This repository contains an autoencoder for multivariate time series forecasting. It features two attention mechanisms described in A Dual-Stage Attention-Based Recurrent Neural Network for Time Series Prediction and was inspired by Seanny123's repository.
To clone the repository please run:
git clone https://github.com/JulesBelveze/time-series-autoencoder.git
Use uv
Then install uv
# install uv curl -LsSf https://astral.sh/uv/install.sh | sh # linux/mac # or brew install uv # mac with homebrew
cd time-series-autoencoder
uv venv
uv pip sync pyproject.tomlInstall directly from requirements.txt
pip install -r requirements.txt
The project uses Hydra as a configuration parser. You can simply change the parameters
directly within your .yaml file or you can override/set parameter using flags (for a complete guide please refer to
the docs).
python3 main.py -cn=[PATH_TO_FOLDER_CONFIG] -cp=[CONFIG_NAME]
Optional arguments:
-h, --help show this help message and exit
--batch-size BATCH_SIZE
batch size
--output-size OUTPUT_SIZE
size of the ouput: default value to 1 for forecasting
--label-col LABEL_COL
name of the target column
--input-att INPUT_ATT
whether or not activate the input attention mechanism
--temporal-att TEMPORAL_ATT
whether or not activate the temporal attention
mechanism
--seq-len SEQ_LEN window length to use for forecasting
--hidden-size-encoder HIDDEN_SIZE_ENCODER
size of the encoder's hidden states
--hidden-size-decoder HIDDEN_SIZE_DECODER
size of the decoder's hidden states
--reg-factor1 REG_FACTOR1
contribution factor of the L1 regularization if using
a sparse autoencoder
--reg-factor2 REG_FACTOR2
contribution factor of the L2 regularization if using
a sparse autoencoder
--reg1 REG1 activate/deactivate L1 regularization
--reg2 REG2 activate/deactivate L2 regularization
--denoising DENOISING
whether or not to use a denoising autoencoder
--do-train DO_TRAIN whether or not to train the model
--do-eval DO_EVAL whether or not evaluating the mode
--data-path DATA_PATH
path to data file
--output-dir OUTPUT_DIR
name of folder to output files
--ckpt CKPT checkpoint path for evaluation
- handles multivariate time series
- attention mechanisms
- denoising autoencoder
- sparse autoencoder
You can find under the examples scripts to train the model in both cases: