I am trying to encode a sequence of image embeddings to one bigger embedding using MultiHeadAttention. (order doesn't matter)
I want an operation similar to passing a sequence of shape (batch_size, sequence_length, embedding_dim) to an LSTM layer and taking the last hidden_state as an embedding holding all importatnt information about the sequence.. or we may say a sequence embedding.
I want to implement that using attention to get rid of the recurrent behavior..
# embeddings: shape(batch_size, sequence_length, embedding_dim)
multihead_attn = nn.MultiheadAttention(embedding_dim, num_heads)
attn_output, attn_output_weights = multihead_attn(embeddings, embeddings, embeddings)
but in this case the attention output will have a shape of (batch_size, sequence_length, embedding_dim) as well..
should I be doing
attn_output = attn_output.mean(1)
or what if I pass the query as the embeddings.mean(1).. will that give the intended behavior? will the output simulate a sequence embedding?
# embeddings: shape(batch_size, sequence_length, embedding_dim)
multihead_attn = nn.MultiheadAttention(embedding_dim, num_heads)
seq_emb, attn_output_weights = multihead_attn(embeddings.mean(1), embeddings, embeddings)
Related
I have implemented Autoencoder using Keras that takes 112*112*3 neurons as input and 100 neurons as the compressed/encoded state. I want to find the neurons out of these 100 that learns the important features. So far i have calculated eigen values(e) and eigen vectors(v) using the following steps. And i found out that around first 30 values of (e) is greater than 0. Does that mean the first 30 modes are the important ones? Is there any other method that could find the important neurons?
Thanks in Advance
x_enc = enc_model.predict(x_train, batch_size=BATCH_SIZE) # shape (3156,100)
x_mean = np.mean(x_enc, axis=0) # shape (100,)
x_stds = np.std(x_enc, axis=0) # shape (100,)
x_cov = np.cov((x_enc - x_mean).T) # shape (100,100)
e, v = np.linalg.eig(x_cov) # shape (100,) and (100,100) respectively
I don't know if the approach you are using will actually give you any useful results since the way the network learns and what it exactly learns aren't known, I suggest you use a different kind of autoencoder, that automatically learns disentangled representations of the data in a latent space, this way you can be sure that all the parameters you find are actually contributing to the representation of your data. check this article
I have a similar open question here on Cross Validated (though not implementation focused, which I intend this question to be, so I think they are both valid).
I'm working on a project that uses sensors to monitor a persons GPS location. The coordinates will then be converted to a simple-grid representation. What I want to try and do is after recording a users routes, train a neural network to predict the next coordinates, i.e. take the example below where a user repeats only two routes over time, Home->A and Home->B.
I want to train an RNN/LSTM with sequences of varying lengths e.g. (14,3), (13,3), (12,3), (11,3), (10,3), (9,3), (8,3), (7,3), (6,3), (5,3), (4,3), (3,3), (2,3), (1,3) and then also predict with sequences of varying lengths e.g. for this example route if I called
route = [(14,3), (13,3), (12,3), (11,3), (10,3)] //pseudocode
pred = model.predict(route)
pred should give me (9,3) (or ideally even a longer prediction e.g. ((9,3), (8,3), (7,3), (6,3), (5,3), (4,3), (3,3), (2,3), (1,3))
How do I feed such training sequences to the init and forward operations identified below?
self.rnn = nn.RNN(input_size, hidden_dim, n_layers, batch_first=True)
out, hidden = self.rnn(x, hidden)
Also, should the entire route be a tensor or each set of coordinates within the route a tensor?
I'm not very experienced with RNNs, but I'll give it a try.
A few things to pay attention to before we start:
1. Your data is not normalized.
2. The output prediction you want (even after normalization) is not bounded to [-1, 1] range and therefore you cannot have tanh or ReLU activations acting on the output predictions.
To address your problem, I propose a recurrent net that given a current state (2D coordinate) predicts the next state (2D coordinates). Note that since this is a recurrent net, there is also a hidden state associated with each location. At first, the hidden state is zero, but as the net sees more steps, it updates its hidden state.
I propose a simple net to address your problem. It has a single RNN layer with 8 hidden states, and a fully connected layer on to to output the prediction.
class MyRnn(nn.Module):
def __init__(self, in_d=2, out_d=2, hidden_d=8, num_hidden=1):
super(MyRnn, self).__init__()
self.rnn = nn.RNN(input_size=in_d, hidden_size=hidden_d, num_layers=num_hidden)
self.fc = nn.Linear(hidden_d, out_d)
def forward(self, x, h0):
r, h = self.rnn(x, h0)
y = self.fc(r) # no activation on the output
return y, h
You can use your two sequences as training data, each sequence is a tensor of shape Tx1x2 where T is the sequence length, and each entry is two dimensional (x-y).
To predict (during training):
rnn = MyRnn()
pred, out_h = rnn(seq[:-1, ...], torch.zeros(1, 1, 8)) # given time t predict t+1
err = criterion(pred, seq[1:, ...]) # compare prediction to t+1
Once the model is trained, you can show it first k steps and continue to predict the next steps:
rnn.eval()
with torch.no_grad():
pred, h = rnn(s[:k,...], torch.zeros(1, 1, 8, dtype=torch.float))
# pred[-1, ...] is the predicted next step
prev = pred[-1:, ...]
for j in range(k+1, s.shape[0]):
pred, h = rnn(prev, h) # note how we keep track of the hidden state of the model. it is no longer init to zero.
prev = pred
I put everything together in a colab notebook so you can play with it.
For simplicity, I ignored the data normalization here, but you can find it in the colab notebook.
What's next?
These types of predictions are prone to error accumulation. This should be addressed during training, by shifting the inputs from the ground truth "clean" sequences to the actual predicted sequences, so the model will be able to compensate for its errors.
The dimensions for the input data for LSTM are [Batch Size, Sequence Length, Input Dimension] in tensorflow.
What is the meaning of Sequence Length & Input Dimension ?
How do we assign the values to them if my input data is of the form :
[[[1.23] [2.24] [5.68] [9.54] [6.90] [7.74] [3.26]]] ?
LSTMs are a subclass of recurrent neural networks. Recurrent neural nets are by definition applied on sequential data, which without loss of generality means data samples that change over a time axis. A full history of a data sample is then described by the sample values over a finite time window, i.e. if your data live in an N-dimensional space and evolve over t-time steps, your input representation must be of shape (num_samples, t, N).
Your data does not fit the above description. I assume, however, that this representation means you have a scalar value x which evolves over 7 time instances, such that x[0] = 1.23, x[1] = 2.24, etc.
If that is the case, you need to reshape your input such that instead of a list of 7 elements, you have an array of shape (7,1). Then, your full data can be described by a 3rd order tensor of shape (num_samples, 7, 1) which can be accepted by a LSTM.
Simply put seq_len is number of time steps that will be inputted into LSTM network, Let's understand this by example...
Suppose you are doing a sentiment classification using LSTM.
Your input sentence to the network is =["I hate to eat apples"]. Every single token would be fed as input at each timestep, So accordingly here the seq_Len would total number of tokens in a sentence that is 5.
Coming to the input_dim you might know we can't directly feed words to the netowrk you would need to encode those words into numbers. In Pytorch/tensorflow embedding layers are used where we have to specify embedding dimension.
Suppose your embedding dimension is 50 that means that embedding layer will take index of respective token and convert it into vector representation of size 50. So the input dim to LSTM network would become 50.
Its commonplace for various neural network architectures in NLP and vision-language problems to tie the weights of an initial word embedding layer to that of an output softmax. Usually this produces a boost to sentence generation quality. (see example here)
In Keras its typical to embed word embedding layers using the Embedding class, however there seems to be no easy way to tie the weights of this layer to the output softmax. Would anyone happen to know how this could be implemented ?
Be aware that Press and Wolf dont't propose to freeze the weights to some pretrained ones, but tie them. That means, to ensure that input and output weights are always the same during training (in the sense of synchronized).
In a typical NLP model (e.g. language modelling/translation), you have an input dimension (vocabulary) of size V and a hidden representation size H. Then, you start with an Embedding layer, which is a matrix VxH. And the output layer is (probably) something like Dense(V, activation='softmax'), which is a matrix H2xV. When tying the weights, we want that those matrices are the same (therefore, H==H2).
For doing this in Keras, I think the way to go is via shared layers:
In your model, you need to instantiate a shared embedding layer (of dimension VxH), and apply it to either your input and output. But you need to transpose it, to have the desired output dimensions (HxV). So, we declare a TiedEmbeddingsTransposed layer, which transposes the embedding matrix from a given layer (and applies an activation function):
class TiedEmbeddingsTransposed(Layer):
"""Layer for tying embeddings in an output layer.
A regular embedding layer has the shape: V x H (V: size of the vocabulary. H: size of the projected space).
In this layer, we'll go: H x V.
With the same weights than the regular embedding.
In addition, it may have an activation.
# References
- [ Using the Output Embedding to Improve Language Models](https://arxiv.org/abs/1608.05859)
"""
def __init__(self, tied_to=None,
activation=None,
**kwargs):
super(TiedEmbeddingsTransposed, self).__init__(**kwargs)
self.tied_to = tied_to
self.activation = activations.get(activation)
def build(self, input_shape):
self.transposed_weights = K.transpose(self.tied_to.weights[0])
self.built = True
def compute_mask(self, inputs, mask=None):
return mask
def compute_output_shape(self, input_shape):
return input_shape[0], K.int_shape(self.tied_to.weights[0])[0]
def call(self, inputs, mask=None):
output = K.dot(inputs, self.transposed_weights)
if self.activation is not None:
output = self.activation(output)
return output
def get_config(self):
config = {'activation': activations.serialize(self.activation)
}
base_config = super(TiedEmbeddingsTransposed, self).get_config()
return dict(list(base_config.items()) + list(config.items()))
The usage of this layer is:
# Declare the shared embedding layer
shared_embedding_layer = Embedding(V, H)
# Obtain word embeddings
word_embedding = shared_embedding_layer(input)
# Do stuff with your model
# Compute output (e.g. a vocabulary-size probability vector) with the shared layer:
output = TimeDistributed(TiedEmbeddingsTransposed(tied_to=shared_embedding_layer, activation='softmax')(intermediate_rep)
I have tested this in NMT-Keras and it trains properly. But, as I try to load a trained model, it gets an error, related to the way Keras loads the models: it doesn't load the weights from the tied_to. I've found several questions regarding this (1, 2, 3), but I haven't managed to solve this issue. If someone have any ideas on the next steps to take, I'd be very glad to hear them :)
As you may read here you should simply set trainable flag to False. E.g.
aux_output = Embedding(..., trainable=False)(input)
....
output = Dense(nb_of_classes, .. ,activation='softmax', trainable=False)
I would like to code with Keras a neural network that acts both as an autoencoder AND a classifier for semi-supervised learning. Take for example this dataset where there is a few labeled images and a lot of unlabeled images: https://cs.stanford.edu/~acoates/stl10/
Some papers listed here achieved that, or very similar things, successfully.
To sum up: if the model would have the same input data shape and the same "encoding" convolutional layers, but would split into two heads (fork-style), so there is a classification head and a decoding head, in a way that the unsupervised autoencoder will contribute to a good learning for the classification head.
With TensorFlow there would be no problem doing that as we have full control over the computational graph.
But with Keras, things are more high-level and I feel that all the calls to ".fit" must always provide all the data at once (so it would force me to tie together the classification head and the autoencoding head into one time-step).
One way in keras to almost do that would be with something that goes like this:
input = Input(shape=(32, 32, 3))
cnn_feature_map = sequential_cnn_trunk(input)
classification_predictions = Dense(10, activation='sigmoid')(cnn_feature_map)
autoencoded_predictions = decode_cnn_head_sequential(cnn_feature_map)
model = Model(inputs=[input], outputs=[classification_predictions, ])
model.compile(optimizer='rmsprop',
loss='binary_crossentropy',
metrics=['accuracy'])
model.fit([images], [labels, images], epochs=10)
However, I think and I fear that if I just want to fit things in that way it will fail and ask for the missing head:
for epoch in range(10):
# classifications step
model.fit([images], [labels, None], epochs=1)
# "semi-unsupervised" autoencoding step
model.fit([images], [None, images], epochs=1)
# note: ".train_on_batch" could probably be used rather than ".fit" to avoid doing a whole epoch each time.
How should one implement that behavior with Keras? And could the training be done jointly without having to split the two calls to the ".fit" function?
Sometimes when you don't have a label you can pass zero vector instead of one hot encoded vector. It should not change your result because zero vector doesn't have any error signal with categorical cross entropy loss.
My custom to_categorical function looks like this:
def tricky_to_categorical(y, translator_dict):
encoded = np.zeros((y.shape[0], len(translator_dict)))
for i in range(y.shape[0]):
if y[i] in translator_dict:
encoded[i][translator_dict[y[i]]] = 1
return encoded
When y contains labels, and translator_dict is a python dictionary witch contains labels and its unique keys like this:
{'unisex':2, 'female': 1, 'male': 0}
If an UNK label can't be found in this dictinary then its encoded label will be a zero vector
If you use this trick you also have to modify your accuracy function to see real accuracy numbers. you have to filter out all zero vectors from our metrics
def tricky_accuracy(y_true, y_pred):
mask = K.not_equal(K.sum(y_true, axis=-1), K.constant(0)) # zero vector mask
y_true = tf.boolean_mask(y_true, mask)
y_pred = tf.boolean_mask(y_pred, mask)
return K.cast(K.equal(K.argmax(y_true, axis=-1), K.argmax(y_pred, axis=-1)), K.floatx())
note: You have to use larger batches (e.g. 32) in order to prevent zero matrix update, because It can make your accuracy metrics crazy, I don't know why
Alternative solution
Use Pseudo Labeling :)
you can train jointly, you have to pass an array insted of single label.
I used fit_generator, e.g.
model.fit_generator(
batch_generator(),
steps_per_epoch=len(dataset) / batch_size,
epochs=epochs)
def batch_generator():
batch_x = np.empty((batch_size, img_height, img_width, 3))
gender_label_batch = np.empty((batch_size, len(gender_dict)))
category_label_batch = np.empty((batch_size, len(category_dict)))
while True:
i = 0
for idx in np.random.choice(len(dataset), batch_size):
image_id = dataset[idx][0]
batch_x[i] = load_and_convert_image(image_id)
gender_label_batch[i] = gender_labels[idx]
category_label_batch[i] = category_labels[idx]
i += 1
yield batch_x, [gender_label_batch, category_label_batch]