optimize_for_inference_lib.fold_batch_norms() preserves data_format (tensorflow#16075)

Fixes tensorflow#15034

Author: yegord

tensorflow/python/tools/optimize_for_inference_lib.py

@@ -349,6 +349,7 @@ def fold_batch_norms(input_graph_def):
     bias_add_op.op = "BiasAdd"
     bias_add_op.name = node.name
     bias_add_op.attr["T"].CopyFrom(conv_op.attr["T"])
+    bias_add_op.attr["data_format"].CopyFrom(conv_op.attr["data_format"])
     bias_add_op.input.extend([new_conv_op.name, offset_op.name])
     new_ops.extend([scaled_weights_op, new_conv_op, offset_op, bias_add_op])
 

tensorflow/python/tools/optimize_for_inference_test.py

@@ -173,48 +173,53 @@ def testFoldBatchNorms(self):
       self.assertNotEqual("BatchNormWithGlobalNormalization", node.op)
 
   def testFoldFusedBatchNorms(self):
-    with self.test_session() as sess:
-      inputs = [1, 4, 2, 5, 3, 6, -1, -4, -2, -5, -3, -6]
-      input_op = constant_op.constant(
-          np.array(inputs), shape=[1, 1, 6, 2], dtype=dtypes.float32)
-      weights = [1, 2, 3, 4, 0.1, 0.2, 0.3, 0.4]
-      weights_op = constant_op.constant(
-          np.array(weights), shape=[1, 2, 2, 2], dtype=dtypes.float32)
-      conv_op = nn_ops.conv2d(
-          input_op, weights_op, [1, 1, 1, 1], padding="SAME", name="conv_op")
-      mean_op = constant_op.constant(
-          np.array([10, 20]), shape=[2], dtype=dtypes.float32)
-      variance_op = constant_op.constant(
-          np.array([0.25, 0.5]), shape=[2], dtype=dtypes.float32)
-      beta_op = constant_op.constant(
-          np.array([0.1, 0.6]), shape=[2], dtype=dtypes.float32)
-      gamma_op = constant_op.constant(
-          np.array([1.0, 2.0]), shape=[2], dtype=dtypes.float32)
-      ops.get_default_graph().graph_def_versions.producer = 9
-      gen_nn_ops._fused_batch_norm(
-          conv_op,
-          gamma_op,
-          beta_op,
-          mean_op,
-          variance_op,
-          0.00001,
-          is_training=False,
-          name="output")
-      original_graph_def = sess.graph_def
-      original_result = sess.run(["output:0"])
-    optimized_graph_def = optimize_for_inference_lib.fold_batch_norms(
-        original_graph_def)
-
-    with self.test_session() as sess:
-      _ = importer.import_graph_def(
-          optimized_graph_def, input_map={}, name="optimized")
-      optimized_result = sess.run(["optimized/output:0"])
-
-    self.assertAllClose(
-        original_result, optimized_result, rtol=1e-04, atol=1e-06)
-
-    for node in optimized_graph_def.node:
-      self.assertNotEqual("FusedBatchNorm", node.op)
+    for data_format, use_gpu in [("NHWC", False), ("NCHW", True)]:
+      with self.test_session(use_gpu=use_gpu) as sess:
+        inputs = [1, 4, 2, 5, 3, 6, -1, -4, -2, -5, -3, -6]
+        input_op = constant_op.constant(
+            np.array(inputs),
+            shape=[1, 1, 6, 2] if data_format == "NHWC" else [1, 2, 1, 6],
+            dtype=dtypes.float32)
+        weights = [1, 2, 3, 4, 0.1, 0.2, 0.3, 0.4]
+        weights_op = constant_op.constant(
+            np.array(weights), shape=[1, 2, 2, 2], dtype=dtypes.float32)
+        conv_op = nn_ops.conv2d(
+            input_op, weights_op, [1, 1, 1, 1], padding="SAME",
+            data_format=data_format, name="conv_op")
+        mean_op = constant_op.constant(
+            np.array([10, 20]), shape=[2], dtype=dtypes.float32)
+        variance_op = constant_op.constant(
+            np.array([0.25, 0.5]), shape=[2], dtype=dtypes.float32)
+        beta_op = constant_op.constant(
+            np.array([0.1, 0.6]), shape=[2], dtype=dtypes.float32)
+        gamma_op = constant_op.constant(
+            np.array([1.0, 2.0]), shape=[2], dtype=dtypes.float32)
+        ops.get_default_graph().graph_def_versions.producer = 9
+        gen_nn_ops._fused_batch_norm(
+            conv_op,
+            gamma_op,
+            beta_op,
+            mean_op,
+            variance_op,
+            0.00001,
+            is_training=False,
+            data_format=data_format,
+            name="output")
+        original_graph_def = sess.graph_def
+        original_result = sess.run(["output:0"])
+      optimized_graph_def = optimize_for_inference_lib.fold_batch_norms(
+          original_graph_def)
+
+      with self.test_session(use_gpu=use_gpu) as sess:
+        _ = importer.import_graph_def(
+            optimized_graph_def, input_map={}, name="optimized")
+        optimized_result = sess.run(["optimized/output:0"])
+
+      self.assertAllClose(
+          original_result, optimized_result, rtol=1e-04, atol=1e-06)
+
+      for node in optimized_graph_def.node:
+        self.assertNotEqual("FusedBatchNorm", node.op)
 
   def testFuseResizePadAndConv(self):
     with self.test_session() as sess: