replace cross_entropy except in python/paddle/fluid/tests/unittests/*.py &&...

replace cross_entropy except in python/paddle/fluid/tests/unittests/*.py && unittests/*/*.py (#48922)

python/paddle/fluid/tests/unittests/collective/fleet/auto_parallel_parallelizer.py

@@ -22,7 +22,6 @@ import paddle.static as static
 import paddle.utils as utils
 from paddle.distributed import fleet
 from paddle.distributed.fleet import auto
-from paddle.fluid import layers
 
 paddle.enable_static()
 _global_parallel_strategy = None
@@ -93,7 +92,9 @@ def mlp_pretrain_forward(train_program, start_program):
 
         predict = mlp(input)
 
-        cost = layers.cross_entropy(input=predict, label=label)
+        cost = paddle.nn.functional.cross_entropy(
+            input=predict, label=label, reduction='none', use_softmax=False
+        )
         avg_cost = paddle.mean(x=cost)
 
     return avg_cost, train_program, start_program

python/paddle/fluid/tests/unittests/collective/fleet/dist_mnist_gradient_merge.py

@@ -34,7 +34,9 @@ class TestDistMnist2x2(TestDistRunnerBase):
 
         # Train program
         predict = cnn_model(images)
-        cost = fluid.layers.cross_entropy(input=predict, label=label)
+        cost = paddle.nn.functional.cross_entropy(
+            input=predict, label=label, reduction='none', use_softmax=False
+        )
         avg_cost = paddle.mean(x=cost)
 
         # Evaluator

python/paddle/fluid/tests/unittests/collective/fleet/parallel_dygraph_se_resnext.py

@@ -350,7 +350,9 @@ class TestSeResNeXt(TestParallelDyGraphRunnerBase):
 
         out = model(img)
         softmax_out = paddle.nn.functional.softmax(out, use_cudnn=False)
-        loss = fluid.layers.cross_entropy(input=softmax_out, label=label)
+        loss = paddle.nn.functional.cross_entropy(
+            input=softmax_out, label=label, reduction='none', use_softmax=False
+        )
         avg_loss = paddle.mean(x=loss)
         return avg_loss
 

python/paddle/fluid/tests/unittests/collective/fleet/pipeline_mnist.py

@@ -100,7 +100,9 @@ class TestDistMnist2x2(TestDistRunnerBase):
             # Train program
             predict = cnn_model(images)
         with fluid.device_guard("gpu:1"):
-            cost = fluid.layers.cross_entropy(input=predict, label=label)
+            cost = paddle.nn.functional.cross_entropy(
+                input=predict, label=label, reduction='none', use_softmax=False
+            )
             avg_cost = paddle.mean(x=cost)
 
         # Evaluator

python/paddle/fluid/tests/unittests/collective/fleet/pipeline_mnist_multi_device.py

@@ -100,7 +100,9 @@ class TestDistMnist2x2(TestDistRunnerBase):
             # Train program
             predict = cnn_model(images)
         with fluid.device_guard("gpu:1"):
-            cost = fluid.layers.cross_entropy(input=predict, label=label)
+            cost = paddle.nn.functional.cross_entropy(
+                input=predict, label=label, reduction='none', use_softmax=False
+            )
             avg_cost = paddle.mean(x=cost)
 
         # Evaluator

python/paddle/fluid/tests/unittests/collective/fleet/pipeline_mnist_one_device.py

@@ -92,7 +92,9 @@ class TestDistMnist2x2(TestDistRunnerBase):
             # Train program
             predict = cnn_model(images)
         with fluid.device_guard("gpu:0"):
-            cost = fluid.layers.cross_entropy(input=predict, label=label)
+            cost = paddle.nn.functional.cross_entropy(
+                input=predict, label=label, reduction='none', use_softmax=False
+            )
             avg_cost = paddle.mean(x=cost)
 
         # Evaluator

python/paddle/fluid/tests/unittests/collective/fleet/test_fleet_checkpoint.py

@@ -41,7 +41,9 @@ class FleetTest(unittest.TestCase):
             feed_list=[image, label], place=fluid.CPUPlace()
         )
         predict = fluid.layers.fc(input=image, size=10, act='softmax')
-        loss = fluid.layers.cross_entropy(input=predict, label=label)
+        loss = paddle.nn.functional.cross_entropy(
+            input=predict, label=label, reduction='none', use_softmax=False
+        )
         avg_loss = paddle.mean(loss)
         optimizer = fluid.optimizer.AdamOptimizer(learning_rate=0.001)
 

python/paddle/fluid/tests/unittests/collective/fleet/test_imperative_auto_mixed_precision.py

@@ -1308,7 +1308,9 @@ class TestResnet(unittest.TestCase):
                 ):
                     out = resnet(img)
 
-                loss = fluid.layers.cross_entropy(input=out, label=label)
+                loss = paddle.nn.functional.cross_entropy(
+                    input=out, label=label, reduction='none', use_softmax=False
+                )
                 avg_loss = paddle.mean(x=loss)
 
                 dy_out = avg_loss.numpy()

python/paddle/fluid/tests/unittests/collective/fleet/test_imperative_auto_mixed_precision_for_eager.py

@@ -1298,7 +1298,9 @@ class TestResnet(unittest.TestCase):
                 ):
                     out = resnet(img)
 
-                loss = fluid.layers.cross_entropy(input=out, label=label)
+                loss = paddle.nn.functional.cross_entropy(
+                    input=out, label=label, reduction='none', use_softmax=False
+                )
                 avg_loss = paddle.mean(x=loss)
 
                 dy_out = avg_loss.numpy()

python/paddle/fluid/tests/unittests/ir/inference/test_trt_conv_quant_dequant_pass.py

@@ -53,7 +53,12 @@ class QuantDequantTensorRTSubgraphPassConvTest(QuantDequantTest):
             elif self.conv_groups == 4:
                 cout = paddle.reshape(conv_out, shape=[1, 1, 10816])
             result = fluid.layers.relu(cout)
-            loss = fluid.layers.cross_entropy(input=result, label=label_shape)
+            loss = paddle.nn.functional.cross_entropy(
+                input=result,
+                label=label_shape,
+                reduction='none',
+                use_softmax=False,
+            )
             avg_loss = paddle.mean(loss)
             return avg_loss, result
 
@@ -156,7 +161,12 @@ class DynamicShapeQuantDequantTensorRTSubgraphPassConvTest(QuantDequantTest):
             )
             cout = paddle.reshape(conv_out, shape=[1, 1, 10816])
             result = fluid.layers.relu(cout)
-            loss = fluid.layers.cross_entropy(input=result, label=label_shape)
+            loss = paddle.nn.functional.cross_entropy(
+                input=result,
+                label=label_shape,
+                reduction='none',
+                use_softmax=False,
+            )
             avg_loss = paddle.mean(loss)
             return avg_loss, result
 
@@ -257,7 +267,12 @@ class QuantDequantTensorRTSubgraphPassConvTransposeTest(QuantDequantTest):
             elif self.conv_groups == 4:
                 cout = paddle.reshape(conv_out, shape=[1, 1, 10816])
             result = fluid.layers.relu(cout)
-            loss = fluid.layers.cross_entropy(input=result, label=label_shape)
+            loss = paddle.nn.functional.cross_entropy(
+                input=result,
+                label=label_shape,
+                reduction='none',
+                use_softmax=False,
+            )
             avg_loss = paddle.mean(loss)
             return avg_loss, result
 

python/paddle/fluid/tests/unittests/ir/inference/test_trt_fc_fuse_quant_dequant_pass.py

@@ -38,7 +38,12 @@ class FCQuantDequantFusePassTRTDims3Cols1Test(QuantDequantTest):
                 act="relu",
             )
             result = fluid.layers.relu(fc_out)
-            loss = fluid.layers.cross_entropy(input=result, label=self.label)
+            loss = paddle.nn.functional.cross_entropy(
+                input=result,
+                label=self.label,
+                reduction='none',
+                use_softmax=False,
+            )
             avg_loss = paddle.mean(loss)
             return avg_loss, result
 
@@ -105,7 +110,12 @@ class FCQuantDequantFusePassTRTDims3Cols2Test(QuantDequantTest):
             )
             c_out = paddle.reshape(fc_out, shape=[0, 784])
             result = fluid.layers.relu(c_out)
-            loss = fluid.layers.cross_entropy(input=result, label=self.label)
+            loss = paddle.nn.functional.cross_entropy(
+                input=result,
+                label=self.label,
+                reduction='none',
+                use_softmax=False,
+            )
             avg_loss = paddle.mean(loss)
             return avg_loss, result
 
@@ -174,7 +184,12 @@ class FCQuantDequantFusePassTRTDims3Cols3Test(QuantDequantTest):
             )
             c_out = paddle.reshape(fc_out, shape=[1, 1, 2744])
             result = fluid.layers.relu(c_out)
-            loss = fluid.layers.cross_entropy(input=result, label=label_shape)
+            loss = paddle.nn.functional.cross_entropy(
+                input=result,
+                label=label_shape,
+                reduction='none',
+                use_softmax=False,
+            )
             avg_loss = paddle.mean(loss)
             return avg_loss, result
 

python/paddle/fluid/tests/unittests/ir/inference/test_trt_matmul_quant_dequant.py

@@ -47,7 +47,12 @@ class TensorRTMatMulQuantDequantDims3Test(QuantDequantTest):
                 act=None,
             )
             result = fluid.layers.relu(fc_out)
-            loss = fluid.layers.cross_entropy(input=result, label=self.label)
+            loss = paddle.nn.functional.cross_entropy(
+                input=result,
+                label=self.label,
+                reduction='none',
+                use_softmax=False,
+            )
             avg_loss = paddle.mean(loss)
             return avg_loss, result
 
@@ -144,7 +149,12 @@ class TensorRTMatMulQuantDequantDims4Test(QuantDequantTest):
                 act=None,
             )
             result = fluid.layers.relu(fc_out)
-            loss = fluid.layers.cross_entropy(input=result, label=self.label)
+            loss = paddle.nn.functional.cross_entropy(
+                input=result,
+                label=self.label,
+                reduction='none',
+                use_softmax=False,
+            )
             avg_loss = paddle.mean(loss)
             return avg_loss, result
 
@@ -240,7 +250,12 @@ class TensorRTMatMulQuantDequantDims3DynamicTest(QuantDequantTest):
                 act=None,
             )
             result = fluid.layers.relu(fc_out)
-            loss = fluid.layers.cross_entropy(input=result, label=self.label)
+            loss = paddle.nn.functional.cross_entropy(
+                input=result,
+                label=self.label,
+                reduction='none',
+                use_softmax=False,
+            )
             avg_loss = paddle.mean(loss)
             return avg_loss, result