKL / dvc.yaml
juhoinkinen's picture
Upload folder using huggingface_hub
0179b9a verified
raw
history blame
3.82 kB
stages:
install:
cmd: python3 -m venv venv && . venv/bin/activate && pip install -U pip wheel setuptools && pip install -r requirements.txt && cp requirements.txt venv-installed
deps:
- requirements.txt
outs:
- venv-installed:
cache: false
load-vocab:
cmd: annif load-vocab ykl corpora/ykl-skos.ttl
deps:
- venv-installed
- corpora/ykl-skos.ttl
outs:
- data/vocabs/ykl
train-omikuji-parabel:
foreach:
- lang2: fi
lang3: fin
- lang2: sv
lang3: swe
- lang2: en
lang3: eng
do:
cmd: annif train ykl-omikuji-parabel-${item.lang2} -j 8 corpora/kirjaesittelyt2021/kvesit-ykl-${item.lang3}-train.tsv
params:
- projects.toml:
- ykl-omikuji-parabel-${item.lang2}
deps:
- venv-installed
- corpora/kirjaesittelyt2021/kvesit-ykl-${item.lang3}-train.tsv
- data/vocabs/ykl
outs:
- data/projects/ykl-omikuji-parabel-${item.lang2}
eval-omikuji-parabel:
foreach:
- lang2: fi
lang3: fin
- lang2: sv
lang3: swe
- lang2: en
lang3: eng
do:
cmd: annif eval ykl-omikuji-parabel-${item.lang2} -j 8 -m Precision@1 -m NDCG
--metrics-file reports/test-omikuji-parabel-${item.lang2}.json
--results-file reports/test-omikuji-parabel-${item.lang2}.csv
corpora/kirjaesittelyt2021/kvesit-ykl-${item.lang3}-test.tsv
deps:
- venv-installed
- corpora/kirjaesittelyt2021/kvesit-ykl-${item.lang3}-test.tsv
- data/projects/ykl-omikuji-parabel-${item.lang2}
outs:
- reports/test-omikuji-parabel-${item.lang2}.csv:
cache: false
metrics:
- reports/test-omikuji-parabel-${item.lang2}.json:
cache: false
train-omikuji-bonsai:
foreach:
- lang2: fi
lang3: fin
- lang2: sv
lang3: swe
- lang2: en
lang3: eng
do:
cmd: annif train ykl-omikuji-bonsai-${item.lang2} -j 8 corpora/kirjaesittelyt2021/kvesit-ykl-${item.lang3}-train.tsv
params:
- projects.toml:
- ykl-omikuji-bonsai-${item.lang2}
deps:
- venv-installed
- corpora/kirjaesittelyt2021/kvesit-ykl-${item.lang3}-train.tsv
- data/vocabs/ykl
outs:
- data/projects/ykl-omikuji-bonsai-${item.lang2}
eval-omikuji-bonsai:
foreach:
- lang2: fi
lang3: fin
- lang2: sv
lang3: swe
- lang2: en
lang3: eng
do:
cmd: annif eval ykl-omikuji-bonsai-${item.lang2} -j 8 -m Precision@1 -m NDCG
--metrics-file reports/test-omikuji-bonsai-${item.lang2}.json
--results-file reports/test-omikuji-bonsai-${item.lang2}.csv
corpora/kirjaesittelyt2021/kvesit-ykl-${item.lang3}-test.tsv
deps:
- venv-installed
- corpora/kirjaesittelyt2021/kvesit-ykl-${item.lang3}-test.tsv
- data/projects/ykl-omikuji-bonsai-${item.lang2}
outs:
- reports/test-omikuji-bonsai-${item.lang2}.csv:
cache: false
metrics:
- reports/test-omikuji-bonsai-${item.lang2}.json:
cache: false
eval-ensemble:
foreach:
- lang2: fi
lang3: fin
- lang2: sv
lang3: swe
- lang2: en
lang3: eng
do:
cmd: annif eval ykl-${item.lang2} -j 8 -m Precision@1 -m NDCG
--metrics-file reports/test-${item.lang2}.json
--results-file reports/test-${item.lang2}.csv
corpora/kirjaesittelyt2021/kvesit-ykl-${item.lang3}-test.tsv
deps:
- venv-installed
- corpora/kirjaesittelyt2021/kvesit-ykl-${item.lang3}-test.tsv
- data/projects/ykl-${item.lang2}
outs:
- reports/test-${item.lang2}.csv:
cache: false
metrics:
- reports/test-${item.lang2}.json:
cache: false