phased_lstm_keras problem? add_weight () got multiple values for argument 'name'

[jabowery]

The below error appeared in https://github.com/abr/neurips2019/blob/master/experiments/psMNIST-phased-lstm.ipynb

This has been reported as a keras issue with a suggested workaround. that may be applied in PhasedLSTM-Keras.

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-10-c56ccd8ad126> in <module>
      4 model.add(PLSTM(200, input_shape=(n_pixels, 1), 
      5                 implementation=0,
----> 6                alpha=0.0001)
      7          )
      8 model.add(Dense(10, activation='softmax'))

~\Miniconda3\envs\lmu\lib\site-packages\keras\engine\sequential.py in add(self, layer)
    164                     # and create the node connecting the current layer
    165                     # to the input layer we just created.
--> 166                     layer(x)
    167                     set_inputs = True
    168             else:

~\Miniconda3\envs\lmu\lib\site-packages\keras\legacy\layers.py in __call__(self, inputs, initial_state, **kwargs)
    511         # modify the input spec to include the state.
    512         if initial_state is None:
--> 513             return super(Recurrent, self).__call__(inputs, **kwargs)
    514 
    515         initial_state = to_list(initial_state, allow_tuple=True)

~\Miniconda3\envs\lmu\lib\site-packages\keras\engine\base_layer.py in __call__(self, inputs, **kwargs)
    461                                          'You can build it manually via: '
    462                                          '`layer.build(batch_input_shape)`')
--> 463                 self.build(unpack_singleton(input_shapes))
    464                 self.built = True
    465 

~\Miniconda3\envs\lmu\lib\site-packages\phased_lstm_keras\PhasedLSTM.py in build(self, input_shape)
    190                                       initializer=self.kernel_initializer,
    191                                       regularizer=self.kernel_regularizer,
--> 192                                       constraint=self.kernel_constraint)
    193 
    194         self.recurrent_kernel = self.add_weight(

TypeError: add_weight() got multiple values for argument 'name'

[AndeYashwanth]

I tried the suggested fix but it doesn't work.
I still get the same error.

self.kernel = self.add_weight(shape=(self.input_dim, self.units * 4),
                                      name='kernel',
                                      initializer=self.kernel_initializer,
                                      regularizer=self.kernel_regularizer,
                                      constraint=self.kernel_constraint)

        self.recurrent_kernel = self.add_weight(
            shape=(self.units, self.units * 4),
            name='recurrent_kernel',
            initializer=self.recurrent_initializer,
            regularizer=self.recurrent_regularizer,
            constraint=self.recurrent_constraint)

        if self.use_bias:
            self.bias = self.add_weight(shape=(self.units * 4,),
                                        name='bias',
                                        initializer=self.bias_initializer,
                                        regularizer=self.bias_regularizer,
                                        constraint=self.bias_constraint)

[doubleapple123]

There are 4 instances where you need to name the first argument as shape= , this seems to have fixed this particular issue for me, I think you're missing one

[zd23]

can you please show the 4 instance cz ive found 3 in mine too thanks in advance !
#doubleapple123

[AutoUpdatingBSoD]

@zd23
This is the code that ended up solving this issue for me.

        self.timegate_kernel = self.add_weight(
            shape=(3, self.units),
            name='timegate_kernel',
            initializer=self.timegate_initializer,
            regularizer=self.timegate_regularizer,
            constraint=self.timegate_constraint)
        self.built = True`

The code before is missing the explicit definition of shape=