diff --git a/dvc.lock b/dvc.lock
index 64c1df08..e4470911 100644
--- a/dvc.lock
+++ b/dvc.lock
@@ -1,30 +1,41 @@
-get_data:
-  cmd: python get_data.py
-  deps:
-  - path: get_data.py
-    md5: 1db5f442403042e0403c75132fe59af4
-  outs:
-  - path: data_raw.csv
-    md5: a6aec8da63a5fa2619af025a76746f29
-process:
-  cmd: python process_data.py
-  deps:
-  - path: data_raw.csv
-    md5: a6aec8da63a5fa2619af025a76746f29
-  - path: process_data.py
-    md5: 79b357c12f171f3d07c76780815b651c
-  outs:
-  - path: data_processed.csv
-    md5: 3b20a3a6ac0570f3de28b77d1e88f932
-train:
-  cmd: python train.py
-  deps:
-  - path: data_processed.csv
-    md5: 3b20a3a6ac0570f3de28b77d1e88f932
-  - path: train.py
-    md5: 80ad33d8caf823fc1d5cdefcb5b9490a
-  outs:
-  - path: by_region.png
-    md5: e7f3818fac35589b0c46dd65f8293e74
-  - path: metrics.json
-    md5: f4844c28505568f336c5f91db3f1beb3
+schema: '2.0'
+stages:
+  get_data:
+    cmd: python get_data.py
+    deps:
+    - path: get_data.py
+      md5: 1db5f442403042e0403c75132fe59af4
+      size: 516
+    outs:
+    - path: data_raw.csv
+      md5: a6aec8da63a5fa2619af025a76746f29
+      size: 68868
+  process:
+    cmd: python process_data.py
+    deps:
+    - path: data_raw.csv
+      md5: a6aec8da63a5fa2619af025a76746f29
+      size: 68868
+    - path: process_data.py
+      md5: 79b357c12f171f3d07c76780815b651c
+      size: 928
+    outs:
+    - path: data_processed.csv
+      md5: 3b20a3a6ac0570f3de28b77d1e88f932
+      size: 24627
+  train:
+    cmd: python train.py
+    deps:
+    - path: data_processed.csv
+      md5: 3b20a3a6ac0570f3de28b77d1e88f932
+      size: 24627
+    - path: train.py
+      md5: 80ad33d8caf823fc1d5cdefcb5b9490a
+      size: 1524
+    outs:
+    - path: by_region.png
+      md5: 47fd5c8b04c5d2d17404e115d8467461
+      size: 7812
+    - path: metrics.json
+      md5: f4844c28505568f336c5f91db3f1beb3
+      size: 89
diff --git a/train.py b/train.py
index e37f0eb5..e33c033e 100644
--- a/train.py
+++ b/train.py
@@ -1,6 +1,6 @@
 import pandas as pd 
 import numpy as np
-from sklearn.linear_model import LogisticRegression
+from sklearn.discriminant_analysis import QuadraticDiscriminantAnalysis
 from sklearn import preprocessing
 from sklearn.model_selection import cross_val_predict
 from sklearn.metrics import confusion_matrix
@@ -13,7 +13,8 @@
 
 df = pd.read_csv("data_processed.csv")
 
-#### Get features ready to model! 
+
+#### Get features ready to model! yeokah
 y = df.pop("cons_general").to_numpy()
 y[y< 4] = 0
 y[y>= 4] = 1
@@ -28,7 +29,7 @@
 
 
 # Linear model
-clf = LogisticRegression()
+clf = QuadraticDiscriminantAnalysis()
 yhat = cross_val_predict(clf, X, y, cv=5)
 
 acc = np.mean(yhat==y)