fix aucop stat shape (#21846)

* fix stat shape back in global auc scenario
* add UT to cover global auc

paddle/fluid/operators/metrics/auc_op.cc

@@ -49,12 +49,15 @@ class AucOp : public framework::OperatorWithKernel {
 
     ctx->SetOutputDim("AUC", {1});
 
-    // slide_steps = slide_steps == 0 ? 1 : slide_steps;
-    int need_batch_id = slide_steps ? 1 : 0;
-    ctx->SetOutputDim("StatPosOut",
-                      {(1 + slide_steps) * num_pred_buckets + need_batch_id});
-    ctx->SetOutputDim("StatNegOut",
-                      {(1 + slide_steps) * num_pred_buckets + need_batch_id});
+    if (slide_steps) {
+      ctx->SetOutputDim("StatPosOut",
+                        {(1 + slide_steps) * num_pred_buckets + 1});
+      ctx->SetOutputDim("StatNegOut",
+                        {(1 + slide_steps) * num_pred_buckets + 1});
+    } else {
+      ctx->SetOutputDim("StatPosOut", {1, num_pred_buckets});
+      ctx->SetOutputDim("StatNegOut", {1, num_pred_buckets});
+    }
   }
 
  protected:

python/paddle/fluid/layers/metric_op.py

@@ -184,9 +184,9 @@ def auc(input,
     # for global auc
     # Needn't maintain the batch id
     stat_pos = helper.create_global_variable(
-        persistable=True, dtype='int64', shape=[num_thresholds + 1])
+        persistable=True, dtype='int64', shape=[1, num_thresholds + 1])
     stat_neg = helper.create_global_variable(
-        persistable=True, dtype='int64', shape=[num_thresholds + 1])
+        persistable=True, dtype='int64', shape=[1, num_thresholds + 1])
 
     for var in [batch_stat_pos, batch_stat_neg, stat_pos, stat_neg]:
         helper.set_variable_initializer(

python/paddle/fluid/tests/unittests/test_auc_op.py

@@ -64,5 +64,45 @@ class TestAucOp(OpTest):
         self.check_output()
 
 
+class TestGlobalAucOp(OpTest):
+    def setUp(self):
+        self.op_type = "auc"
+        pred = np.random.random((128, 2)).astype("float32")
+        labels = np.random.randint(0, 2, (128, 1)).astype("int64")
+        num_thresholds = 200
+        slide_steps = 0
+
+        stat_pos = np.zeros((1, (num_thresholds + 1))).astype("int64")
+        stat_neg = np.zeros((1, (num_thresholds + 1))).astype("int64")
+
+        self.inputs = {
+            'Predict': pred,
+            'Label': labels,
+            "StatPos": stat_pos,
+            "StatNeg": stat_neg
+        }
+        self.attrs = {
+            'curve': 'ROC',
+            'num_thresholds': num_thresholds,
+            "slide_steps": slide_steps
+        }
+
+        python_auc = metrics.Auc(name="auc",
+                                 curve='ROC',
+                                 num_thresholds=num_thresholds)
+        python_auc.update(pred, labels)
+
+        pos = python_auc._stat_pos
+        neg = python_auc._stat_neg
+        self.outputs = {
+            'AUC': np.array(python_auc.eval()),
+            'StatPosOut': np.array(pos),
+            'StatNegOut': np.array(neg)
+        }
+
+    def test_check_output(self):
+        self.check_output()
+
+
 if __name__ == "__main__":
     unittest.main()

python/paddle/fluid/tests/unittests/test_auc_single_pred_op.py

@@ -67,5 +67,48 @@ class TestAucSinglePredOp(OpTest):
         self.check_output()
 
 
+class TestAucGlobalSinglePredOp(OpTest):
+    def setUp(self):
+        self.op_type = "auc"
+        pred = np.random.random((128, 2)).astype("float32")
+        pred0 = pred[:, 0].reshape(128, 1)
+        labels = np.random.randint(0, 2, (128, 1)).astype("int64")
+        num_thresholds = 200
+        slide_steps = 0
+
+        stat_pos = np.zeros((1, (num_thresholds + 1))).astype("int64")
+        stat_neg = np.zeros((1, (num_thresholds + 1))).astype("int64")
+
+        self.inputs = {
+            'Predict': pred0,
+            'Label': labels,
+            "StatPos": stat_pos,
+            "StatNeg": stat_neg
+        }
+        self.attrs = {
+            'curve': 'ROC',
+            'num_thresholds': num_thresholds,
+            "slide_steps": slide_steps
+        }
+
+        python_auc = metrics.Auc(name="auc",
+                                 curve='ROC',
+                                 num_thresholds=num_thresholds)
+        for i in range(128):
+            pred[i][1] = pred[i][0]
+        python_auc.update(pred, labels)
+
+        pos = python_auc._stat_pos
+        neg = python_auc._stat_neg
+        self.outputs = {
+            'AUC': np.array(python_auc.eval()),
+            'StatPosOut': np.array(pos),
+            'StatNegOut': np.array(neg)
+        }
+
+    def test_check_output(self):
+        self.check_output()
+
+
 if __name__ == "__main__":
     unittest.main()