keras-unet-deploy icon indicating copy to clipboard operation
keras-unet-deploy copied to clipboard

savepb error

Open pele228 opened this issue 4 years ago • 0 comments

你好,我按照你的流程,train和test都没有问题, 使用TensorFlow版本为1.14.0, keras版本为2.2.4 python demo.py --op savepb 命令行报错如下:

E:\keras-unet-deploy\py3>python demo.py --op savepb Using TensorFlow backend. models/myunet_64_1.2.h5 WARNING:tensorflow:From demo.py:142: The name tf.Session is deprecated. Please use tf.compat.v1.Session instead.

2021-02-05 14:31:41.618167: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 WARNING:tensorflow:From C:\Users\admin\AppData\Roaming\Python\Python35\site-packages\tensorflow\python\ops\nn_impl.py:180: add_dispatch_support..wrapper (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version. Instructions for updating: Use tf.where in 2.0, which has the same broadcast rule as np.where [ INFO ] inputs: ['input_1'] [ INFO ] outputs: ['output_0'] WARNING:tensorflow:From E:\admin\openvino\keras-unet-deploy\py3\utils.py:93: convert_variables_to_constants (from tensorflow.python.framework.graph_util_impl) is deprecated and will be removed in a future version. Instructions for updating: Use tf.compat.v1.graph_util.convert_variables_to_constants WARNING:tensorflow:From C:\Users\admin\AppData\Roaming\Python\Python35\site-packages\tensorflow\python\framework\graph_util_impl.py:270: extract_sub_graph (from tensorflow.python.framework.graph_util_impl) is deprecated and will be removed in a future version. Instructions for updating: Use tf.compat.v1.graph_util.extract_sub_graph Traceback (most recent call last): File "demo.py", line 216, in run_savepb(h5fpath) File "demo.py", line 146, in run_savepb utils.savepb(model, sess, pbname = pbfpath) File "E:\admin\openvino\keras-unet-deploy\py3\utils.py", line 93, in savepb graphdef_frozen = graph_util.convert_variables_to_constants(sess, graph_def, output_names) File "C:\Users\admin\AppData\Roaming\Python\Python35\site-packages\tensorflow\python\util\deprecation.py", line 324, in new_func return func(*args, **kwargs) File "C:\Users\admin\AppData\Roaming\Python\Python35\site-packages\tensorflow\python\framework\graph_util_impl.py", line 297, in convert_variables_to_constants source_op_name = get_input_name(node) File "C:\Users\admin\AppData\Roaming\Python\Python35\site-packages\tensorflow\python\framework\graph_util_impl.py", line 254, in get_input_name raise ValueError("Tensor name '{0}' is invalid.".format(node.input[0])) ValueError: Tensor name 'encoder_11_bn/cond/ReadVariableOp/Switch:1' is invalid.

pele228 avatar Feb 05 '21 06:02 pele228