Fixed bugs in accuracy test

tests/test_imputed_accuracy.py

@@ -10,7 +10,7 @@ def make_dataset(seed):
     random_state = np.random.RandomState(seed)
     iris = pd.concat(load_iris(return_X_y=True, as_frame=True), axis=1)
     iris["bi"] = random_state.binomial(
-        1, (iris["target"] == 0).map({True: 0.85, False: 0.15}), size=150
+        1, (iris["target"] == 0).map({True: 0.9, False: 0.10}), size=150
     )
     iris["bi"] = iris["bi"].astype("category")
     iris["sp"] = iris["target"].map({0: "A", 1: "B", 2: "C"}).astype("category")
@@ -25,7 +25,7 @@ def make_dataset(seed):
         axis=1,
         inplace=True,
     )
-    iris_amp = mf.utils.ampute_data(iris, perc=0.20)
+    iris_amp = mf.utils.ampute_data(iris, perc=0.20, random_state=random_state)
 
     return iris, iris_amp
 
@@ -83,7 +83,7 @@ def get_categorical_performance(kernel: mf.ImputationKernel, variables, iris):
         rocs[col] = roc_auc_score(orig, preds, multi_class="ovr", average="macro")
         accs[col] = (imps == orig).mean()
         rand_accs[col] = np.sum(
-            cand.value_counts(normalize=True) * imps.value_counts(normalize=True)
+            cand.value_counts(normalize=True) * orig.value_counts(normalize=True)
         )
     rocs = pd.Series(rocs)
     accs = pd.Series(accs)
@@ -94,7 +94,8 @@ def get_categorical_performance(kernel: mf.ImputationKernel, variables, iris):
 def test_defaults():
 
     for i in range(10):
-        # i = 0
+        # i = 3
+        print(i)
         iris, iris_amp = make_dataset(i)
         kernel_1 = mf.ImputationKernel(
             iris_amp,
@@ -166,7 +167,7 @@ def test_custom_params():
             iterations=4,
             verbose=False,
             boosting="random_forest",
-            num_iterations=500,
+            num_iterations=200,
             min_data_in_leaf=2,
         )
         kernel_1.complete_data(0, inplace=True)