iwyoo / convlstmcell-tensorflow Goto Github PK

      X = tf.placeholder([batch_size,time_steps,dim_input])             
      # [batchsize, time_steps, diminput] => [nsteps, batchsize, dim_input]
      X = tf.transpose(X, [1, 0, 2])
      # 2. Reshape input to [nsteps*batchsize, dim_input]
      X = tf.reshape( X, [-1, dim_input])
      # 3. Input layer => Hidden layer
      input_split = tf.matmul(X, _W['hidden']) + _b['hidden']
      # 4. Splite data to 'nsteps' chunks. An i-th chunck indicates i-th batch data
      input_split = tf.split(0,time_steps, input_split)
      lstm_cell = tf.nn.rnn_cell.BasicLSTMCell(num_hidden, \
                                         forget_bias=1.0,state_is_tuple=True)
      lstm_cell = tf.nn.rnn_cell.MultiRNNCell([lstm_cell]*num_layers,\
                                         state_is_tuple=True)
      state     = lstm_cell.zero_state(batch_size,dtype=tf.float32)
      LSTM_O, LSTM_S = tf.nn.rnn(lstm_cell, input_split, \
                                         initial_state=state)

cell = ConvLSTMCell(13)
state = cell.zero_state(16, 46, 46) #batchsize is 16, height and width is 46.
x_unwrap = []
with tf.variable_scope("ConvLSTM") as scope:  # as BasicLSTMCell
    for i, p_input_ in enumerate(_Hsplit):
        if i > 0:
            scope.reuse_variables()
        # ConvCell takes Tensor with size [batch_size, height, width, channel].
        t_output, state = cell(p_input_, state)
        x_unwrap.append(t_output)

x_unwrap = tf.stack(x_unwrap)
x_unwrap = tf.transpose(x_unwrap, [1, 0, 2, 3, 4])
hidden_unwrap = tf.stack(hidden_unwrap)
hidden_unwrap = tf.transpose(hidden_unwrap, [1, 0, 2, 3, 4])