experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

tf_saved_model: Test that variable values are imported correctly.

Browse Source 
Now that http://cl/282439157 has landed, we can now use different variable values in this test :)

PiperOrigin-RevId: 282459295
Change-Id: I615e30b75fe769023e8f9747580d027ecb05e67b
This commit is contained in:
Sean Silva 2019-11-25 16:38:45 -08:00 committed by TensorFlower Gardener
parent b9e83ea2b2
commit 29f12759ab
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
tensorflow/compiler/mlir/tensorflow/tests/tf_saved_model/partially_shaped_variables.py
View File

@ -29,9 +29,9 @@ class TestModule(tf.Module):
def __init__(self):
super(TestModule, self).__init__()
# CHECK: "tf_saved_model.global_tensor"() {is_mutable, {{.*}} tf_saved_model.exported_names = ["v0"], type = tensor<*xf32>, value = dense<0.000000e+00> : tensor<1xf32>} : () -> ()
# CHECK: "tf_saved_model.global_tensor"() {is_mutable, {{.*}} tf_saved_model.exported_names = ["v1"], type = tensor<?xf32>, value = dense<0.000000e+00> : tensor<1xf32>} : () -> ()
# CHECK: "tf_saved_model.global_tensor"() {is_mutable, {{.*}} tf_saved_model.exported_names = ["v1"], type = tensor<?xf32>, value = dense<[0.000000e+00, 1.000000e+00]> : tensor<2xf32>} : () -> ()
self.v0 = tf.Variable([0.], shape=tf.TensorShape(None))
self.v1 = tf.Variable([0.], shape=[None])
self.v1 = tf.Variable([0., 1.], shape=[None])
if __name__ == '__main__':