recurrent-neural-network

I'm trying to make RNN using LSTM . I made LSTM model, and after it, there is two DNN network, and one regression output layer. I trained my data, and the final training loss become about 0.009 . However, when i applied the model to test data, the loss become about 0.5 . The 1th epoch training loss is about 0.5 . So, I think the trained variable do not used in test model. The only difference between training and test model is batch size. Trainning Batch = 100~200 , Test Batch Size = 1 . in main function i made LSTM instance. In LSTM innitializer, the model is made. def __init__(self,config

with tf.variable_scope('forward'): cell_img_fwd = tf.nn.rnn_cell.GRUCell(hidden_state_size, hidden_state_size) img_init_state_fwd = rnn_img_mapped[:, 0, :] img_init_state_fwd = tf.multiply( img_init_state_fwd, tf.zeros([batch_size, hidden_state_size])) rnn_outputs2, final_state2 = tf.nn.dynamic_rnn( cell_img_fwd, rnn_img_mapped, initial_state=img_init_state_fwd, dtype=tf.float32) This is my code for a GRU for input of dimension 100x196x50, it should be unpacked along the second dimension (that is 196). hidden_state_size is 50, batch_size is 100. However I get the following error: ValueError:

My question is about the TensorFlow method tf.nn.dynamic_rnn . It returns the output of every time step and the final state. I would like to know if the returned final state is the state of the cell at the maximum sequence length or if it is determined individually by the sequence_length argument. For better understanding an example: I have 3 sequences with length [10,20,30] and getting back the final state [3,512] (if the hidden state of the cell has the length 512). Are the three returned hidden states for the three sequences the state of the cell at time step 30 or am I getting back the

Or, more specific: what is the difference between ConvLSTM2D and ConvLSTM2DCell ? What is the difference between SimpleRNN and SimpleRNNCell ? Same question for GRU and GRUCell Keras manuals are not very verbose here. I can see from RTFS (reading those fine sources) that these classes are descendants of different base classes. Those, with names, ending with Cell , are subclasses of Layer . In my task I need to classify video sequences. That is, my classifier's input is a sequence of video frames, and the output is a single label (one-hot encoded vector). What class should I use? The difference