ImportError: No module named 'quantized_ops'

[MrLinNing]

sorry, i run your code, but it happens this

(my_env) luhang@intelligence:~/meqnn/QuantizedNeuralNetworks-Keras-Tensorflow$ python train.py config_CIFAR-10 -o lr=0.01 wbits=4 abits=4 network_type='full-qnn'
/home/luhang/my_env/lib/python3.5/site-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.
  from ._conv import register_converters as _register_converters
Using TensorFlow backend.
Traceback (most recent call last):
  File "train.py", line 8, in <module>
    from models.model_factory import build_model
  File "/home/luhang/meqnn/QuantizedNeuralNetworks-Keras-Tensorflow/models/model_factory.py", line 8, in <module>
    from layers.quantized_layers import QuantizedConv2D,QuantizedDense
  File "/home/luhang/meqnn/QuantizedNeuralNetworks-Keras-Tensorflow/layers/quantized_layers.py", line 10, in <module>
    from quantized_ops import quantize, clip_through
ImportError: No module named 'quantized_ops'

[BertMoons]

Hi,

• Make sure the quantized_ops.py file exists under ./layers/
• Your command is not exactly right, that might have something to do with this. You should either run:
  python train.py -c config_CIFAR-10 -o lr=0.01 wbits=4 abits=4 network_type='full-qnn', or:
  ./train.sh config_CIFAR-10 -o lr=0.01 wbits=4 abits=4 network_type='full-qnn'

Hopefully this works, good luck!

Bert

[MrLinNing]

but, i am sure the quantized_ops.py file exists under ./layers/

And most import thing that I used python 3.5， how can I run the code？

[xmicroshell]

Hi, LinNing
This problem is a common error.and i dont think it has anything to do with the python version.
You should replace all the sentences:
from quantized_ops import quantize
with:
from layers.quantized_ops import quantize
i meet other similar import errors, just follow the above to resolve it

[MrLinNing]

Thank you！BertMoons and xmicroshell.
I want to deploy trained weights and the network to my embedding devices, and how to see the trained weights value ？

[xmicroshell]

If you want to use the model in your device,i think the function model.save_weights(filepath) could help.
If you want to see the weight by yourself, maybe tensorboard can help you (but i've used it）