Train problem....

[daniele777]

i follow structure of readme and collect wav file …but …i have this error… help !

precise-train -e 60 tars-start.net tars-start/ 2021-04-30 19:05:34.881606: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library ‘libcudart.so.11.0’; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory 2021-04-30 19:05:34.881702: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine. Using TensorFlow backend. Traceback (most recent call last): File “/home/dani/Scaricati/mycroft-precise-dev/.venv/bin/precise-train”, line 33, in sys.exit(load_entry_point(‘mycroft-precise’, ‘console_scripts’, ‘precise-train’)()) File “/home/dani/Scaricati/mycroft-precise-dev/precise/scripts/base_script.py”, line 43, in run_main script = cls(args) File “/home/dani/Scaricati/mycroft-precise-dev/precise/scripts/train.py”, line 87, in init self.model = create_model(args.model, params) File “/home/dani/Scaricati/mycroft-precise-dev/precise/model.py”, line 76, in create_model model = Sequential() File “/home/dani/Scaricati/mycroft-precise-dev/.venv/lib/python3.9/site-packages/tensorflow/python/training/tracking/base.py”, line 522, in _method_wrapper result = method(self, *args, kwargs) File “/home/dani/Scaricati/mycroft-precise-dev/.venv/lib/python3.9/site-packages/keras/engine/sequential.py”, line 109, in init super(functional.Functional, self).init( # pylint: disable=bad-super-call File “/home/dani/Scaricati/mycroft-precise-dev/.venv/lib/python3.9/site-packages/tensorflow/python/training/tracking/base.py”, line 522, in _method_wrapper result = method(self, args, kwargs) File “/home/dani/Scaricati/mycroft-precise-dev/.venv/lib/python3.9/site-packages/keras/engine/training.py”, line 245, in init super(Model, self).init(kwargs) File “/home/dani/Scaricati/mycroft-precise-dev/.venv/lib/python3.9/site-packages/tensorflow/python/training/tracking/base.py”, line 522, in _method_wrapper result = method(self, args, kwargs) File “/home/dani/Scaricati/mycroft-precise-dev/.venv/lib/python3.9/site-packages/keras/engine/base_layer.py”, line 350, in init self._init_set_name(name) File “/home/dani/Scaricati/mycroft-precise-dev/.venv/lib/python3.9/site-packages/keras/engine/functional.py”, line 501, in _init_set_name self._name = backend.unique_object_name( AttributeError: module ‘keras.backend’ has no attribute ‘unique_object_name’

[daniele777]

No suggestion?? i try to modify keras version ... same problem

[el-tocino]

what version of tensorflow? keras? did setup.sh complete without errors?

[daniele777]

Hi!! Tensorflow 2.5.0rc1 keras 2.1.5... no problem during installation ...

[el-tocino]

Did you download the tf2 fork of precise?

[daniele777]

I download and install from main page menù .. so i can't have Nvidia card possibile Is this problem ? I have AMD video card

[el-tocino]

It doesn't require a gpu.

You definitely had problems running setup, or you didn't run it: 'tensorflow>=1.13,<1.14'

[daniele777]

When i try to run pip install tensorflow==1.13.1 he says : Error no matching distribuition found for tensorflow 1.13.1

[el-tocino]

Yeah, you're on python3.9, you'd need to be on 3.7 or less or else compile your own locally.

[daniele777]

You are right and can i install python 3.7 and leave 3.9 ? And create venv from python 3.7 ?

[el-tocino]

It really depends on your skill and comfort. Maybe use a different host with older versions?

[daniele777]

Yeah best solution i use different host !!!

Thanks Bro !!