filip/Stalwart
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

Spam filter: Fix training sample size checks

Browse Source
This commit is contained in:
mdecimus
2025-12-26 10:22:48 +01:00
parent 7d29cfbccf
commit 5764a8580b
1 changed files with 21 additions and 29 deletions
Download Patch File Download Diff File Expand all files Collapse all files

50
crates/spam-filter/src/modules/classifier.rs
View File

@@ -262,8 +262,12 @@ impl SpamClassifier for Server {
Spam(SpamEvent::ModelNotReady), Spam(SpamEvent::ModelNotReady),
Reason = "Not enough samples for training", Reason = "Not enough samples for training",
Details = vec![ Details = vec![
trc::Value::from(ham_count + trainer.reservoir.ham.total_seen), trc::Value::from(trainer.reservoir.ham.total_seen),
trc::Value::from(spam_count + trainer.reservoir.spam.total_seen) trc::Value::from(trainer.reservoir.spam.total_seen)
],
Limit = vec![
trc::Value::from(config.min_ham_samples),
trc::Value::from(config.min_spam_samples)
], ],
Elapsed = started.elapsed() Elapsed = started.elapsed()
); );
@@ -273,7 +277,7 @@ impl SpamClassifier for Server {
// Balance classes if needed // Balance classes if needed
if spam_count > ham_count { if spam_count > ham_count {
// We have too much spam today. We need to replay old HAM. // We have too much spam this time. We need to replay old HAM.
samples.extend( samples.extend(
trainer trainer
.reservoir .reservoir
@@ -286,7 +290,7 @@ impl SpamClassifier for Server {
}), }),
); );
} else if ham_count > spam_count { } else if ham_count > spam_count {
// We have too much ham today. We need to replay old SPAM. // We have too much ham this time. We need to replay old SPAM.
samples.extend( samples.extend(
trainer trainer
.reservoir .reservoir
@@ -491,32 +495,20 @@ impl SpamClassifier for Server {
) )
.await .await
.caused_by(trc::location!())?; .caused_by(trc::location!())?;
if ham_count >= config.min_ham_samples && spam_count >= config.min_spam_samples { self.blob_store()
self.blob_store() .put_blob(
.put_blob( SPAM_CLASSIFIER_KEY,
SPAM_CLASSIFIER_KEY, &classifier.serialize().caused_by(trc::location!())?,
&classifier.serialize().caused_by(trc::location!())?, )
) .await
.await .caused_by(trc::location!())?;
.caused_by(trc::location!())?;
self.inner self.inner
.data .data
.spam_classifier .spam_classifier
.store(Arc::new(classifier.inner)); .store(Arc::new(classifier.inner));
self.cluster_broadcast(BroadcastEvent::ReloadSpamFilter) self.cluster_broadcast(BroadcastEvent::ReloadSpamFilter)
.await; .await;
} else {
self.blob_store()
.delete_blob(SPAM_CLASSIFIER_KEY)
.await
.caused_by(trc::location!())?;
trc::event!(
Spam(SpamEvent::ModelNotReady),
Details = vec![trc::Value::from(ham_count), trc::Value::from(spam_count)],
);
}
trc::event!( trc::event!(
Spam(SpamEvent::TrainCompleted), Spam(SpamEvent::TrainCompleted),