Switch all tf.concat(concat_dim, value, name) calls in third_party/tensorflow to tf.concat_v2(value, axis, name).

Change: 141255675

1 files changed, 5 insertions, 5 deletions

diff --git a/tensorflow/contrib/rnn/python/ops/lstm_ops.py b/tensorflow/contrib/rnn/python/ops/lstm_ops.py
index 3e30f24310..4ad269ab4f 100644
--- a/tensorflow/contrib/rnn/python/ops/lstm_ops.py
+++ b/tensorflow/contrib/rnn/python/ops/lstm_ops.py
@@ -277,7 +277,7 @@ def _LSTMBlockCellGrad(op, *grad):
   h_prev_grad.get_shape().merge_with(h_prev.get_shape())
 
   # Backprop from dicfo to w.
-  xh = array_ops.concat(1, [x, h_prev])
+  xh = array_ops.concat_v2([x, h_prev], 1)
   w_grad = math_ops.matmul(xh, dicfo, transpose_a=True)
   w_grad.get_shape().merge_with(w.get_shape())
 
@@ -527,10 +527,10 @@ class LSTMBlockWrapper(fused_rnn_cell.FusedRNNCell):
         # correctly,since we want to access the last valid state at
         # sequence_length - 1, which can even be -1, corresponding to the
         # initial state.
-        mod_cell_states = array_ops.concat(
-            0, [array_ops.expand_dims(initial_cell_state, [0]), cell_states])
-        mod_outputs = array_ops.concat(
-            0, [array_ops.expand_dims(initial_output, [0]), outputs])
+        mod_cell_states = array_ops.concat_v2(
+            [array_ops.expand_dims(initial_cell_state, [0]), cell_states], 0)
+        mod_outputs = array_ops.concat_v2(
+            [array_ops.expand_dims(initial_output, [0]), outputs], 0)
         final_cell_state = self._gather_states(mod_cell_states, sequence_length,
                                                batch_size)
         final_output = self._gather_states(mod_outputs, sequence_length,