keras - tensorflow - LSTM - csv - how to use fit_generator

Question

Iam tyring to implement a simple RNN LSTM model but stuck. The problem itself is simple. I will be giving 5 consecutive digits to the model (but 1 digit at a time) and then I want the model to predict the 6th one.

Example: Input data: 1, 2, 3, 4, 5 (1 digit at each time step) And the output for this sequence should be 6 .

I have a csv file in which:

• The first row is the headers
• There are 6 columns
• The first column is the ID only. Not used in training.
• The next 5 columns are input data (x)
• And the last column is the label (y)

I want to develop a model with Keras and make it successfully guess the 6th number.

HEre is what I do:

1) First implement some constants that we will need.

NR_FEATURES = 5
ITERATOR_BATCH_SIZE = 1
NR_EPOCHS = 15

2) Define the generator that will be used when training.

def train_data_generator():

    dataset = tf.contrib.data.make_csv_dataset(train_path1, 
                                               batch_size=ITERATOR_BATCH_SIZE, 
                                               num_epochs=NR_EPOCHS, 
                                               shuffle=True)

    iter = dataset.make_one_shot_iterator()
    next = iter.get_next()
    ID = next['ID']
    features = [next['nr1'], next['nr2'], next['nr3'], next['nr4'], next['nr5']]
    features = tf.reshape(features, [NR_FEATURES, 1])
    label = next['next_nr']
    yield (features, label)

3) Create the model and start training.

input_data = Input(shape=(5, 1), name='input_data')
layer1_out = LSTM(1, return_sequences=False)(input_data)  # only return the last output
lstm_model = Model(inputs=input_data, outputs=layer1_out)

lstm_model.compile(loss='mean_absolute_error', optimizer='adam', metrics=['accuracy'])

lstm_model.fit_generator(train_data_generator(), 
                         steps_per_epoch=(150/ITERATOR_BATCH_SIZE),
                         epochs=NR_EPOCHS, 
                         verbose=1)

But it crashes right away...

The error message I get:

Epoch 1/15

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-5-af9dcbcbe289> in <module>()
      8                          steps_per_epoch=(150/ITERATOR_BATCH_SIZE),
      9                          epochs=NR_EPOCHS,
---> 10                          verbose=1)

~/anaconda3/envs/tensorflow/lib/python3.6/site-packages/keras/legacy/interfaces.py in wrapper(*args, **kwargs)
     89                 warnings.warn('Update your `' + object_name +
     90                               '` call to the Keras 2 API: ' + signature, stacklevel=2)
---> 91             return func(*args, **kwargs)
     92         wrapper._original_function = func
     93         return wrapper

~/anaconda3/envs/tensorflow/lib/python3.6/site-packages/keras/engine/training.py in fit_generator(self, generator, steps_per_epoch, epochs, verbose, callbacks, validation_data, validation_steps, class_weight, max_queue_size, workers, use_multiprocessing, shuffle, initial_epoch)
   2212                     # build batch logs
   2213                     batch_logs = {}
-> 2214                     if x is None or len(x) == 0:
   2215                         # Handle data tensors support when no input given
   2216                         # step-size = 1 for data tensors

TypeError: object of type 'Tensor' has no len()

I just do not get it. Does anyone have any idea?

Answer 1

You can convert tensor to numpy by eval() directly.

features = tf.reshape(features, [NR_FEATURES, 1])
# convert tensor to numpy
with tf.Session() as sess:
    features = features.eval()
# Your data shape needs to be adjusted relative to your model input.
features = features.reshape(-1,NR_FEATURES,1) 
label = next['next_nr']
label = np.array([label])
yield (features, label)