I'm trying a very simple example for tensorflow RNN.
In that example, I use dynamic rnn. The code is as follows:
data = tf.placeholder(tf.float32, [None, 10,1]) #Number of examples, number of input, dimension of each input
target = tf.placeholder(tf.float32, [None, 11])
num_hidden = 24
cell = tf.nn.rnn_cell.LSTMCell(num_hidden,state_is_tuple=True)
val, _ = tf.nn.dynamic_rnn(cell, data, dtype=tf.float32)
val = tf.transpose(val, [1, 0, 2])
last = tf.gather(val, int(val.get_shape()[0]) - 1)
weight = tf.Variable(tf.truncated_normal([num_hidden, int(target.get_shape()[1])]))
bias = tf.Variable(tf.constant(0.1, shape=[target.get_shape()[1]]))
prediction = tf.nn.softmax(tf.matmul(last, weight) + bias)
cross_entropy = -tf.reduce_sum(target * tf.log(tf.clip_by_value(prediction,1e-10,1.0)))
optimizer = tf.train.AdamOptimizer()
minimize = optimizer.minimize(cross_entropy)
mistakes = tf.not_equal(tf.argmax(target, 1), tf.argmax(prediction, 1))
error = tf.reduce_mean(tf.cast(mistakes, tf.float32))
Actually, the code is taken from this tutorial.
The input to this RNN network is a sequence of binary numbers. Each number is put into an array. For example, a seuquence has format:
[[1],[0],[0],[1],[1],[0],[1],[1],[1],[0]]
The shape of the input is [None,10,1] which are batch size, sequence size and embedding size, respectively. Now because dynamic rnn can accept variable input shape, I change the code as follows:
data = tf.placeholder(tf.float32, [None, None,1])
Basically, I want to use variable-length sequences (of course same length for all sequences in the same batch, but different between batches). However, it throws the error:
Traceback (most recent call last):
File "rnn-lstm-variable-length.py", line 48, in <module>
last = tf.gather(val, int(val.get_shape()[0]) - 1)
TypeError: __int__ returned non-int (type NoneType)
I understand that the second dimension is None, which cannot be used in get_shape()[0]. However, I believe that there must be a way to overcome this because RNN accepts variable lenth inputs, in general.
How can I do it?
tl;dr: try using tf.batch(..., dynamic_pad=True) to batch your data.
#chris_anderson's comment is correct. Ultimately your network needs a dense matrix of numbers to work with and there are a couple of strategies to convert variable length data into hyperrectangles:
Pad all batches to a fixed size (e.g. assume a maximum length of say 500 items per input and every item in every batch is padded to 500). There is nothing dynamic about this strategy.
Apply padding per-batch to the length of the longest item in the batch (dynamic padding).
Bucket your input based on length and apply padding per-batch. This is the same as #2, but with less overall padding.
There are other strategies that you could use too.
To do this batching, you use:
tf.train.batch - by default it does no padding, you need to implement it yourself.
tf.train.batch(..., dynamic_pad=True)
tf.contrib.training.bucket_by_sequence_length
I suspect you're also confused by the use of tf.nn.dynamic_rnn. It's important to note that the dynamic in dynamic_rnn refers to the way that TensorFlow unrolls the recurrent part of the network. in tf.nn.rnn, the recurrence is done statically in the graph (there is no internal loop, it's unrolled at graph construction time). In dynamic_rnn however, TensorFlow uses tf.while_loop to iterate inside the graph at run time. To use dynamic padding, you need to use dynamic unrolling, but it does not do it automatically.
tf.gather expects a tensor, so you can use tf.shape(val) to get a tensor, calculated at run-time, for the shape of val - e.g. tf.gather(val, tf.shape(val)[0] - 1)
Related
I'm working with padded sequences of maximum length 50. I have two types of sequence data:
1) A sequence, seq1, of integers (1-100) that correspond to event types (e.g. [3,6,3,1,45,45....3]
2) A sequence, seq2, of integers representing time, in minutes, from the last event in seq1. So the last element is zero, by definition. So for example [100, 96, 96, 45, 44, 12,... 0]. seq1 and seq2 are the same length, 50.
I'm trying to run the LSTM primarily on the event/seq1 data, but have the time/seq2 strongly influence the forget gate within the LSTM. The reason for this is I want the LSTM to tend to really penalize older events and be more likely to forget them. I was thinking about multiplying the forget weight by the inverse of the current value of the time/seq2 sequence. Or maybe (1/seq2_element + 1), to handle cases where it's zero minutes.
I see in the keras code (LSTMCell class) where the change would have to be:
f = self.recurrent_activation(x_f + K.dot(h_tm1_f,self.recurrent_kernel_f))
So I need to modify keras' LSTM code to accept multiple inputs. As an initial test, within the LSTMCell class, I changed the call function to look like this:
def call(self, inputs, states, training=None):
time_input = inputs[1]
inputs = inputs[0]
So that it can handle two inputs given as a list.
When I try running the model with the Functional API:
# Input 1: event type sequences
# Take the event integer sequences, run them through an embedding layer to get float vectors, then run through LSTM
main_input = Input(shape =(max_seq_length,), dtype = 'int32', name = 'main_input')
x = Embedding(output_dim = embedding_length, input_dim = num_unique_event_symbols, input_length = max_seq_length, mask_zero=True)(main_input)
## Input 2: time vectors
auxiliary_input = Input(shape=(max_seq_length,1), dtype='float32', name='aux_input')
m = Masking(mask_value = 99999999.0)(auxiliary_input)
lstm_out = LSTM(32)(x, time_vector = m)
# Auxiliary loss here from first input
auxiliary_output = Dense(1, activation='sigmoid', name='aux_output')(lstm_out)
# An abitrary number of dense, hidden layers here
x = Dense(64, activation='relu')(lstm_out)
# The main output node
main_output = Dense(1, activation='sigmoid', name='main_output')(x)
## Compile and fit the model
model = Model(inputs=[main_input, auxiliary_input], outputs=[main_output, auxiliary_output])
model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'], loss_weights=[1., 0.2])
print(model.summary())
np.random.seed(21)
model.fit([train_X1, train_X2], [train_Y, train_Y], epochs=1, batch_size=200)
However, I get the following error:
An `initial_state` was passed that is not compatible with `cell.state_size`. Received `state_spec`=[InputSpec(shape=(None, 50, 1), ndim=3)]; however `cell.state_size` is (32, 32)
Any advice?
You can't pass a list of inputs to default recurrent layers in Keras. The input_spec is fixed and the recurrent code is implemented based on single tensor input also pointed out in the documentation, ie it doesn't magically iterate over 2 inputs of same timesteps and pass that to the cell. This is partly because of how the iterations are optimised and assumptions made if the network is unrolled etc.
If you like 2 inputs, you can pass constants (doc) to the cell which will pass the tensor as is. This is mainly to implement attention models in the future. So 1 input will iterate over timesteps while the other will not. If you really like 2 inputs to be iterated like a zip() in python, you will have to implement a custom layer.
I would like to throw in a different ideas here. They don't require you to modify the Keras code.
After the embedding layer of the event types, stack the embeddings with the elapsed time. The Keras function is keras.layers.Concatenate(axis=-1). Imagine this, a single even type is mapped to a n dimensional vector by the embedding layer. You just add the elapsed time as one more dimension after the embedding so that it becomes a n+1 vector.
Another idea, sort of related to your problem/question and may help here, is 1D convolution. The convolution can happen right after the concatenated embeddings. The intuition for applying convolution to event types and elapsed time is actually 1x1 convolution. In such a way that you linearly combine the two together and the parameters are trained. Note in terms of convolution, the dimensions of the vectors are called channels. Of course, you can also convolve more than 1 event at a step. Just try it. It may or may not help.
I am new to Keras and I am trying to implement a model for an image captioning project.
I am trying to reproduce the model from Image captioning pre-inject architecture (The picture is taken from this paper: Where to put the image in an image captioning generator) (but with a minor difference: generating a word at each time step instead of only generating a single word at the end), in which the inputs for the LSTM at the first time step are the embedded CNN features. The LSTM should support variable input length and in order to do this I padded all the sequences with zeros so that all of them have maxlen time steps.
The code for the model I have right now is the following:
def get_model(model_name, batch_size, maxlen, voc_size, embed_size,
cnn_feats_size, dropout_rate):
# create input layer for the cnn features
cnn_feats_input = Input(shape=(cnn_feats_size,))
# normalize CNN features
normalized_cnn_feats = BatchNormalization(axis=-1)(cnn_feats_input)
# embed CNN features to have same dimension with word embeddings
embedded_cnn_feats = Dense(embed_size)(normalized_cnn_feats)
# add time dimension so that this layer output shape is (None, 1, embed_size)
final_cnn_feats = RepeatVector(1)(embedded_cnn_feats)
# create input layer for the captions (each caption has max maxlen words)
caption_input = Input(shape=(maxlen,))
# embed the captions
embedded_caption = Embedding(input_dim=voc_size,
output_dim=embed_size,
input_length=maxlen)(caption_input)
# concatenate CNN features and the captions.
# Ouput shape should be (None, maxlen + 1, embed_size)
img_caption_concat = concatenate([final_cnn_feats, embedded_caption], axis=1)
# now feed the concatenation into a LSTM layer (many-to-many)
lstm_layer = LSTM(units=embed_size,
input_shape=(maxlen + 1, embed_size), # one additional time step for the image features
return_sequences=True,
dropout=dropout_rate)(img_caption_concat)
# create a fully connected layer to make the predictions
pred_layer = TimeDistributed(Dense(units=voc_size))(lstm_layer)
# build the model with CNN features and captions as input and
# predictions output
model = Model(inputs=[cnn_feats_input, caption_input],
outputs=pred_layer)
optimizer = Adam(lr=0.0001,
beta_1=0.9,
beta_2=0.999,
epsilon=1e-8)
model.compile(loss='categorical_crossentropy',optimizer=optimizer)
model.summary()
return model
The model (as it is above) compiles without any errors (see: model summary) and I managed to train it using my data. However, it doesn't take into account the fact that my sequences are zero-padded and the results won't be accurate because of this. When I try to change the Embedding layer in order to support masking (also making sure that I use voc_size + 1 instead of voc_size, as it's mentioned in the documentation) like this:
embedded_caption = Embedding(input_dim=voc_size + 1,
output_dim=embed_size,
input_length=maxlen, mask_zero=True)(caption_input)
I get the following error:
Traceback (most recent call last):
File "/export/home/.../py3_env/lib/python3.5/site-packages/tensorflow/python/framework/ops.py", line 1567, in _create_c_op
c_op = c_api.TF_FinishOperation(op_desc)
tensorflow.python.framework.errors_impl.InvalidArgumentError: Dimension 0 in both shapes must be equal, but are 200 and 1. Shapes are [200] and [1]. for 'concatenate_1/concat_1' (op: 'ConcatV2') with input shapes: [?,1,200], [?,25,1], [] and with computed input tensors: input[2] = <1>
I don't know why it says the shape of the second array is [?, 25, 1], as I am printing its shape before the concatenation and it's [?, 25, 200] (as it should be).
I don't understand why there'd be an issue with a model that compiles and works fine without that parameter, but I assume there's something I am missing.
I have also been thinking about using a Masking layer instead of mask_zero=True, but it should be before the Embedding and the documentation says that the Embedding layer should be the first layer in a model (after the input).
Is there anything I could change in order to fix this or is there a workaround to this ?
The non-equal shape error refers to the mask rather than the tensors/inputs. With concatenate supporting masking, it need to handle mask propagation. Your final_cnn_feats doesn't have mask (None), while your embedded_caption has a mask of shape (?, 25). You can find this out by doing:
print(embedded_caption._keras_history[0].compute_mask(caption_input))
Since final_cnn_feats has no mask, concatenate will give it a all non-zero mask for proper mask propagation. While this is correct, the shape of the mask, however, has the same shape as final_cnn_feats which is (?, 1, 200) rather than (?, 1), i.e. masking all features at all time step rather than just all time step. This is where the non-equal shape error comes from ((?, 1, 200) vs (?, 25)).
To fix it, you need to give final_cnn_feats a correct/matching mask. Now I'm not familiar with your project here. One option is to apply a Masking layer to final_cnn_feats, since it is designed to mask timestep(s).
final_cnn_feats = Masking()(RepeatVector(1)(embedded_cnn_feats))
This can be correct only when not all 200 features in final_cnn_feats are zero, i.e. there is always at least one non-zero value in final_cnn_feats. With that condition, Masking layer will give a (?, 1) mask and will not mask the single time step in final_cnn_feats.
I have a 1D input signal. I want to compute autocorrelation as the part of the neural net for further use inside the network.
I need to perform convolution of input with input itself.
To perform convolution in keras custom layer/ tensorflow. We need the following parameters
data shape is "[batch, in_height, in_width, in_channels]",
filter shape is "[filter_height, filter_width, in_channels, out_channels]
There is no batch present in filter shape, which needs to be input in my case
TensorFlow now has an auto_correlation function. It should be in release 1.6. If you build from source you can use it right now (see e.g. the github code).
Here is a possible solution.
By self convolution, I understood a regular convolution where the filter is exactly the same as the input (if it's not that, sorry for my misunderstanding).
We need a custom function for that, and a Lambda layer.
At first I used padding = 'same' which brings outputs with the same length as the inputs. I'm not sure about what output length you want exactly, but if you want more, you should add padding yourself before doing the convolution. (In the example with length 7, for a complete convolution from one end to another, this manual padding would include 6 zeros before and 6 zeros after the input length, and use padding = 'valid'. Find the backend functions here)
Working example - Input (5,7,2)
from keras.models import Model
from keras.layers import *
import keras.backend as K
batch_size = 5
length = 7
channels = 2
channels_batch = batch_size*channels
def selfConv1D(x):
#this function unfortunately needs to know previously the shapes
#mainly because of the for loop, for other lines, there are workarounds
#but these workarounds are not necessary since we'll have this limitation anyway
#original x: (batch_size, length, channels)
#bring channels to the batch position:
x = K.permute_dimensions(x,[2,0,1]) #(channels, batch_size, length)
#suppose channels are just individual samples (since we don't mix channels)
x = K.reshape(x,(channels_batch,length,1))
#here, we get a copy of x reshaped to match filter shapes:
filters = K.permute_dimensions(x,[1,2,0]) #(length, 1, channels_batch)
#now, in the lack of a suitable available conv function, we make a loop
allChannels = []
for i in range (channels_batch):
f = filters[:,:,i:i+1]
allChannels.append(
K.conv1d(
x[i:i+1],
f,
padding='same',
data_format='channels_last'))
#although channels_last is my default config, I found this bug:
#https://github.com/fchollet/keras/issues/8183
#convolution output: (1, length, 1)
#concatenate all results as samples
x = K.concatenate(allChannels, axis=0) #(channels_batch,length,1)
#restore the original form (passing channels to the end)
x = K.reshape(x,(channels,batch_size,length))
return K.permute_dimensions(x,[1,2,0]) #(batch_size, length, channels)
#input data for the test:
x = np.array(range(70)).reshape((5,7,2))
#little model that just performs the convolution
inp= Input((7,2))
out = Lambda(selfConv1D)(inp)
model = Model(inp,out)
#checking results
p = model.predict(x)
for i in range(5):
print("x",x[i])
print("p",p[i])
You can just use tf.nn.conv3d by treating the "batch size" as "depth":
# treat the batch size as depth.
data = tf.reshape(input_data, [1, batch, in_height, in_width, in_channels])
kernel = [filter_depth, filter_height, filter_width, in_channels, out_channels]
out = tf.nn.conv3d(data, kernel, [1,1,1,1,1], padding='SAME')
I have the following situation:
I want to deploy a face detector model using Tensorflow Serving: https://www.tensorflow.org/serving/.
In Tensorflow Serving, there is a command line option called --enable_batching. This causes the model server to automatically batch the requests to maximize throughput. I want this to be enabled.
My model takes in a set of images (called images), which is a tensor of shape (batch_size, 640, 480, 3).
The model has two outputs: (number_of_faces, 4) and (number_of_faces,). The first output will be called faces. The last output, which we can call partitions is the index in the original batch for the corresponding face. For example, if I pass in a batch of 4 images and get 7 faces, then I might have this tensor as [0, 0, 1, 2, 2, 2, 3]. The first two faces correspond to the first image, the third face for the second image, the 3rd image has 3 faces, etc.
My issue is this:
In order for the --enable_batching flag to work, the output from my model needs to have the 0th dimension the same as the input. That is, I need a tensor with the following shape: (batch_size, ...). I suppose this is so that the model server can know which grpc connection to send each output in the batch towards.
What I want to do is to convert my output tensor from the face detector from this shape (number_of_faces, 4) to this shape (batch_size, None, 4). That is, an array of batches, where each batch can have a variable number of faces (e.g. one image in the batch may have no faces, and another might have 3).
What I tried:
tf.dynamic_partition. On the surface, this function looks perfect. However, I ran into difficulties after realizing that the num_partitions parameter cannot be a tensor, only an integer:
tensorflow_serving_output = tf.dynamic_partition(faces, partitions, batch_size)
If the tf.dynamic_partition function were to accept tensor values for num_partition, then it seems that my problem would be solved. However, I am back to square one since this is not the case.
Thank you all for your help! Let me know if anything is unclear
P.S. Here is a visual representation of the intended process:
I ended up finding a solution to this using TensorArray and tf.while_loop:
def batch_reconstructor(tensor, partitions, batch_size):
"""
Take a tensor of shape (batch_size, 4) and a 1-D partitions tensor as well as the scalar batch_size
And reconstruct a TensorArray that preserves the original batching
From the partitions, we can get the maximum amount of tensors within a batch. This will inform the padding we need to use.
Params:
- tensor: The tensor to convert to a batch
- partitions: A list of batch indices. The tensor at position i corresponds to batch # partitions[i]
"""
tfarr = tf.TensorArray(tf.int32, size=batch_size, infer_shape=False)
_, _, count = tf.unique_with_counts(partitions)
maximum_tensor_size = tf.cast(tf.reduce_max(count), tf.int32)
padding_tensor_index = tf.cast(tf.gather(tf.shape(tensor), 0), tf.int32)
padding_tensor = tf.expand_dims(tf.cast(tf.fill([4], -1), tf.float32), axis=0) # fill with [-1, -1, -1, -1]
tensor = tf.concat([tensor, padding_tensor], axis=0)
def cond(i, acc):
return tf.less(i, batch_size)
def body(i, acc):
partition_indices = tf.reshape(tf.cast(tf.where(tf.equal(partitions, i)), tf.int32), [-1])
partition_size = tf.gather(tf.shape(partition_indices), 0)
# concat the partition_indices with padding_size * padding_tensor_index
padding_size = tf.subtract(maximum_tensor_size, partition_size)
padding_indices = tf.reshape(tf.fill([padding_size], padding_tensor_index), [-1])
partition_indices = tf.concat([partition_indices, padding_indices], axis=0)
return (tf.add(i, 1), acc.write(i, tf.gather(tensor, partition_indices)))
_, reconstructed = tf.while_loop(
cond,
body,
(tf.constant(0), tfarr),
name='batch_reconstructor'
)
reconstructed = reconstructed.stack()
return reconstructed
I am aware that there is a similar topic at LSTM Followed by Mean Pooling, but that is about Keras and I work in pure TensorFlow.
I have an LSTM network where the recurrence is handled by:
outputs, final_state = tf.nn.dynamic_rnn(cell,
embed,
sequence_length=seq_lengths,
initial_state=initial_state)
where I pass the correct sequence lengths for each sample (padding by zeros). In any case, outputs contains irrelevant outputs since some samples produce longer outputs than others, based on sequence lengths.
Right now I'm extracting the last relevant output by means of the following method:
def extract_axis_1(data, ind):
"""
Get specified elements along the first axis of tensor.
:param data: Tensorflow tensor that will be subsetted.
:param ind: Indices to take (one for each element along axis 0 of data).
:return: Subsetted tensor.
"""
batch_range = tf.range(tf.shape(data)[0])
indices = tf.stack([batch_range, ind], axis=1)
res = tf.reduce_mean(tf.gather_nd(data, indices), axis=0)
where I pass sequence_length - 1 as indices. In reference to the last topic, I would like to select all relevant outputs followed by average pooling, instead of just the last one.
Now, I tried passing nested lists as indeces to extract_axis_1 but tf.stack does not accept this.
Any solution directions for this?
You can exploit the weight parameter of the tf.contrib.seq2seq.sequence_loss function.
From the documentation:
weights: A Tensor of shape [batch_size, sequence_length] and dtype float. weights constitutes the weighting of each prediction in the sequence. When using weights as masking, set all valid timesteps to 1 and all padded timesteps to 0, e.g. a mask returned by tf.sequence_mask.
You need to compute a binary mask that distinguish between your valid outputs and invalid ones. Then you can just provide this mask to the weights parameter of the loss function (probably, you will want to use a loss like this one); the function will not consider the outputs with a 0 weight in the computation of the loss.
If you can't/don't need to use a sequence loss you can do exactly the same thing manually. You compute a binarymask and then multiply your outputs by this mask and provide these as inputs to your fully connected layer.