Initial commit: Add model and vectorizer

README.md

@@ -0,0 +1,10 @@
+# Multi-Label Hate Speech Classifier
+
+This model uses a OneVsRest Logistic Regression classifier with TF-IDF vectorization to detect multiple types of hate speech:
+- toxic
+- obscene
+- insult
+- threat
+- identity_hate
+
+Custom thresholds have been applied for each label.

model.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f56d854b87388a1fb6838c9b7ab071bf40272067289cb346a21eaf645a9ab5a
+size 802676

tfidf_vectorizer.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a2534c76b8bb47dd33f7c895dc08d079f0cb7237731b77545d953d59801fa70
+size 746426