jikechao opened a new issue, #14721:
URL: https://github.com/apache/tvm/issues/14721

   
   
   ### Expected behavior
   
   the predicted result in tvm should be the same as keras.
   
   ### Actual behavior
   
   inconsistent inference results between keras and tvm.
   
![image](https://user-images.githubusercontent.com/29506758/234322139-b7159ced-fc75-4477-b24e-e53685c15a08.png)
   
   
   
   
   ### Steps to reproduce
   
   ```
   import tvm
   import tvm.relay as relay
   import numpy as np
   from tensorflow import keras
   from tensorflow.keras import layers, models
   
   
   input_shape = (1, 2, 1)
   input_data = np.random.random(input_shape)
   x = layers.Input(shape=input_shape[1:], dtype='float32')
   
   layer = keras.layers.Softmax()
   layer.set_weights(layer.get_weights())
   
   y = layer(x)
   model = models.Model(x, y)
   # print(model.summary())
   res_keras = model.predict(input_data)
   
   shape_dict = {'input_1': input_shape}
   mod, params = relay.frontend.from_keras(model, shape_dict)
   
   with tvm.transform.PassContext(opt_level=3):
       model = relay.build_module.create_executor("graph", mod, tvm.cpu(0), 
'llvm', params).evaluate()
   
   res_tvm = model(tvm.nd.array(input_data.astype('float32'))).numpy()
   
   print('keras infer result:', res_keras)
   print('tvm infer result:', res_tvm)
   np.testing.assert_allclose(res_keras, res_tvm, atol=1e-3, rtol=1e-3)
   ```
   *  frontend:keras
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to