KeyError: "The name 'deepspeech/logits:0' refers to a Tensor which does not exist.
FrancisChen888 opened this issue · 7 comments
While excute --- Step1: extract images from vids ---
Traceback (most recent call last):
File "data_util/deepspeech_features/extract_ds_features.py", line 128, in
main()
File "data_util/deepspeech_features/extract_ds_features.py", line 120, in main
extract_features(
File "data_util/deepspeech_features/extract_ds_features.py", line 80, in extract_features
conv_audios_to_deepspeech(
File "/data/chenxianhao/AD-NeRF/data_util/deepspeech_features/deepspeech_features.py", line 41, in conv_audios_to_deepspeech
graph, logits_ph, input_node_ph, input_lengths_ph = prepare_deepspeech_net(
File "/data/chenxianhao/AD-NeRF/data_util/deepspeech_features/deepspeech_features.py", line 105, in prepare_deepspeech_net
logits_ph = graph.get_tensor_by_name("deepspeech/logits:0")
File "/opt/anaconda3/lib/python3.8/site-packages/tensorflow/python/framework/ops.py", line 3931, in get_tensor_by_name
return self.as_graph_element(name, allow_tensor=True, allow_operation=False)
File "/opt/anaconda3/lib/python3.8/site-packages/tensorflow/python/framework/ops.py", line 3755, in as_graph_element
return self._as_graph_element_locked(obj, allow_tensor, allow_operation)
File "/opt/anaconda3/lib/python3.8/site-packages/tensorflow/python/framework/ops.py", line 3795, in _as_graph_element_locked
raise KeyError("The name %s refers to a Tensor which does not "
KeyError: "The name 'deepspeech/logits:0' refers to a Tensor which does not exist. The operation, 'deepspeech/logits', does not exist in the graph."
Hey, What exactly did u do to resolve the issue?.
Have you solved this problem?
I wonder have you solved this problem too.
Downgrading tensorflow-gpu to 1.15.0 helped me solve this problem.
Look for "logits:0" instead of "deepspeech/logits:0" will do the trick. This can be found after inspecting the name of all the nodes in the graph.
tensorflow-gpu to 1.15.0 is right
mark