fix: new c=0.1 for training baseline and add custom_stopwords. #5

configs/features.yaml

@@ -8,6 +8,19 @@ preprocessing:
   lemmatize: true
   min_token_length: 2
   language: en
+  custom_stopwords:
+    - youtube
+    - video
+    - watch
+    - like
+    - comment
+    - channel
+    - stefan
+    - peggy
+    - masri
+    - ferguson
+    - cigar
+    - hubbard
 
 vectorization:
   method: tfidf  # tfidf | bow | both

configs/models.yaml

@@ -0,0 +1,30 @@
+models:
+  logistic_regression:
+    C: 0.1
+    max_iter: 1000
+    class_weight: balanced
+    solver: lbfgs
+
+  random_forest:
+    n_estimators: 100
+    max_depth: null
+    min_samples_split: 2
+    class_weight: balanced
+    n_jobs: -1
+
+  xgboost:
+    n_estimators: 100
+    max_depth: 6
+    learning_rate: 0.1
+    subsample: 0.8
+    colsample_bytree: 0.8
+    scale_pos_weight: 1
+
+evaluation:
+  primary_metric: f1_weighted
+  metrics:
+    - accuracy
+    - f1_weighted
+    - precision_weighted
+    - recall_weighted
+    - roc_auc