The using-lstm-network-for-sentiment-analysis from sumanismcse

sumanismcse / using-lstm-network-for-sentiment-analysis Goto Github PK

View Code? Open in Web Editor NEW

Using Long Short Term Memory Model and Recursive Neural Network for Sentiment Analysis to use in the automated therapeutic app as an input to read intent of user in Google's cloud natural language processing service 'dialogflow' and on that basis we provide low cost therapy

Jupyter Notebook 76.21% Python 23.79%

using-lstm-network-for-sentiment-analysis's People

Contributors

Stargazers

Watchers

using-lstm-network-for-sentiment-analysis's Issues

Convert pytorch model to onnx file and onnx file to tensorflow model for better data serving in the app

### When I tried to convert pytorch model to onnx file,This Happened:

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-33-a9d25627b347> in <module>()
     10 # Export the trained model to ONNX
     11 dummy_input = Variable(torch.randn(1, 1, 28, 28)) # one black and white 28 x 28 picture will be the input to the model
---> 12 torch.onnx.export(trained_model, dummy_input, "sentiment.onnx")

9 frames
/usr/local/lib/python3.6/dist-packages/torch/nn/modules/module.py in _slow_forward(self, *input, **kwargs)
    529         tracing_state._traced_module_stack.append(self)
    530         try:
--> 531             result = self.forward(*input, **kwargs)
    532         finally:
    533             tracing_state.pop_scope()

TypeError: forward() missing 1 required positional argument: 'hidden'

###The same happened on the onnx file to tensorflow model:

----------------------------------------------------------------------------------------------------
---------------------------------------------------------------------------
ModuleNotFoundError                       Traceback (most recent call last)
<ipython-input-34-a405ea78dbb6> in <module>()
      1 
----> 2 import onnx
      3 import warnings
      4 from onnx_tf.backend import prepare
      5 

ModuleNotFoundError: No module named 'onnx'

---------------------------------------------------------------------------
NOTE: If your import is failing due to a missing package, you can
manually install dependencies using either !pip or !apt.

To view examples of installing some common dependencies, click the
"Open Examples" button below.
---------------------------------------------------------------------------

Recommend Projects