LTSF-Linear
[AAAI-23 Oral] Official implementation of the paper "Are Transformers Effective for Time Series Forecasting?"
File Explorer
- __init__.py
- data_factory.py
- data_loader.py
- exp_basic.py
- exp_main.py
- exp_stat.py
- data_factory.py
- data_loader.py
- exp_basic.py
- exp_main.py
- AutoCorrelation.py
- Autoformer_EncDec.py
- Embed.py
- FourierCorrelation.py
- MultiWaveletCorrelation.py
- SelfAttention_Family.py
- Transformer_EncDec.py
- utils.py
- Autoformer.py
- FEDformer.py
- Informer.py
- Transformer.py
- LongForecasting.sh
- LookBackWindow.sh
- masking.py
- metrics.py
- timefeatures.py
- tools.py
- LICENSE
- README.md
- run.py
- AutoCorrelation.py
- Autoformer_EncDec.py
- Embed.py
- SelfAttention_Family.py
- Transformer_EncDec.py
- Autoformer.py
- DLinear.py
- Informer.py
- Linear.py
- NLinear.py
- Stat_models.py
- Transformer.py
- .DS_Store
- DLinear.png
- DLinear_results.png
- DLinear_Univariate_Results.png
- LookBackWindow.png
- results.png
- efficiency.png
- Linear.png
- Mul-results.png
- Uni-results.png
- Visualization_DLinear.png
- lib_hierarchical_mm_float32_cuda.so
- embed.py
- graph_attention.py
- hierarchical_mm_tvm.py
- Layers.py
- Modules.py
- PAM_TVM.py
- Pyraformer_LR.py
- Pyraformer_SS.py
- SubLayers.py
- LongForecasting.sh
- LookBackWindow.sh
- timefeatures.cpython-37.pyc
- tools.cpython-37.pyc
- timefeatures.py
- tools.py
- data_loader.py
- LEGAL.md
- LICENSE
- long_range_main.py
- preprocess_elect.py
- preprocess_flow.py
- preprocess_wind.py
- README.md
- requirements.txt
- simulate_sin.py
- single_step_main.py
- Formers_Embedding.sh
- etth1.sh
- etth2.sh
- ettm1.sh
- ettm2.sh
- .DS_Store
- electricity.sh
- etth1.sh
- etth2.sh
- ettm1.sh
- ettm2.sh
- exchange_rate.sh
- ili.sh
- traffic.sh
- weather.sh
- Formers_Long.sh
- Linear-I.sh
- Stat_Long.sh
- Formers_LookBackWindow.sh
- Linear_DiffWindow.sh
- masking.py
- metrics.py
- timefeatures.py
- tools.py
- .gitignore
- LICENSE
- LTSF-Benchmark.md
- README.md
- requirements.txt
- run_longExp.py
- run_stat.py
- weight_plot.py
# Installation Guide
pythonSetup Steps
conda create -n LTSF_Linear python=3.6.9
Create a conda virtual environment.
conda activate LTSF_Linear
Activate the conda environment.
pip install -r requirements.txt
Install required python packages.
mkdir dataset
Create the dataset directory.
sh scripts/EXP-LongForecasting/Linear/exchange_rate.sh
Start training the model on the exchange rate dataset.
Key Commands
pip install -r requirements.txt
Install python libraries required for the project.
sh scripts/EXP-LongForecasting/Linear/exchange_rate.sh
Run LTSF-Linear model training via the specified shell script.
python weight_plot.py
Run weight visualization for the trained model.
Datasets must be downloaded from Google Drive and placed in the ./dataset directory.
# Use via CDN
jsDelivrjsDelivr serves any public GitHub repository as a CDN with zero setup. Pick a version and a file to get a ready-to-paste link and snippet.
Link
Example
// repository documentation
Was this content helpful?
(0 ratings)
