Error during trie creation

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

No, re-running will download the latest binaries from master branch as I said earlier. To retrain, re-run the training ?

[pra978]

Ohkay, so you are saying latest binaries from Master branch are > r1.4.

Actually I am trying to confirm because my tensorflow is 1.6 and I trained the new model 3 days ago only and I got the error.

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

It’s easy to confirm: we have the version from DeepSpeech and TensorFlow being outputted on standard error now, so you should see v1.6.0-… If you don’t see it, then you are not downloading or running the proper binaries.

[pra978]

To download latest binaries from master branch, I re-ran the command :

python3 util/taskcluster.py --arch osx --target .

and it showed something like this while downloading:

Downloading https://index.taskcluster.net/v1/task/project.deepspeech.deepspeech.native_client.master.osx/artifacts/public/native_client.tar.xz …

Then i re-run the training.

Then i got model, i ran inference but it threw exact same error : ‘identical_element_shapes’.

Did i miss anything here?

My tensorflow version is still v1.6.0

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

I cannot help you if you don’t provide me with only partial informations. Please provide the full output of all your process, please verify that the downloaded files are the ones you are calling.

[pra978]

okay.

This is my output while running inference.

Loading model from file /Users/naveen/Downloads/DeepSpeech/results/model_export/output_graph.pb
Loaded model in 0.090s.
Running inference.
2018-04-27 13:38:04.067366: E tensorflow/core/framework/op_segment.cc:53] Create kernel failed: Invalid argument: NodeDef mentions attr ‘identical_element_shapes’ not in Op<name=TensorArrayV3; signature=size:int32 -> handle:resource, flow:float; attr=dtype:type; attr=element_shape:shape,default=; attr=dynamic_size:bool,default=false; attr=clear_after_read:bool,default=true; attr=tensor_array_name:string,default=""; is_stateful=true>; NodeDef: bidirectional_rnn/bw/bw/TensorArray_1 = TensorArrayV3clear_after_read=true, dtype=DT_FLOAT, dynamic_size=false, element_shape=[?,750], identical_element_shapes=true, tensor_array_name=“bidirectional_rnn/bw/bw/dynamic_rnn/input_0”, _device="/job:localhost/replica:0/task:0/device:CPU:0". (Check whether your GraphDef-interpreting binary is up to date with your GraphDef-generating binary.).
2018-04-27 13:38:04.067432: E tensorflow/core/common_runtime/executor.cc:643] Executor failed to create kernel. Invalid argument: NodeDef mentions attr ‘identical_element_shapes’ not in Op<name=TensorArrayV3; signature=size:int32 -> handle:resource, flow:float; attr=dtype:type; attr=element_shape:shape,default=; attr=dynamic_size:bool,default=false; attr=clear_after_read:bool,default=true; attr=tensor_array_name:string,default=""; is_stateful=true>; NodeDef: bidirectional_rnn/bw/bw/TensorArray_1 = TensorArrayV3clear_after_read=true, dtype=DT_FLOAT, dynamic_size=false, element_shape=[?,750], identical_element_shapes=true, tensor_array_name=“bidirectional_rnn/bw/bw/dynamic_rnn/input_0”, _device="/job:localhost/replica:0/task:0/device:CPU:0". (Check whether your GraphDef-interpreting binary is up to date with your GraphDef-generating binary.).
[[Node: bidirectional_rnn/bw/bw/TensorArray_1 = TensorArrayV3clear_after_read=true, dtype=DT_FLOAT, dynamic_size=false, element_shape=[?,750], identical_element_shapes=true, tensor_array_name=“bidirectional_rnn/bw/bw/dynamic_rnn/input_0”, _device="/job:localhost/replica:0/task:0/device:CPU:0"]]
Error running session: Invalid argument: NodeDef mentions attr ‘identical_element_shapes’ not in Op<name=TensorArrayV3; signature=size:int32 -> handle:resource, flow:float; attr=dtype:type; attr=element_shape:shape,default=; attr=dynamic_size:bool,default=false; attr=clear_after_read:bool,default=true; attr=tensor_array_name:string,default=""; is_stateful=true>; NodeDef: bidirectional_rnn/bw/bw/TensorArray_1 = TensorArrayV3clear_after_read=true, dtype=DT_FLOAT, dynamic_size=false, element_shape=[?,750], identical_element_shapes=true, tensor_array_name=“bidirectional_rnn/bw/bw/dynamic_rnn/input_0”, _device="/job:localhost/replica:0/task:0/device:CPU:0". (Check whether your GraphDef-interpreting binary is up to date with your GraphDef-generating binary.).
[[Node: bidirectional_rnn/bw/bw/TensorArray_1 = TensorArrayV3clear_after_read=true, dtype=DT_FLOAT, dynamic_size=false, element_shape=[?,750], identical_element_shapes=true, tensor_array_name=“bidirectional_rnn/bw/bw/dynamic_rnn/input_0”, _device="/job:localhost/replica:0/task:0/device:CPU:0"]]
None
Inference took 0.269s for 7.920s audio file.

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

No version number in this, you are not running the latest binaries.

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

$ python ../util/taskcluster.py --arch osx --target . 
Downloading https://index.taskcluster.net/v1/task/project.deepspeech.deepspeech.native_client.master.osx/artifacts/public/native_client.tar.xz ...
Downloading: 100%

generate_trie
libctc_decoder_with_kenlm.so
libdeepspeech.so
libdeepspeech_utils.so
LICENSE
deepspeech
README.mozilla
$ LC_ALL=C ls -hal libdeepspeech.so deepspeech 
-rwxr-xr-x 1 alex alex 26K Apr 20 00:31 deepspeech
-r-xr-xr-x 1 alex alex 75M Apr 20 00:31 libdeepspeech.so

[pra978]

This is what i got while downloading/installing binaries inside DeepSpeech folder:

Prafful’s MacBook Pro:DeepSpeech naveen$ python3 util/taskcluster.py --arch osx --target .
Downloading https://index.taskcluster.net/v1/task/project.deepspeech.deepspeech.native_client.master.osx/artifacts/public/native_client.tar.xz …
Downloading: 100%

x generate_trie
x libctc_decoder_with_kenlm.so
x libdeepspeech.so
x libdeepspeech_utils.so
x LICENSE
x deepspeech
x README.mozilla

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

$ sha1sum deepspeech libdeepspeech.so 
b0455dd60674ad49d858a161902b37d7c87f4282  deepspeech
0c3f7496930eaed4759e8a78d05b99f965ef6615  libdeepspeech.so

[pra978]

$ sha1sum deepspeech libdeepspeech.so
b0455dd60674ad49d858a161902b37d7c87f4282 deepspeech
0c3f7496930eaed4759e8a78d05b99f965ef6615 libdeepspeech.so

What does this signify?

[pra978]

This command :
$ python …/util/taskcluster.py --arch osx --target .
Downloading https://index.taskcluster.net/v1/task/project.deepspeech.deepspeech.native_client.master.osx/artifacts/public/native_client.tar.xz …
Downloading: 100%

This is exact same command which i ran. And from exact same URL, its downloading in both cases. Only difference that i see is that i used “python3” while you used “python”.

Will that make any difference?

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

No, it means you have downoaded the proper binaries. But I still don’t know exactly how you run your inference …

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

You can check for example this log from C++ tests running on OSX: https://taskcluster-artifacts.net/Xcn9cfUDSz2HEjc4uXfJeQ/0/public/logs/live_backing.log

There you will see output with TensorFlow and DeepSpeech versions given.

[pra978]

Okay. My bad.

This is the complete code for running inference:

Prafful’s MacBook Pro:DeepSpeech naveen$ deepspeech /Users/naveen/Downloads/DeepSpeech/results/model_export/output_graph.pb /Users/naveen/Downloads/DeepSpeech/TEST/engtext_3488.wav /Users/naveen/Downloads/DeepSpeech/alphabet.txt
Loading model from file /Users/naveen/Downloads/DeepSpeech/results/model_export/output_graph.pb
Loaded model in 0.090s.
Running inference.
2018-04-27 13:38:04.067366: E tensorflow/core/framework/op_segment.cc:53] Create kernel failed: Invalid argument: NodeDef mentions attr ‘identical_element_shapes’ not in Op<name=TensorArrayV3; signature=size:int32 -> handle:resource, flow:float; attr=dtype:type; attr=element_shape:shape,default=; attr=dynamic_size:bool,default=false; attr=clear_after_read:bool,default=true; attr=tensor_array_name:string,default=""; is_stateful=true>; NodeDef: bidirectional_rnn/bw/bw/TensorArray_1 = TensorArrayV3clear_after_read=true, dtype=DT_FLOAT, dynamic_size=false, element_shape=[?,750], identical_element_shapes=true, tensor_array_name=“bidirectional_rnn/bw/bw/dynamic_rnn/input_0”, _device="/job:localhost/replica:0/task:0/device:CPU:0". (Check whether your GraphDef-interpreting binary is up to date with your GraphDef-generating binary.).
2018-04-27 13:38:04.067432: E tensorflow/core/common_runtime/executor.cc:643] Executor failed to create kernel. Invalid argument: NodeDef mentions attr ‘identical_element_shapes’ not in Op<name=TensorArrayV3; signature=size:int32 -> handle:resource, flow:float; attr=dtype:type; attr=element_shape:shape,default=; attr=dynamic_size:bool,default=false; attr=clear_after_read:bool,default=true; attr=tensor_array_name:string,default=""; is_stateful=true>; NodeDef: bidirectional_rnn/bw/bw/TensorArray_1 = TensorArrayV3clear_after_read=true, dtype=DT_FLOAT, dynamic_size=false, element_shape=[?,750], identical_element_shapes=true, tensor_array_name=“bidirectional_rnn/bw/bw/dynamic_rnn/input_0”, _device="/job:localhost/replica:0/task:0/device:CPU:0". (Check whether your GraphDef-interpreting binary is up to date with your GraphDef-generating binary.).
[[Node: bidirectional_rnn/bw/bw/TensorArray_1 = TensorArrayV3clear_after_read=true, dtype=DT_FLOAT, dynamic_size=false, element_shape=[?,750], identical_element_shapes=true, tensor_array_name=“bidirectional_rnn/bw/bw/dynamic_rnn/input_0”, _device="/job:localhost/replica:0/task:0/device:CPU:0"]]
Error running session: Invalid argument: NodeDef mentions attr ‘identical_element_shapes’ not in Op<name=TensorArrayV3; signature=size:int32 -> handle:resource, flow:float; attr=dtype:type; attr=element_shape:shape,default=; attr=dynamic_size:bool,default=false; attr=clear_after_read:bool,default=true; attr=tensor_array_name:string,default=""; is_stateful=true>; NodeDef: bidirectional_rnn/bw/bw/TensorArray_1 = TensorArrayV3clear_after_read=true, dtype=DT_FLOAT, dynamic_size=false, element_shape=[?,750], identical_element_shapes=true, tensor_array_name=“bidirectional_rnn/bw/bw/dynamic_rnn/input_0”, _device="/job:localhost/replica:0/task:0/device:CPU:0". (Check whether your GraphDef-interpreting binary is up to date with your GraphDef-generating binary.).
[[Node: bidirectional_rnn/bw/bw/TensorArray_1 = TensorArrayV3clear_after_read=true, dtype=DT_FLOAT, dynamic_size=false, element_shape=[?,750], identical_element_shapes=true, tensor_array_name=“bidirectional_rnn/bw/bw/dynamic_rnn/input_0”, _device="/job:localhost/replica:0/task:0/device:CPU:0"]]
None
Inference took 0.269s for 7.920s audio file.

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

Please:

sha1sum `which deepspeech`

[pra978]

sha1sum which deepspeech

Prafful’s MacBook Pro:DeepSpeech naveen$ sha1sum which deepspeech
-bash: sha1sum: command not found

[pra978]

I used this to check version of my tensorflow:

python3 -c ‘import tensorflow as tf; print(tf.version)’

Is there any method/command to check version of my downloaded and running binaries?

[pra978]

fe649aafec251357ff8a1e14b3ed842e79d510c2 /anaconda3/bin/deepspeech

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

And bingo, you’re running the binary from python package v0.1.1 that you installed in anaconda3. Please use the proper path to your extracted deepspeech binary.