Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
46 changes: 30 additions & 16 deletions Makefile.manual
Original file line number Diff line number Diff line change
Expand Up @@ -339,7 +339,7 @@ translatewiki_tuning_reports: \
datasets/wikidatawiki.autolabeled_revisions.20k_2015.json: \
datasets/wikidatawiki.balanced_revisions.20k_2015.json
cat $< | \
./utility autolabel --host=https://wikidata.org \
./utility autolabel --host=https://www.wikidata.org \
--trusted-groups=abusefilter,arbcom,bureaucrat,checkuser,rollbacker,sysop,bot \
--trusted-edits=1000 \
--verbose > $@
Expand All @@ -360,9 +360,23 @@ datasets/wikidatawiki.labeled_revisions.w_cache.20k_2015.json: \
editquality.feature_lists.wikidatawiki.reverted \
editquality.feature_lists.wikidatawiki.damaging \
editquality.feature_lists.wikidatawiki.goodfaith \
--host https://wikidata.org \
--host https://www.wikidata.org \
--verbose > $@

datasets/wikidatawiki.labeled_revisions.6k_2018.json:
./utility fetch_labels \
https://labels.wmflabs.org/campaigns/wikidatawiki/81/ > $@

datasets/wikidatawiki.labeled_revisions.w_cache.6k_2018.json: \
datasets/wikidatawiki.labeled_revisions.6k_2018.json
cat $< | \
revscoring extract \
editquality.feature_lists.wikidatawiki.reverted \
editquality.feature_lists.wikidatawiki.damaging \
editquality.feature_lists.wikidatawiki.goodfaith \
--host https://www.wikidata.org \
--verbose > $@

# https://quarry.wmflabs.org/query/27882
datasets/wikidatawiki.sampled_revisions.500k_2018.json:
wget -qO- https://quarry.wmflabs.org/run/275172/output/0/json-lines?download=true > $@
Expand All @@ -382,29 +396,29 @@ datasets/wikidatawiki.autolabeled_revisions.500k_2018.review.json: \
cat $< | grep -E '"needs_review": (true|"True")' | grep -v '"review_reason": "blocked user"' > $@

tuning_reports/wikidatawiki.damaging.md: \
datasets/wikidatawiki.labeled_revisions.w_cache.20k_2015.json
cat $< | \
datasets/wikidatawiki.labeled_revisions.w_cache.20k_2015.json \
datasets/wikidatawiki.labeled_revisions.w_cache.6k_2018.json
cat $^ | \
revscoring tune \
config/classifiers.params.yaml \
editquality.feature_lists.wikidatawiki.damaging \
damaging \
roc_auc.labels.true \
--label-weight $(damaging_weight) \
--pop-rate "true=0.0008668694143782405" \
--pop-rate "false=0.9991331305856218" \
--labels "true,false" \
--center --scale \
--cv-timeout=60 \
--debug > $@

models/wikidatawiki.damaging.gradient_boosting.model: \
datasets/wikidatawiki.labeled_revisions.w_cache.20k_2015.json
cat $< | \
datasets/wikidatawiki.labeled_revisions.w_cache.20k_2015.json \
datasets/wikidatawiki.labeled_revisions.w_cache.6k_2018.json
cat $^ | \
revscoring cv_train \
revscoring.scoring.models.GradientBoosting \
editquality.feature_lists.wikidatawiki.damaging \
damaging \
--version=$(damaging_major_minor).0 \
--version=$(damaging_major_minor).1 \
-p 'max_depth=7' \
-p 'learning_rate=0.01' \
-p 'max_features="log2"' \
Expand All @@ -414,28 +428,28 @@ models/wikidatawiki.damaging.gradient_boosting.model: \
--center --scale > $@

tuning_reports/wikidatawiki.goodfaith.md: \
datasets/wikidatawiki.labeled_revisions.w_cache.20k_2015.json
cat $< | \
datasets/wikidatawiki.labeled_revisions.w_cache.20k_2015.json \
datasets/wikidatawiki.labeled_revisions.w_cache.6k_2018.json
cat $^ | \
revscoring tune \
config/classifiers.params.yaml \
editquality.feature_lists.wikidatawiki.goodfaith \
goodfaith \
roc_auc.labels.true \
--label-weight $(goodfaith_weight) \
--pop-rate "true=0.9998525516181488" \
--pop-rate "false=0.00014744838185121178" \
--labels "true,false" \
--cv-timeout=60 \
--debug > $@

models/wikidatawiki.goodfaith.gradient_boosting.model: \
datasets/wikidatawiki.labeled_revisions.w_cache.20k_2015.json
cat $< | \
datasets/wikidatawiki.labeled_revisions.w_cache.20k_2015.json \
datasets/wikidatawiki.labeled_revisions.w_cache.6k_2018.json
cat $^ | \
revscoring cv_train \
revscoring.scoring.models.GradientBoosting \
editquality.feature_lists.wikidatawiki.goodfaith \
goodfaith \
--version=$(goodfaith_major_minor).0 \
--version=$(goodfaith_major_minor).1 \
-p 'max_depth=7' \
-p 'learning_rate=0.1' \
-p 'max_features="log2"' \
Expand Down
4 changes: 2 additions & 2 deletions models/wikidatawiki.damaging.gradient_boosting.model
Git LFS file not shown
4 changes: 2 additions & 2 deletions models/wikidatawiki.goodfaith.gradient_boosting.model
Git LFS file not shown
Loading