Runtimeerror createmodel failed with error code 8195

[JRMeyer]

>>> shubhamc247
[November 19, 2019, 6:13am]

I trained my own data , it train very well. slash
but when i predict by using following command slash
(deepspeech-venv)
shubh slash shubh-Lenovo-Flex-2-14: slash ~/Downloads/DeepSpeech slash $ deepspeech
models/output_graph.pbmm models/alphabet.txt

it get 8195 error.

TensorFlow: v1.13.1-10-g3e0cc53 slash
DeepSpeech: v0.5.1-0-g4b29b78 slash
Warning: reading entire model file into memory. Transform model file
into an mmapped graph to reduce heap usage. slash
2019-11-19 11:23:13.351350: I
tensorflow/core/platform/cpu_feature_guard.cc:141 slash ] Your CPU supports
instructions that this TensorFlow binary was not compiled to use: AVX2
FMA slash
2019-11-19 11:23:13.497345: E
tensorflow/core/framework/op_kernel.cc:1325 slash ] OpKernel ('op:
'UnwrapDatasetVariant' device_type: 'CPU'') for unknown op:
UnwrapDatasetVariant slash
2019-11-19 11:23:13.497400: E
tensorflow/core/framework/op_kernel.cc:1325 slash ] OpKernel ('op:
'WrapDatasetVariant' device_type: 'GPU' host_memory_arg: 'input_handle'
host_memory_arg: 'output_handle'') for unknown op: WrapDatasetVariant slash
2019-11-19 11:23:13.497411: E
tensorflow/core/framework/op_kernel.cc:1325 slash ] OpKernel ('op:
'WrapDatasetVariant' device_type: 'CPU'') for unknown op:
WrapDatasetVariant slash
2019-11-19 11:23:13.497533: E
tensorflow/core/framework/op_kernel.cc:1325 slash ] OpKernel ('op:
'UnwrapDatasetVariant' device_type: 'GPU' host_memory_arg:
'input_handle' host_memory_arg: 'output_handle'') for unknown op:
UnwrapDatasetVariant slash
Specified model file version (0) is incompatible with minimum version
supported by this client (1). See
https://github.com/mozilla/DeepSpeech/#model-compatibility for more
information slash
Traceback (most recent call last): slash
File '/home/shubh/tmp/deepspeech-venv/bin/deepspeech', line 8, in slash
sys.exit(main()) slash
File
'/home/shubh/tmp/deepspeech-venv/lib/python3.6/site-packages/deepspeech/client.py',
line 88, in main slash
ds = Model(args.model, N_FEATURES, N_CONTEXT, args.alphabet,
BEAM_WIDTH) slash
File
'/home/shubh/tmp/deepspeech-venv/lib/python3.6/site-packages/deepspeech/init.py',
line 23, in init slash
raise RuntimeError('CreateModel failed with error code
{}'.format(status)) slash
RuntimeError: CreateModel failed with error code 8195

i used: slash
Tensorflow 1.14.0 but when i used deepspeech slash --v it shows 1.13.0 slash
and used 0.6.0 binary files slash
i was unable to download convert graphdef slash _ 1.13 version so i
downloaded 1.14 convert graphdef and then run .

please help

thanx in advance

[This is an archived TTS discussion thread from discourse.mozilla.org/t/runtimeerror-createmodel-failed-with-error-code-8195]

[JRMeyer]

>>> lissyx
[November 19, 2019, 8:50am]

> Tensorflow 1.14.0 but when i used deepspeech slash --v it shows 1.13.0 slash
> and used 0.6.0 binary files slash
> i was unable to download convert graphdef slash _ 1.13 version so i
> downloaded 1.14 convert graphdef and then run .

It looks like you already figured it out yourself, you mixed versions,
it's not supported. Just ensure you install latest v0.6.0a15 binary.
You may need to force pip with --upgrade deepspeech==0.6.0a15

[Archived Post]

[JRMeyer]

>>> shubhamc247
[November 20, 2019, 8:57am]

if i am installed deepspeech==0.6.0a15 so Tensorflow 1.14.0 and 0.6.0
binary file supported on deepspeech? slash
or i need to install latest version of binary and tensorflow?

[Archived Post]

[JRMeyer]

>>> shubhamc247
[November 20, 2019, 9:03am]

if i write command pip3 list it shows ds-ctcdecoder 0.6.0a15 so it can
ok with upgraded version of deepspeech?

[Archived Post]

[JRMeyer]

>>> lissyx
[November 20, 2019, 9:07am]

> if i am installed deepspeech==0.6.0a15 so Tensorflow 1.14.0 and 0.6.0
> binary file supported on deepspeech? slash
> or i need to install latest version of binary and tensorflow?

You don't need to care about TensorFlow at all.

[Archived Post]

[JRMeyer]

>>> lissyx
[November 20, 2019, 9:08am]

> if i write command pip3 list it shows ds-ctcdecoder 0.6.0a15 so it can
> ok with upgraded version of deepspeech?

Yes, ds_ctcdecoder should match the DeepSpeech version as well.

[Archived Post]

[JRMeyer]

>>> shubhamc247
[November 22, 2019, 9:11am]

hello, slash
I trained data and then create .pbmm file but when i predict audio file
by following command slash
./export/alphabet.txt slash --lm ./export/lm.binary slash --trie ./export/trie
slash --audio ./export/40196.wav

it predict only 'i' ? slash
please tell me where i wrong?

deepspeech slash --model ./export/output_graph.pb slash --alphabet
./export/alphabet.txt slash --lm ./export/lm.binary slash --trie ./export/trie
slash --audio ./export/40196.wav slash
Loading model from file ./export/output_graph.pb slash
TensorFlow: v1.13.1-10-g3e0cc53 slash
DeepSpeech: v0.5.1-0-g4b29b78 slash
Warning: reading entire model file into memory. Transform model file
into an mmapped graph to reduce heap usage. slash
2019-11-22 14:34:58.722356: I
tensorflow/core/platform/cpu_feature_guard.cc:141 slash ] Your CPU supports
instructions that this TensorFlow binary was not compiled to use: AVX2
FMA slash
2019-11-22 14:35:00.561618: E
tensorflow/core/framework/op_kernel.cc:1325 slash ] OpKernel ('op:
'UnwrapDatasetVariant' device_type: 'CPU'') for unknown op:
UnwrapDatasetVariant slash
2019-11-22 14:35:00.561687: E
tensorflow/core/framework/op_kernel.cc:1325 slash ] OpKernel ('op:
'WrapDatasetVariant' device_type: 'GPU' host_memory_arg: 'input_handle'
host_memory_arg: 'output_handle'') for unknown op: WrapDatasetVariant slash
2019-11-22 14:35:00.561706: E
tensorflow/core/framework/op_kernel.cc:1325 slash ] OpKernel ('op:
'WrapDatasetVariant' device_type: 'CPU'') for unknown op:
WrapDatasetVariant slash
2019-11-22 14:35:00.561874: E
tensorflow/core/framework/op_kernel.cc:1325 slash ] OpKernel ('op:
'UnwrapDatasetVariant' device_type: 'GPU' host_memory_arg:
'input_handle' host_memory_arg: 'output_handle'') for unknown op:
UnwrapDatasetVariant slash
Loaded model in 1.84s. slash
Loading language model from files ./export/lm.binary ./export/trie slash
Loaded language model in 1.01e+02s. slash
Running inference. slash
2019-11-22 14:36:42.269189: W
tensorflow/core/framework/allocator.cc:124 slash ] Allocation of 16777216
exceeds 10% of system memory. slash
2019-11-22 14:36:42.283837: W
tensorflow/core/framework/allocator.cc:124 slash ] Allocation of 16777216
exceeds 10% of system memory. slash
2019-11-22 14:36:42.305110: W
tensorflow/core/framework/allocator.cc:124 slash ] Allocation of 134217728
exceeds 10% of system memory. slash
2019-11-22 14:36:42.485505: W
tensorflow/core/framework/allocator.cc:124 slash ] Allocation of 16777216
exceeds 10% of system memory. slash
2019-11-22 14:36:42.497546: W
tensorflow/core/framework/allocator.cc:124 slash ] Allocation of 16777216
exceeds 10% of system memory. slash
i slash
Inference took 18.559s for 6.976s audio file.

[Archived Post]

[JRMeyer]

>>> lissyx
[November 22, 2019, 12:27pm]

> please tell me where i wrong?

We can't help you without more informations on the training you
performed.

[Archived Post]

[JRMeyer]

>>> shubhamc247
[November 23, 2019, 5:38am]

ok slash
i used deepspeech 5.0 & ds-ctcdecoder 5.0 slash
and trained model on 910 (.wav) file using following command: slash
python3 DeepSpeech.py slash --checkpoint_dir /cpoint/ slash --train_batch_size 10
slash --test_batch_size 10 slash --dev_batch_size 10 slash --epochs 2 slash --train_files
clips/TRAIN/train.csv slash --dev_files clips/DEV/dev.csv slash --test_files
clips/TEST/test.csv slash --learning_rate 0.01 slash --export_dir export/ slash
is this information required slash
?

[Archived Post]

[JRMeyer]

>>> lissyx
[November 23, 2019, 9:16am]

So basically, small dataset, your model was train on 2 epochs, it learnt
mostly nothing.

[Archived Post]

[JRMeyer]

>>> shubhamc247
[November 23, 2019, 9:42am]

ok...i will train on 5 or more epochs. slash
thanx

[Archived Post]

[JRMeyer]

>>> lissyx
[November 23, 2019, 10:46am]

> ok...i will train on 5 or more epochs.

You would need much more than that, and smaller geometry as well, given
your dataset.

[Archived Post]

[JRMeyer]

>>> shubhamc247
[November 25, 2019, 1:44pm]

if i am trained on 700 .wav file and each file has 10 second , so total
116 min data. when i used all default parameter then it was taked more
than 2 day, Can you provide some hints about the total training time ? slash
system information : slash
8gb ram slash
i5 intel processor slash
and used cpu.

if i am using GPU instead of cpu then how much it will affected for
training? slash
Thanx in Advance

[Archived Post]

[JRMeyer]

>>> lissyx
[November 25, 2019, 2:00pm]

> if i am trained on 700 .wav file and each file has 10 second , so
> total 116 min data.

This amount of data is really really really low. I don't think you can
get anything out of just that.

> if i am using GPU instead of cpu then how much it will affected for
> training?

That depends on the GPU you choose, but anything decent enough should
take care of that amount of data in a matter of minutes.

[Archived Post]

[JRMeyer]

>>> shubhamc247
[November 25, 2019, 2:15pm]

ok, can u please tell me that how much (minimum) size of data required
for training?

[Archived Post]

[JRMeyer]

>>> lissyx
[November 25, 2019, 2:20pm]

It's already documented and it depends on languages, but you can go with
thousands of hours of audio.

[Archived Post]