Autoencoders for sequence data that work out-of-the-box

## Project description

# sequitur

`sequitur`

is a small library of autoencoders for different types of sequence data, ranging from sequences of numbers (e.g. time series) to sequences of images (e.g. videos). It's built on PyTorch and very simple to use –– training and using an autoencoder takes only two lines of code:

```
import torch
from sequitur.models import LINEAR_AE
from sequitur import quick_train
train_seqs = [
torch.tensor([1, 2, 3, 4]),
torch.tensor([5, 6, 7, 8]),
torch.tensor([9, 10, 11, 12])
]
encoder, decoder, _, _ = quick_train(LINEAR_AE, train_seqs, encoding_dim=2, denoise=True)
encoder(torch.tensor([13, 14, 15, 16])) # => torch.tensor([0.19, 0.84])
```

Each autoencoder learns to represent input sequences as lower-dimensional, fixed-size vectors. This can be useful for finding patterns among sequences, clustering sequences, or converting sequences into inputs for other algorithms.

## Installation

Requires Python 3.X and PyTorch 1.2.X

You can install `sequitur`

with `pip`

:

```
$ pip install sequitur
```

## Getting Started

First, you need to prepare a set of example sequences to train an autoencoder on. This training set should be a list of `torch.Tensor`

s, where each tensor has shape `[num_elements, *num_features]`

. So, if each example in your training set is a sequence of 10 5x5 matrices, then each example would be a tensor with shape `[10, 5, 5]`

.

Next, you need to choose an autoencoder model. If you're working with sequences of numbers (e.g. time series) or 1D vectors (e.g. word vectors), then you should use the `LINEAR_AE`

or `LSTM_AE`

model. For sequences of 2D matrices (e.g. videos) or 3D matrices (e.g. fMRI scans), you'll want to use `CONV_LSTM_AE`

. Each model is a PyTorch module, and can be imported like so:

```
from sequitur.models import CONV_LSTM_AE
```

From here, you can either initialize the model yourself and write your own training loop, or import the `quick_train`

function and plug in the model, training set, and desired encoding size, like so:

```
import torch
from sequitur.models import CONV_LSTM_AE
from sequitur import quick_train
train_set = [torch.randn(10, 5, 5) for _ in range(100)]
encoder, decoder, _, _ = quick_train(CONV_LSTM_AE, train_set, encoding_dim=4)
```

After training, `quick_train`

returns the `encoder`

and `decoder`

models, which are PyTorch modules that can encode and decode new sequences. These can be used like so:

```
x = torch.randn(10, 5, 5)
z = encoder(x) # Tensor with shape [4]
x_prime = decoder(z) # Tensor with shape [10, 5, 5]
```

## API

#### quick_train

**`quick_train(model, train_set, encoding_dim, verbose=False, lr=1e-3, epochs=50, denoise=False, **kwargs)`

**

Lets you train an autoencoder with just one line of code. Useful if you don't want to create your own training loop. Training involves learning a vector encoding of each input sequence, reconstructing the original sequence from the encoding, and calculating the loss (mean-squared error) between the reconstructed input and the original input. The autoencoder weights are updated using the Adam optimizer.

If `denoise=True`

, then each input sequence is injected with Gaussian noise before being fed into the autoencoder. The autoencoder is then trained to reconstruct the original undistorted input.

**Parameters:**

`model`

*(torch.nn.Module)*: Autoencoder model to train (imported from`sequitur.models`

)`train_set`

*(list)*: List of sequences (each a`torch.Tensor`

) to train the model on; has shape`[num_examples, seq_len, *num_features]`

`encoding_dim`

*(int)*: Desired size of the vector encoding`verbose`

*(bool, optional (default=False))*: Whether or not to print the loss at each epoch`lr`

*(float, optional (default=1e-3))*: Learning rate`epochs`

*(int, optional (default=50))*: Number of epochs to train for`denoise`

*(bool, optional=(default=False))*: If`True`

, converts autoencoder into a Denoising Autoencoder (DAE)`**kwargs`

: Parameters to pass into`model`

when it's instantiated

**Returns:**

`encoder`

*(torch.nn.Module)*: Trained encoder model; takes a sequence (as a tensor) as input and returns an encoding of the sequence as a tensor of shape`[encoding_dim]`

`decoder`

*(torch.nn.Module)*: Trained decoder model; takes an encoding (as a tensor) and returns a decoded sequence`encodings`

*(list)*: List of tensors corresponding to the final vector encodings of each sequence in the training set`losses`

*(list)*: List of average MSE values at each epoch

### Models

Every autoencoder inherits from `torch.nn.Module`

and has an `encoder`

attribute and a `decoder`

attribute, both of which also inherit from `torch.nn.Module`

.

#### LINEAR_AE

`LINEAR_AE(input_dim, encoding_dim, h_dims=[], h_activ=torch.nn.Sigmoid(), out_activ=torch.nn.Tanh())`

Consists of fully-connected layers stacked on top of each other. Can only be used if you're dealing with sequences of numbers, not vectors or matrices.

**Parameters:**

`input_dim`

*(int)*: Size of each input sequence`encoding_dim`

*(int)*: Size of the vector encoding`h_dims`

*(list, optional (default=[]))*: List of hidden layer sizes for the encoder`h_activ`

*(torch.nn.Module or None, optional (default=torch.nn.Sigmoid()))*: Activation function to use for hidden layers; if`None`

, no activation function is used`out_activ`

*(torch.nn.Module or None, optional (default=torch.nn.Tanh()))*: Activation function to use for the output layer in the encoder; if`None`

, no activation function is used

**Example:**

To create the autoencoder shown in the diagram above, use the following arguments:

```
from sequitur.models import LINEAR_AE
model = LINEAR_AE(
input_dim=10,
encoding_dim=4,
h_dims=[8, 6],
h_activ=None,
out_activ=None
)
```

#### LSTM_AE

`LSTM_AE(input_dim, encoding_dim, h_dims=[], h_activ=torch.nn.Sigmoid(), out_activ=torch.nn.Tanh())`

Autoencoder for sequences of 1D vectors which consists of stacked LSTMs. Can be trained on sequences of varying length.

**Parameters:**

`input_dim`

*(int)*: Size of each sequence element (vector)`encoding_dim`

*(int)*: Size of the vector encoding`h_dims`

*(list, optional (default=[]))*: List of hidden layer sizes for the encoder`h_activ`

*(torch.nn.Module or None, optional (default=torch.nn.Sigmoid()))*: Activation function to use for hidden layers; if`None`

, no activation function is used`out_activ`

*(torch.nn.Module or None, optional (default=torch.nn.Tanh()))*: Activation function to use for the output layer in the encoder; if`None`

, no activation function is used

**Example:**

To create the autoencoder shown in the diagram above, use the following arguments:

```
from sequitur.models import LSTM_AE
model = LSTM_AE(
input_dim=3,
encoding_dim=7,
h_dims=[64],
h_activ=None,
out_activ=None
)
```

Note that the `model.decoder`

module requires an argument in addition to a vector encoding, `seq_len`

, to decode a sequence. For example:

```
import torch
x = torch.randn(10, 3)
z = model.encoder(x)
x_prime = model.decoder(z, seq_len=10)
```

#### CONV_LSTM_AE

`CONV_LSTM_AE(input_dims, encoding_dim, in_channels, h_conv_channels=[], h_lstm_channels=[], kernel=None, stride=None)`

Autoencoder for sequences of 2D or 3D matrices/images, loosely based on the CNN-LSTM architecture described in *Beyond Short Snippets: Deep Networks for Video Classification.* Uses a CNN to create vector encodings of each image in an input sequence, and then an LSTM to create encodings of the sequence of vectors.

**Parameters:**

`input_dims`

*(list)*: Shape of each 2D or 3D image in the input sequences`encoding_dim`

*(int)*: Size of the vector encoding`in_channels`

*(int)*: Number of channels in each image`kernel`

*(tuple or None, optional (default=None))*: TODO`stride`

*(tuple or None, optional (default=None))*: TODO`h_conv_channels`

*(list, optional (default=[]))*: List of hidden channel sizes for the convolutional layers`h_lstm_channels`

*(list, optional (default=[]))*: List of hidden channel sizes for the LSTM layers

**Example:**

```
from sequitur.models import CONV_LSTM_AE
model = CONV_LSTM_AE(
input_dims=[5, 5],
encoding_dim=16,
in_channels=1,
kernel=(2, 2),
stride=(1, 1),
)
```

## Project details

## Release history Release notifications | RSS feed

## Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.