├── .gitignore ├── Dockerfile ├── LICENSE ├── README.md ├── characteristics_extractor ├── Characteristics_Extractor.py ├── DemoDatasets │ ├── Exchange.csv │ └── m4_hourly_dataset_69.csv ├── Readme_chn.md └── Readme_en.md ├── config ├── fixed_forecast_config_daily.json ├── fixed_forecast_config_hourly.json ├── fixed_forecast_config_monthly.json ├── fixed_forecast_config_other.json ├── fixed_forecast_config_quarterly.json ├── fixed_forecast_config_weekly.json ├── fixed_forecast_config_yearly.json └── rolling_forecast_config.json ├── docs ├── figures │ ├── Comparison_with_Related_Libraries.png │ ├── Drop-last.png │ ├── MultiGPUs.png │ ├── Non-Gaussianity.png │ ├── Pipeline.png │ ├── QR.png │ ├── TFB-LOGO.png │ ├── correlation.png │ ├── seasonal.png │ ├── shifting.png │ ├── stationarity.png │ ├── transition.png │ └── trend.png └── tutorials │ ├── converted-Exchange.csv │ ├── introduction_and_pseudocode_for_time_series_characteristics.md │ ├── ori-Exchange.csv │ ├── other_usage.sh │ ├── steps_to_develop_your_own_method.md │ ├── steps_to_evaluate_your_own_time_series.md │ ├── steps_to_get_predict_and_actual_data.md │ ├── steps_to_predict_only_a_subset_of_input_variables.md │ ├── steps_to_predict_only_a_subset_of_input_variables_chn.pdf │ └── steps_to_train_models_with_multi_gpus_using_dp.md ├── requirements-docker.txt ├── requirements.txt ├── scripts ├── __init__.py ├── multivariate_forecast │ ├── AQShunyi_script │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── AQWan_script │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── Covid-19_script │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── CzeLan_script │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── ETTh1_script │ │ ├── Amplifier.sh │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── HDMixer │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeKAN.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ ├── iTransformer.sh │ │ └── xPatch.sh │ ├── ETTh2_script │ │ ├── Amplifier.sh │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeKAN.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── ETTm1_script │ │ ├── Amplifier.sh │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeKAN.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── ETTm2_script │ │ ├── Amplifier.sh │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeKAN.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── Electricity_script │ │ ├── Amplifier.sh │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeKAN.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── Exchange_script │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── FRED-MD_script │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── ILI_script │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── METR-LA_script │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── NASDAQ_script │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── NN5_script │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── NYSE_script │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── PEMS-BAY_script │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── PEMS04_script │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── PEMS08_script │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── Solar_script │ │ ├── Amplifier.sh │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeKAN.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── Traffic_script │ │ ├── Amplifier.sh │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeKAN.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── Weather_script │ │ ├── Amplifier.sh │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchMLP.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeKAN.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── Wike2000_script │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ ├── Wind_script │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh │ └── ZafNoo_script │ │ ├── Crossformer.sh │ │ ├── DLinear.sh │ │ ├── DUET.sh │ │ ├── FEDformer.sh │ │ ├── FITS.sh │ │ ├── FiLM.sh │ │ ├── Informer.sh │ │ ├── MICN.sh │ │ ├── NLinear.sh │ │ ├── Nonstationary_Transformer.sh │ │ ├── PDF.sh │ │ ├── PatchTST.sh │ │ ├── Pathformer.sh │ │ ├── RNNModel.sh │ │ ├── RegressionModel.sh │ │ ├── TCNModel.sh │ │ ├── TimeMixer.sh │ │ ├── TimesNet.sh │ │ ├── Triformer.sh │ │ ├── VAR_model.sh │ │ └── iTransformer.sh ├── run_benchmark.py └── univariate_forecast │ └── univariate-forecast.sh └── ts_benchmark ├── __init__.py ├── baselines ├── __init__.py ├── amplifier │ ├── __init__.py │ ├── amplifier.py │ ├── layers │ │ ├── RevIN.py │ │ └── __init__.py │ ├── models │ │ ├── DLinear.py │ │ ├── __init__.py │ │ └── amplifier_model.py │ └── utils │ │ └── tools.py ├── darts │ ├── __init__.py │ └── darts_models.py ├── duet │ ├── __init__.py │ ├── duet.py │ ├── layers │ │ ├── Autoformer_EncDec.py │ │ ├── Embed.py │ │ ├── RevIN.py │ │ ├── SelfAttention_Family.py │ │ ├── Transformer_EncDec.py │ │ ├── __init__.py │ │ ├── distributional_router_encoder.py │ │ ├── linear_extractor_cluster.py │ │ └── linear_pattern_extractor.py │ ├── models │ │ ├── __init__.py │ │ └── duet_model.py │ └── utils │ │ ├── __init__.py │ │ ├── losses.py │ │ ├── masked_attention.py │ │ ├── masking.py │ │ ├── timefeatures.py │ │ └── tools.py ├── fits │ ├── __init__.py │ ├── fits.py │ └── fits_model.py ├── hdmixer │ ├── __init__.py │ ├── hdmixer.py │ ├── layers │ │ ├── PatchTST_layers.py │ │ ├── RevIN.py │ │ └── box_coder1D.py │ ├── models │ │ └── HDMixer.py │ └── utils │ │ └── tools.py ├── merlion │ ├── __init__.py │ └── merlion_models.py ├── patchmlp │ ├── __init__.py │ ├── layers │ │ ├── Embed.py │ │ └── __init__.py │ ├── models │ │ ├── __init__.py │ │ └── patchmlp_model.py │ ├── patchmlp.py │ └── utils │ │ ├── __init__.py │ │ ├── masking.py │ │ ├── metrics.py │ │ ├── timefeatures.py │ │ └── tools.py ├── pathformer │ ├── __init__.py │ ├── layers │ │ ├── AMS.py │ │ ├── Embedding.py │ │ ├── Layer.py │ │ ├── RevIN.py │ │ └── __init__.py │ ├── models │ │ ├── __init__.py │ │ └── pathformer_model.py │ ├── pathformer.py │ └── utils │ │ ├── Other.py │ │ ├── __init__.py │ │ ├── decomposition.py │ │ ├── masking.py │ │ ├── metrics.py │ │ ├── timefeatures.py │ │ └── tools.py ├── pdf │ ├── PDF.py │ ├── __init__.py │ ├── layers │ │ ├── PDF_backbone.py │ │ ├── PDF_layers.py │ │ ├── RevIN.py │ │ └── __init__.py │ ├── models │ │ ├── PDF.py │ │ └── __init__.py │ └── utils │ │ ├── Other.py │ │ ├── __init__.py │ │ ├── decomposition.py │ │ ├── masking.py │ │ ├── metrics.py │ │ ├── timefeatures.py │ │ └── tools.py ├── self_impl │ ├── VAR │ │ └── VAR.py │ └── __init__.py ├── tag_csv │ ├── darts_tag.csv │ ├── other_methods_tag.csv │ └── tslib_tag.csv ├── time_series_library │ ├── __init__.py │ ├── adapters_for_transformers.py │ ├── layers │ │ ├── AutoCorrelation.py │ │ ├── Autoformer_EncDec.py │ │ ├── Conv_Blocks.py │ │ ├── Crossformer_EncDec.py │ │ ├── ETSformer_EncDec.py │ │ ├── Embed.py │ │ ├── FourierCorrelation.py │ │ ├── MultiWaveletCorrelation.py │ │ ├── Pyraformer_EncDec.py │ │ ├── SelfAttention_Family.py │ │ ├── StandardNorm.py │ │ ├── Transformer_EncDec.py │ │ └── __init__.py │ ├── models │ │ ├── Autoformer.py │ │ ├── Crossformer.py │ │ ├── DLinear.py │ │ ├── ETSformer.py │ │ ├── FEDformer.py │ │ ├── FiLM.py │ │ ├── FreTS.py │ │ ├── Informer.py │ │ ├── Koopa.py │ │ ├── LightTS.py │ │ ├── MICN.py │ │ ├── Nonstationary_Transformer.py │ │ ├── PatchTST.py │ │ ├── Pyraformer.py │ │ ├── Reformer.py │ │ ├── SegRNN.py │ │ ├── TSMixer.py │ │ ├── TemporalFusionTransformer.py │ │ ├── TiDE.py │ │ ├── TimeMixer.py │ │ ├── TimesNet.py │ │ ├── Transformer.py │ │ └── iTransformer.py │ ├── patchs │ │ ├── Linear.py │ │ ├── NLinear.py │ │ └── Triformer.py │ └── utils │ │ ├── __init__.py │ │ ├── losses.py │ │ ├── masking.py │ │ ├── timefeatures.py │ │ └── tools.py ├── timekan │ ├── __init__.py │ ├── layers │ │ ├── Autoformer_EncDec.py │ │ ├── ChebyKANLayer.py │ │ ├── Embed.py │ │ ├── StandardNorm.py │ │ └── __init__.py │ ├── models │ │ └── timekan_model.py │ ├── timekan.py │ └── utils │ │ ├── __init__.py │ │ ├── masking.py │ │ ├── metrics.py │ │ ├── timefeatures.py │ │ └── tools.py ├── utils.py └── xpatch │ ├── __init__.py │ ├── layers │ ├── decomp.py │ ├── dema.py │ ├── ema.py │ ├── network.py │ ├── network_cnn.py │ ├── network_mlp.py │ └── revin.py │ ├── models │ └── xPatch.py │ ├── utils │ ├── metrics.py │ ├── timefeatures.py │ └── tools.py │ └── xpatch.py ├── common ├── __init__.py └── constant.py ├── data ├── __init__.py ├── data_pool.py ├── data_pool_impl_base.py ├── data_server_base.py ├── data_source.py ├── dataset.py ├── suites │ ├── __init__.py │ └── global_storage.py └── utils.py ├── evaluation ├── __init__.py ├── evaluate_model.py ├── evaluator.py ├── metrics │ ├── __init__.py │ ├── regression_metrics.py │ └── utils.py └── strategy │ ├── __init__.py │ ├── constants.py │ ├── fixed_forecast.py │ ├── forecasting.py │ ├── rolling_forecast.py │ └── strategy.py ├── models ├── __init__.py ├── model_base.py └── model_loader.py ├── pipeline.py ├── recording.py ├── report ├── __init__.py ├── report_csv.py ├── report_dash │ ├── __init__.py │ ├── app.py │ ├── layout.py │ ├── memory.py │ └── pages │ │ ├── leaderboard.py │ │ └── query.py └── utils │ ├── __init__.py │ └── leaderboard.py └── utils ├── __init__.py ├── compress.py ├── data_processing.py ├── design_pattern.py ├── get_file_name.py ├── parallel ├── __init__.py ├── base.py ├── ray_backend.py └── sequential_backend.py └── random_utils.py /.gitignore: -------------------------------------------------------------------------------- 1 | dataset/ 2 | result/ 3 | .idea/ 4 | .vscode/ 5 | .git 6 | *.DS_Store 7 | venv/ -------------------------------------------------------------------------------- /Dockerfile: -------------------------------------------------------------------------------- 1 | FROM ubuntu:20.04 2 | 3 | # setup packages 4 | RUN apt-get update -y 5 | RUN apt-get install -y python3 python-is-python3 python3-pip python3-venv 6 | 7 | # copy local files 8 | RUN mkdir /home/TFB 9 | COPY . /home/TFB 10 | 11 | # create virtualenv, activate and upgrade the env-oriented pip 12 | RUN python -m venv /env 13 | ENV PATH="/env/bin:$PATH" 14 | RUN pip install --upgrade pip 15 | 16 | # install dependencies 17 | RUN pip install -r /home/TFB/requirements-docker.txt -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- 1 | MIT License 2 | 3 | Copyright (c) 2024 Huawei Technologies Co., Ltd 4 | 5 | Permission is hereby granted, free of charge, to any person obtaining a copy 6 | of this software and associated documentation files (the "Software"), to deal 7 | in the Software without restriction, including without limitation the rights 8 | to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 9 | copies of the Software, and to permit persons to whom the Software is 10 | furnished to do so, subject to the following conditions: 11 | 12 | The above copyright notice and this permission notice shall be included in all 13 | copies or substantial portions of the Software. 14 | 15 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 16 | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 17 | FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 18 | AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 19 | LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, 20 | OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE 21 | SOFTWARE. 22 | -------------------------------------------------------------------------------- /config/fixed_forecast_config_daily.json: -------------------------------------------------------------------------------- 1 | { 2 | "data_config" : { 3 | "feature_dict": { 4 | "if_univariate": true, 5 | "if_trend": null, 6 | "has_timestamp": null, 7 | "if_season": null, 8 | "freq": "daily" 9 | }, 10 | "data_set_name": "large_forecast" 11 | }, 12 | 13 | "model_config": { 14 | "models": [ 15 | ], 16 | "recommend_model_hyper_params": { 17 | "input_chunk_length":17, 18 | "output_chunk_length": 14, 19 | "norm": false 20 | 21 | } 22 | }, 23 | 24 | "evaluation_config" : { 25 | "metrics": ["mae", "mse", "rmse", "mape", "smape", {"name": "mase", "seasonality": 7}, "wape", "msmape"], 26 | "strategy_args": { 27 | "strategy_name": "fixed_forecast", 28 | "horizon": 14, 29 | "train_ratio_in_tv": 1, 30 | "seed": 2021, 31 | "deterministic": "efficient", 32 | "save_true_pred": false, 33 | "target_channel": null 34 | } 35 | }, 36 | 37 | 38 | "report_config" : { 39 | "aggregate_type": "mean", 40 | "report_metrics": ["mae", "mse", "rmse", "mape", "smape", "mase", "wape", "msmape"], 41 | "fill_type": "mean_value", 42 | "null_value_threshold": "0.3" 43 | } 44 | } 45 | -------------------------------------------------------------------------------- /config/fixed_forecast_config_hourly.json: -------------------------------------------------------------------------------- 1 | { 2 | "data_config" : { 3 | "feature_dict": { 4 | "if_univariate": true, 5 | "if_trend": null, 6 | "has_timestamp": null, 7 | "if_season": null, 8 | "freq": "hourly" 9 | }, 10 | "data_set_name": "large_forecast" 11 | }, 12 | 13 | "model_config": { 14 | "models": [ 15 | ], 16 | "recommend_model_hyper_params": { 17 | "input_chunk_length":60, 18 | "output_chunk_length": 48, 19 | "norm": false 20 | } 21 | }, 22 | 23 | "evaluation_config" : { 24 | "metrics": ["mae", "mse", "rmse", "mape", "smape", {"name": "mase", "seasonality": 24}, "wape", "msmape"], 25 | "strategy_args": { 26 | "strategy_name": "fixed_forecast", 27 | "horizon": 48, 28 | "train_ratio_in_tv": 1, 29 | "seed": 2021, 30 | "deterministic": "efficient", 31 | "save_true_pred": false, 32 | "target_channel": null 33 | } 34 | }, 35 | 36 | 37 | "report_config" : { 38 | "aggregate_type": "mean", 39 | "report_metrics": ["mae", "mse", "rmse", "mape", "smape", "mase", "wape", "msmape"], 40 | "fill_type": "mean_value", 41 | "null_value_threshold": "0.3" 42 | } 43 | } 44 | -------------------------------------------------------------------------------- /config/fixed_forecast_config_monthly.json: -------------------------------------------------------------------------------- 1 | { 2 | "data_config" : { 3 | "feature_dict": { 4 | "if_univariate": true, 5 | "if_trend": null, 6 | "has_timestamp": null, 7 | "if_season": null, 8 | "freq": "monthly" 9 | }, 10 | "data_set_name": "large_forecast" 11 | }, 12 | 13 | "model_config": { 14 | "models": [ 15 | ], 16 | "recommend_model_hyper_params": { 17 | "input_chunk_length":22, 18 | "output_chunk_length": 18, 19 | "norm": false 20 | 21 | } 22 | }, 23 | 24 | "evaluation_config" : { 25 | "metrics": ["mae", "mse", "rmse", "mape", "smape", {"name": "mase", "seasonality": 12}, "wape", "msmape"], 26 | "strategy_args": { 27 | "strategy_name": "fixed_forecast", 28 | "horizon": 18, 29 | "train_ratio_in_tv": 1, 30 | "seed": 2021, 31 | "deterministic": "efficient", 32 | "save_true_pred": false, 33 | "target_channel": null 34 | } 35 | }, 36 | 37 | 38 | "report_config" : { 39 | "aggregate_type": "mean", 40 | "report_metrics": ["mae", "mse", "rmse", "mape", "smape", "mase", "wape", "msmape"], 41 | "fill_type": "mean_value", 42 | "null_value_threshold": "0.3" 43 | } 44 | } 45 | -------------------------------------------------------------------------------- /config/fixed_forecast_config_other.json: -------------------------------------------------------------------------------- 1 | { 2 | "data_config" : { 3 | "feature_dict": { 4 | "if_univariate": true, 5 | "if_trend": null, 6 | "has_timestamp": null, 7 | "if_season": null, 8 | "freq": "other" 9 | }, 10 | "data_set_name": "large_forecast" 11 | }, 12 | 13 | "model_config": { 14 | "models": [ 15 | ], 16 | "recommend_model_hyper_params": { 17 | "input_chunk_length":10, 18 | "output_chunk_length": 8, 19 | "norm": false 20 | 21 | } 22 | }, 23 | 24 | "evaluation_config" : { 25 | "metrics": ["mae", "mse", "rmse", "mape", "smape", {"name": "mase", "seasonality": 1}, "wape", "msmape"], 26 | "strategy_args": { 27 | "strategy_name": "fixed_forecast", 28 | "horizon": 8, 29 | "train_ratio_in_tv": 1, 30 | "seed": 2021, 31 | "deterministic": "efficient", 32 | "save_true_pred": false, 33 | "target_channel": null 34 | } 35 | }, 36 | 37 | 38 | "report_config" : { 39 | "aggregate_type": "mean", 40 | "report_metrics": ["mae", "mse", "rmse", "mape", "smape", "mase", "wape", "msmape"], 41 | "fill_type": "mean_value", 42 | "null_value_threshold": "0.3" 43 | } 44 | } 45 | -------------------------------------------------------------------------------- /config/fixed_forecast_config_quarterly.json: -------------------------------------------------------------------------------- 1 | { 2 | "data_config" : { 3 | "feature_dict": { 4 | "if_univariate": true, 5 | "if_trend": null, 6 | "has_timestamp": null, 7 | "if_season": null, 8 | "freq": "quarterly" 9 | }, 10 | "data_set_name": "large_forecast" 11 | }, 12 | 13 | "model_config": { 14 | "models": [ 15 | ], 16 | "recommend_model_hyper_params": { 17 | "input_chunk_length":10, 18 | "output_chunk_length": 8, 19 | "norm": false 20 | 21 | } 22 | }, 23 | 24 | "evaluation_config" : { 25 | "metrics": ["mae", "mse", "rmse", "mape", "smape", {"name": "mase", "seasonality": 4}, "wape", "msmape"], 26 | "strategy_args": { 27 | "strategy_name": "fixed_forecast", 28 | "horizon": 8, 29 | "train_ratio_in_tv": 1, 30 | "seed": 2021, 31 | "deterministic": "efficient", 32 | "save_true_pred": false, 33 | "target_channel": null 34 | } 35 | }, 36 | 37 | 38 | "report_config" : { 39 | "aggregate_type": "mean", 40 | "report_metrics": ["mae", "mse", "rmse", "mape", "smape", "mase", "wape", "msmape"], 41 | "fill_type": "mean_value", 42 | "null_value_threshold": "0.3" 43 | } 44 | } 45 | -------------------------------------------------------------------------------- /config/fixed_forecast_config_weekly.json: -------------------------------------------------------------------------------- 1 | { 2 | "data_config" : { 3 | "feature_dict": { 4 | "if_univariate": true, 5 | "if_trend": null, 6 | "has_timestamp": null, 7 | "if_season": null, 8 | "freq": "weekly" 9 | }, 10 | "data_set_name": "large_forecast" 11 | }, 12 | 13 | "model_config": { 14 | "models": [ 15 | ], 16 | "recommend_model_hyper_params": { 17 | "input_chunk_length":16, 18 | "output_chunk_length": 13, 19 | "norm": false 20 | } 21 | }, 22 | 23 | "evaluation_config" : { 24 | "metrics": ["mae", "mse", "rmse", "mape", "smape", {"name": "mase", "seasonality": 1}, "wape", "msmape"], 25 | "strategy_args": { 26 | "strategy_name": "fixed_forecast", 27 | "horizon": 13, 28 | "train_ratio_in_tv": 1, 29 | "seed": 2021, 30 | "deterministic": "efficient", 31 | "save_true_pred": false, 32 | "target_channel": null 33 | } 34 | }, 35 | 36 | 37 | "report_config" : { 38 | "aggregate_type": "mean", 39 | "report_metrics": ["mae", "mse", "rmse", "mape", "smape", "mase", "wape", "msmape"], 40 | "fill_type": "mean_value", 41 | "null_value_threshold": "0.3" 42 | } 43 | } 44 | -------------------------------------------------------------------------------- /config/fixed_forecast_config_yearly.json: -------------------------------------------------------------------------------- 1 | { 2 | "data_config" : { 3 | "feature_dict": { 4 | "if_univariate": true, 5 | "if_trend": null, 6 | "has_timestamp": null, 7 | "if_season": null, 8 | "freq": "yearly" 9 | }, 10 | "data_set_name": "large_forecast" 11 | }, 12 | 13 | "model_config": { 14 | "models": [ 15 | ], 16 | "recommend_model_hyper_params": { 17 | "input_chunk_length":7, 18 | "output_chunk_length": 6, 19 | "norm": false 20 | 21 | } 22 | }, 23 | 24 | "evaluation_config" : { 25 | "metrics": ["mae", "mse", "rmse", "mape", "smape", {"name": "mase", "seasonality": 1}, "wape", "msmape"], 26 | "strategy_args": { 27 | "strategy_name": "fixed_forecast", 28 | "horizon": 6, 29 | "train_ratio_in_tv": 1, 30 | "seed": 2021, 31 | "deterministic": "efficient", 32 | "save_true_pred": false, 33 | "target_channel": null 34 | } 35 | }, 36 | 37 | 38 | "report_config" : { 39 | "aggregate_type": "mean", 40 | "report_metrics": ["mae", "mse", "rmse", "mape", "smape", "mase", "wape", "msmape"], 41 | "fill_type": "mean_value", 42 | "null_value_threshold": "0.3" 43 | } 44 | } 45 | -------------------------------------------------------------------------------- /docs/figures/Comparison_with_Related_Libraries.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/docs/figures/Comparison_with_Related_Libraries.png -------------------------------------------------------------------------------- /docs/figures/Drop-last.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/docs/figures/Drop-last.png -------------------------------------------------------------------------------- /docs/figures/MultiGPUs.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/docs/figures/MultiGPUs.png -------------------------------------------------------------------------------- /docs/figures/Non-Gaussianity.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/docs/figures/Non-Gaussianity.png -------------------------------------------------------------------------------- /docs/figures/Pipeline.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/docs/figures/Pipeline.png -------------------------------------------------------------------------------- /docs/figures/QR.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/docs/figures/QR.png -------------------------------------------------------------------------------- /docs/figures/TFB-LOGO.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/docs/figures/TFB-LOGO.png -------------------------------------------------------------------------------- /docs/figures/correlation.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/docs/figures/correlation.png -------------------------------------------------------------------------------- /docs/figures/seasonal.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/docs/figures/seasonal.png -------------------------------------------------------------------------------- /docs/figures/shifting.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/docs/figures/shifting.png -------------------------------------------------------------------------------- /docs/figures/stationarity.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/docs/figures/stationarity.png -------------------------------------------------------------------------------- /docs/figures/transition.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/docs/figures/transition.png -------------------------------------------------------------------------------- /docs/figures/trend.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/docs/figures/trend.png -------------------------------------------------------------------------------- /docs/tutorials/steps_to_predict_only_a_subset_of_input_variables_chn.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/docs/tutorials/steps_to_predict_only_a_subset_of_input_variables_chn.pdf -------------------------------------------------------------------------------- /requirements-docker.txt: -------------------------------------------------------------------------------- 1 | darts==0.25.0 2 | matplotlib==3.7.5 3 | numpy==1.24.4 4 | pandas==1.5.3 5 | salesforce_merlion==2.0.2 6 | scikit_learn==1.3.2 7 | scipy==1.10.1 8 | statsmodels==0.14.1 9 | torch==2.4.1 10 | ray==2.10.0 11 | tqdm==4.66.4 12 | dash==2.17.0 13 | dash-bootstrap-components==1.6.0 14 | reformer-pytorch==1.4.4 15 | lightgbm==4.1.0 -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- 1 | darts==0.25.0 2 | matplotlib>=3.6.2 3 | numpy>=1.22.3 4 | pandas>=1.5.1 5 | scikit_learn>=1.1.3 6 | scipy>=1.8.0 7 | statsmodels>=0.14.0 8 | torch>=1.11.0 9 | ray>=2.6.3 10 | tqdm>=4.64.0 11 | dash>=2.9.3 12 | dash-bootstrap-components>=1.5.0 13 | reformer-pytorch==1.4.4 14 | lightgbm>=4.1.0 -------------------------------------------------------------------------------- /scripts/__init__.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/AQShunyi_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQShunyi.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQShunyi/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQShunyi.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQShunyi/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQShunyi.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQShunyi/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQShunyi.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQShunyi/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/AQShunyi_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQShunyi.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQShunyi/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQShunyi.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQShunyi/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQShunyi.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQShunyi/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQShunyi.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQShunyi/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/AQShunyi_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQShunyi.csv" --strategy-args '{"horizon": 96}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQShunyi/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQShunyi.csv" --strategy-args '{"horizon": 192}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQShunyi/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQShunyi.csv" --strategy-args '{"horizon": 336}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQShunyi/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQShunyi.csv" --strategy-args '{"horizon": 720}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 960, "n_epochs": 100, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQShunyi/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/AQShunyi_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQShunyi.csv" --strategy-args '{"horizon": 96}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQShunyi/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQShunyi.csv" --strategy-args '{"horizon": 192}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQShunyi/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQShunyi.csv" --strategy-args '{"horizon": 336}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQShunyi/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQShunyi.csv" --strategy-args '{"horizon": 720}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQShunyi/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/AQWan_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQWan.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQWan/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQWan.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQWan/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQWan.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQWan/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQWan.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQWan/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/AQWan_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQWan.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQWan/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQWan.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQWan/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQWan.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQWan/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQWan.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQWan/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/AQWan_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQWan.csv" --strategy-args '{"horizon": 96}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQWan/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQWan.csv" --strategy-args '{"horizon": 192}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQWan/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQWan.csv" --strategy-args '{"horizon": 336}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQWan/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQWan.csv" --strategy-args '{"horizon": 720}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 960, "n_epochs": 100, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQWan/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/AQWan_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQWan.csv" --strategy-args '{"horizon": 96}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQWan/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQWan.csv" --strategy-args '{"horizon": 192}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQWan/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQWan.csv" --strategy-args '{"horizon": 336}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQWan/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "AQWan.csv" --strategy-args '{"horizon": 720}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "AQWan/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Covid-19_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Covid-19.csv" --strategy-args '{"horizon": 24}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Covid-19/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Covid-19.csv" --strategy-args '{"horizon": 36}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Covid-19/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Covid-19.csv" --strategy-args '{"horizon": 48}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Covid-19/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Covid-19.csv" --strategy-args '{"horizon": 60}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Covid-19/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Covid-19_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Covid-19.csv" --strategy-args '{"horizon": 24}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 24}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Covid-19/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Covid-19.csv" --strategy-args '{"horizon": 36}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 36}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Covid-19/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Covid-19.csv" --strategy-args '{"horizon": 48}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 48}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Covid-19/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Covid-19.csv" --strategy-args '{"horizon": 60}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 60}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Covid-19/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Covid-19_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Covid-19.csv" --strategy-args '{"horizon": 24}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 24}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Covid-19/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Covid-19.csv" --strategy-args '{"horizon": 36}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 36}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Covid-19/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Covid-19.csv" --strategy-args '{"horizon": 48}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 48}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Covid-19/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Covid-19.csv" --strategy-args '{"horizon": 60}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 60}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Covid-19/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Covid-19_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Covid-19.csv" --strategy-args '{"horizon": 24}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Covid-19/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Covid-19.csv" --strategy-args '{"horizon": 36}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Covid-19/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Covid-19.csv" --strategy-args '{"horizon": 48}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Covid-19/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Covid-19.csv" --strategy-args '{"horizon": 60}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Covid-19/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/CzeLan_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "CzeLan.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "CzeLan/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "CzeLan.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "CzeLan/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "CzeLan.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "CzeLan/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "CzeLan.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "CzeLan/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/CzeLan_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "CzeLan.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "CzeLan/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "CzeLan.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "CzeLan/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "CzeLan.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "CzeLan/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "CzeLan.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "CzeLan/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/CzeLan_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "CzeLan.csv" --strategy-args '{"horizon": 96}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "CzeLan/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "CzeLan.csv" --strategy-args '{"horizon": 192}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "CzeLan/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "CzeLan.csv" --strategy-args '{"horizon": 336}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "CzeLan/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "CzeLan.csv" --strategy-args '{"horizon": 720}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 960, "n_epochs": 100, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "CzeLan/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/CzeLan_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "CzeLan.csv" --strategy-args '{"horizon": 96}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "CzeLan/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "CzeLan.csv" --strategy-args '{"horizon": 192}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "CzeLan/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "CzeLan.csv" --strategy-args '{"horizon": 336}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "CzeLan/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "CzeLan.csv" --strategy-args '{"horizon": 720}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "CzeLan/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTh1_script/Amplifier.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 192}' --model-name "amplifier.Amplifier" --model-hyper-params '{"SCI": 0, "batch_size": 32, "hidden_size": 64, "horizon": 192, "lr": 0.0005, "norm": true, "seq_len": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/Amplifier" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 336}' --model-name "amplifier.Amplifier" --model-hyper-params '{"SCI": 0, "batch_size": 32, "hidden_size": 64, "horizon": 336, "lr": 0.0005, "norm": true, "seq_len": 512}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/Amplifier" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 720}' --model-name "amplifier.Amplifier" --model-hyper-params '{"SCI": 0, "batch_size": 32, "hidden_size": 512, "horizon": 720, "lr": 0.03, "norm": true, "seq_len": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/Amplifier" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 96}' --model-name "amplifier.Amplifier" --model-hyper-params '{"SCI": 0, "batch_size": 32, "hidden_size": 64, "horizon": 96, "lr": 0.0005, "norm": true, "seq_len": 512}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/Amplifier" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTh1_script/FITS.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 96}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 96, "loss": "MSE", "lr": 0.0005, "norm": true, "patience": 20, "seq_len": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/FITS" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 192}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 192, "loss": "MSE", "lr": 0.0005, "norm": true, "patience": 20, "seq_len": 512}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/FITS" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 336}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 336, "loss": "MSE", "lr": 0.0005, "norm": true, "patience": 20, "seq_len": 512}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/FITS" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 720}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 720, "loss": "MSE", "lr": 0.0005, "norm": true, "patience": 20, "seq_len": 512}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/FITS" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTh1_script/HDMixer: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 96}' --model-name "hdmixer.HDMixer" --model-hyper-params '{"n_heads":4, "d_model": 16, "d_ff":32, "dropout":0.8, "e_layers":1, "batch_size":256, "lr": 5e-4, "fc_dropout":0.3, "head_dropout":0, "num_epochs": 50, "patience": 20, "horizon": 96, "norm": true, "seq_len": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/HDMixer" 2 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTh1_script/MICN.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 96}' --model-name "time_series_library.MICN" --model-hyper-params '{"d_ff": 512, "d_model": 128, "horizon": 96, "norm": true, "seq_len": 96}' --adapter "transformer_adapter" --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/MICN" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 192}' --model-name "time_series_library.MICN" --model-hyper-params '{"d_ff": 128, "d_model": 32, "horizon": 192, "norm": true, "seq_len": 512}' --adapter "transformer_adapter" --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/MICN" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 336}' --model-name "time_series_library.MICN" --model-hyper-params '{"d_ff": 128, "d_model": 32, "horizon": 336, "norm": true, "seq_len": 336}' --adapter "transformer_adapter" --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/MICN" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 720}' --model-name "time_series_library.MICN" --model-hyper-params '{"d_ff": 128, "d_model": 32, "horizon": 720, "norm": true, "seq_len": 512}' --adapter "transformer_adapter" --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/MICN" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTh1_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTh1_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTh1_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 96}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 192}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 336}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 720}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 960, "n_epochs": 100, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTh1_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 96}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 192}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 336}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 720}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTh1_script/xPatch.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh1.csv" --strategy-args '{"horizon": 96}' --model-name "xpatch.xPatch" --model-hyper-params '{"batch_size": 128, "lr": 0.0005, "num_epochs": 20, "patience": 5, "horizon": 96, "norm": true, "seq_len": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh1/xPatch" 2 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTh2_script/Amplifier.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 192}' --model-name "amplifier.Amplifier" --model-hyper-params '{"SCI": 0, "batch_size": 256, "hidden_size": 512, "horizon": 192, "lr": 0.02, "norm": true, "seq_len": 512}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/Amplifier" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 336}' --model-name "amplifier.Amplifier" --model-hyper-params '{"SCI": 0, "batch_size": 32, "hidden_size": 512, "horizon": 336, "lr": 0.03, "norm": true, "seq_len": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/Amplifier" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 720}' --model-name "amplifier.Amplifier" --model-hyper-params '{"SCI": 0, "batch_size": 256, "hidden_size": 512, "horizon": 720, "lr": 0.03, "norm": true, "seq_len": 512}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/Amplifier" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 96}' --model-name "amplifier.Amplifier" --model-hyper-params '{"SCI": 0, "batch_size": 32, "hidden_size": 64, "horizon": 96, "lr": 0.02, "norm": true, "seq_len": 512}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/Amplifier" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTh2_script/FITS.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 96}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 96, "loss": "MSE", "lr": 0.0005, "norm": true, "patience": 20, "seq_len": 512}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/FITS" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 192}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 192, "loss": "MSE", "lr": 0.0005, "norm": true, "patience": 20, "seq_len": 512}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/FITS" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 336}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 336, "loss": "MSE", "lr": 0.0005, "norm": true, "patience": 20, "seq_len": 512}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/FITS" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 720}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 720, "loss": "MSE", "lr": 0.0005, "norm": true, "patience": 20, "seq_len": 512}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/FITS" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTh2_script/MICN.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 96}' --model-name "time_series_library.MICN" --model-hyper-params '{"d_ff": 512, "d_model": 128, "horizon": 96, "norm": true, "seq_len": 336}' --adapter "transformer_adapter" --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/MICN" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 192}' --model-name "time_series_library.MICN" --model-hyper-params '{"d_ff": 512, "d_model": 128, "horizon": 192, "norm": true, "seq_len": 336}' --adapter "transformer_adapter" --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/MICN" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 336}' --model-name "time_series_library.MICN" --model-hyper-params '{"d_ff": 128, "d_model": 32, "horizon": 336, "norm": true, "seq_len": 336}' --adapter "transformer_adapter" --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/MICN" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 720}' --model-name "time_series_library.MICN" --model-hyper-params '{"d_ff": 128, "d_model": 32, "horizon": 720, "norm": true, "seq_len": 336}' --adapter "transformer_adapter" --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/MICN" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTh2_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTh2_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTh2_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 96}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 192}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 336}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 720}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 960, "n_epochs": 100, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTh2_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 96}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 192}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 336}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTh2.csv" --strategy-args '{"horizon": 720}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTh2/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTm1_script/MICN.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm1.csv" --strategy-args '{"horizon": 96}' --model-name "time_series_library.MICN" --model-hyper-params '{"d_ff": 64, "d_model": 32, "horizon": 96, "norm": true, "seq_len": 336}' --adapter "transformer_adapter" --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm1/MICN" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm1.csv" --strategy-args '{"horizon": 192}' --model-name "time_series_library.MICN" --model-hyper-params '{"d_ff": 64, "d_model": 32, "horizon": 192, "norm": true, "seq_len": 336}' --adapter "transformer_adapter" --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm1/MICN" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm1.csv" --strategy-args '{"horizon": 336}' --model-name "time_series_library.MICN" --model-hyper-params '{"d_ff": 64, "d_model": 32, "horizon": 336, "norm": true, "seq_len": 512}' --adapter "transformer_adapter" --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm1/MICN" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm1.csv" --strategy-args '{"horizon": 720}' --model-name "time_series_library.MICN" --model-hyper-params '{"d_ff": 64, "d_model": 32, "horizon": 720, "norm": true, "seq_len": 512}' --adapter "transformer_adapter" --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm1/MICN" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTm1_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm1.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm1/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm1.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm1/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm1.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm1/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm1.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm1/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTm1_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm1.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm1/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm1.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm1/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm1.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm1/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm1.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm1/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTm1_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm1.csv" --strategy-args '{"horizon": 96}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm1/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm1.csv" --strategy-args '{"horizon": 192}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm1/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm1.csv" --strategy-args '{"horizon": 336}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm1/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm1.csv" --strategy-args '{"horizon": 720}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 960, "n_epochs": 100, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm1/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTm1_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm1.csv" --strategy-args '{"horizon": 96}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm1/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm1.csv" --strategy-args '{"horizon": 192}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm1/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm1.csv" --strategy-args '{"horizon": 336}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm1/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm1.csv" --strategy-args '{"horizon": 720}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm1/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTm2_script/Amplifier.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 192}' --model-name "amplifier.Amplifier" --model-hyper-params '{"SCI": 0, "batch_size": 32, "hidden_size": 64, "horizon": 192, "lr": 0.0001, "norm": true, "seq_len": 512}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/Amplifier" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 336}' --model-name "amplifier.Amplifier" --model-hyper-params '{"SCI": 0, "batch_size": 256, "hidden_size": 512, "horizon": 336, "lr": 0.03, "norm": true, "seq_len": 512}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/Amplifier" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 720}' --model-name "amplifier.Amplifier" --model-hyper-params '{"SCI": 0, "batch_size": 32, "hidden_size": 64, "horizon": 720, "lr": 0.0001, "norm": true, "seq_len": 512}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/Amplifier" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 96}' --model-name "amplifier.Amplifier" --model-hyper-params '{"SCI": 0, "batch_size": 256, "hidden_size": 64, "horizon": 96, "lr": 0.02, "norm": true, "seq_len": 512}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/Amplifier" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTm2_script/MICN.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 96}' --model-name "time_series_library.MICN" --model-hyper-params '{"d_ff": 64, "d_model": 32, "horizon": 96, "norm": true, "seq_len": 512}' --adapter "transformer_adapter" --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/MICN" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 192}' --model-name "time_series_library.MICN" --model-hyper-params '{"d_ff": 64, "d_model": 32, "horizon": 192, "norm": true, "seq_len": 512}' --adapter "transformer_adapter" --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/MICN" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 336}' --model-name "time_series_library.MICN" --model-hyper-params '{"d_ff": 64, "d_model": 32, "horizon": 336, "norm": true, "seq_len": 512}' --adapter "transformer_adapter" --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/MICN" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 720}' --model-name "time_series_library.MICN" --model-hyper-params '{"d_ff": 64, "d_model": 32, "horizon": 720, "norm": true, "seq_len": 512}' --adapter "transformer_adapter" --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/MICN" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTm2_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTm2_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTm2_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 96}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 192}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 336}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 720}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 960, "n_epochs": 100, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ETTm2_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 96}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 192}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 336}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ETTm2.csv" --strategy-args '{"horizon": 720}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ETTm2/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Electricity_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Electricity.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Electricity/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Electricity.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Electricity/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Electricity.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Electricity/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Electricity.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Electricity/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Electricity_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Electricity.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Electricity/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Electricity.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Electricity/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Electricity.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Electricity/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Electricity.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Electricity/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Electricity_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Electricity.csv" --strategy-args '{"horizon": 96}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Electricity/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Electricity.csv" --strategy-args '{"horizon": 192}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Electricity/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Electricity.csv" --strategy-args '{"horizon": 336}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Electricity/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Electricity.csv" --strategy-args '{"horizon": 720}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 960, "n_epochs": 10, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Electricity/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Electricity_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Electricity.csv" --strategy-args '{"horizon": 96}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Electricity/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Electricity.csv" --strategy-args '{"horizon": 192}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Electricity/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Electricity.csv" --strategy-args '{"horizon": 336}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Electricity/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Electricity.csv" --strategy-args '{"horizon": 720}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Electricity/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Exchange_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Exchange.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Exchange/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Exchange.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Exchange/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Exchange.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Exchange/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Exchange.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Exchange/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Exchange_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Exchange.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Exchange/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Exchange.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Exchange/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Exchange.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Exchange/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Exchange.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Exchange/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Exchange_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Exchange.csv" --strategy-args '{"horizon": 96}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Exchange/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Exchange.csv" --strategy-args '{"horizon": 192}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Exchange/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Exchange.csv" --strategy-args '{"horizon": 336}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Exchange/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Exchange.csv" --strategy-args '{"horizon": 720}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 960, "n_epochs": 100, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Exchange/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Exchange_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Exchange.csv" --strategy-args '{"horizon": 96}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Exchange/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Exchange.csv" --strategy-args '{"horizon": 192}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Exchange/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Exchange.csv" --strategy-args '{"horizon": 336}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Exchange/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Exchange.csv" --strategy-args '{"horizon": 720}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Exchange/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/FRED-MD_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "FRED-MD.csv" --strategy-args '{"horizon": 24}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "FRED-MD/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "FRED-MD.csv" --strategy-args '{"horizon": 36}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "FRED-MD/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "FRED-MD.csv" --strategy-args '{"horizon": 48}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "FRED-MD/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "FRED-MD.csv" --strategy-args '{"horizon": 60}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "FRED-MD/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/FRED-MD_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "FRED-MD.csv" --strategy-args '{"horizon": 24}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 24}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "FRED-MD/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "FRED-MD.csv" --strategy-args '{"horizon": 36}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 36}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "FRED-MD/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "FRED-MD.csv" --strategy-args '{"horizon": 48}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 48}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "FRED-MD/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "FRED-MD.csv" --strategy-args '{"horizon": 60}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 60}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "FRED-MD/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/FRED-MD_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "FRED-MD.csv" --strategy-args '{"horizon": 24}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 24}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "FRED-MD/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "FRED-MD.csv" --strategy-args '{"horizon": 36}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 36}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "FRED-MD/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "FRED-MD.csv" --strategy-args '{"horizon": 48}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 48}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "FRED-MD/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "FRED-MD.csv" --strategy-args '{"horizon": 60}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 60}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "FRED-MD/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/FRED-MD_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "FRED-MD.csv" --strategy-args '{"horizon": 24}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "FRED-MD/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "FRED-MD.csv" --strategy-args '{"horizon": 36}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "FRED-MD/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "FRED-MD.csv" --strategy-args '{"horizon": 48}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "FRED-MD/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "FRED-MD.csv" --strategy-args '{"horizon": 60}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "FRED-MD/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ILI_script/FITS.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ILI.csv" --strategy-args '{"horizon": 24}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 24, "loss": "MSE", "lr": 0.005, "norm": true, "patience": 20, "seq_len": 104}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ILI/FITS" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ILI.csv" --strategy-args '{"horizon": 36}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 8, "batch_size": 64, "horizon": 36, "loss": "MSE", "lr": 0.005, "norm": true, "patience": 20, "seq_len": 104}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ILI/FITS" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ILI.csv" --strategy-args '{"horizon": 48}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 8, "batch_size": 64, "horizon": 48, "loss": "MSE", "lr": 0.005, "norm": true, "patience": 20, "seq_len": 104}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ILI/FITS" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ILI.csv" --strategy-args '{"horizon": 60}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 8, "batch_size": 64, "horizon": 60, "loss": "MSE", "lr": 0.005, "norm": true, "patience": 20, "seq_len": 104}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ILI/FITS" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ILI_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ILI.csv" --strategy-args '{"horizon": 24}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ILI/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ILI.csv" --strategy-args '{"horizon": 36}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ILI/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ILI.csv" --strategy-args '{"horizon": 48}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ILI/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ILI.csv" --strategy-args '{"horizon": 60}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ILI/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ILI_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ILI.csv" --strategy-args '{"horizon": 24}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 24}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ILI/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ILI.csv" --strategy-args '{"horizon": 36}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 36}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ILI/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ILI.csv" --strategy-args '{"horizon": 48}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 48}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ILI/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ILI.csv" --strategy-args '{"horizon": 60}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 60}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ILI/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ILI_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ILI.csv" --strategy-args '{"horizon": 24}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ILI/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ILI.csv" --strategy-args '{"horizon": 36}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 10, "norm": true, "output_chunk_length": 36}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ILI/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ILI.csv" --strategy-args '{"horizon": 48}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 10, "norm": true, "output_chunk_length": 48}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ILI/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ILI.csv" --strategy-args '{"horizon": 60}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 10, "norm": true, "output_chunk_length": 60}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ILI/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ILI_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ILI.csv" --strategy-args '{"horizon": 24}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ILI/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ILI.csv" --strategy-args '{"horizon": 36}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ILI/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ILI.csv" --strategy-args '{"horizon": 48}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ILI/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ILI.csv" --strategy-args '{"horizon": 60}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ILI/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/METR-LA_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "METR-LA.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "METR-LA/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "METR-LA.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "METR-LA/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "METR-LA.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "METR-LA/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "METR-LA.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "METR-LA/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/METR-LA_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "METR-LA.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "METR-LA/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "METR-LA.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "METR-LA/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "METR-LA.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "METR-LA/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "METR-LA.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "METR-LA/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/METR-LA_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "METR-LA.csv" --strategy-args '{"horizon": 96}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "METR-LA/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "METR-LA.csv" --strategy-args '{"horizon": 192}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "METR-LA/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "METR-LA.csv" --strategy-args '{"horizon": 336}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "METR-LA/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "METR-LA.csv" --strategy-args '{"horizon": 720}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 960, "n_epochs": 10, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "METR-LA/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/METR-LA_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "METR-LA.csv" --strategy-args '{"horizon": 96}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "METR-LA/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "METR-LA.csv" --strategy-args '{"horizon": 192}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "METR-LA/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "METR-LA.csv" --strategy-args '{"horizon": 336}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "METR-LA/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "METR-LA.csv" --strategy-args '{"horizon": 720}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "METR-LA/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/NASDAQ_script/FITS.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NASDAQ.csv" --strategy-args '{"horizon": 24}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 24, "loss": "MSE", "lr": 0.005, "norm": true, "patience": 20, "seq_len": 104}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NASDAQ/FITS" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NASDAQ.csv" --strategy-args '{"horizon": 36}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 36, "loss": "MSE", "lr": 0.005, "norm": true, "patience": 20, "seq_len": 104}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NASDAQ/FITS" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NASDAQ.csv" --strategy-args '{"horizon": 48}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 48, "loss": "MSE", "lr": 0.001, "norm": true, "patience": 5, "seq_len": 104}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NASDAQ/FITS" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NASDAQ.csv" --strategy-args '{"horizon": 60}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 60, "loss": "MSE", "lr": 0.001, "norm": true, "patience": 5, "seq_len": 104}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NASDAQ/FITS" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/NASDAQ_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NASDAQ.csv" --strategy-args '{"horizon": 24}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NASDAQ/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NASDAQ.csv" --strategy-args '{"horizon": 36}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NASDAQ/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NASDAQ.csv" --strategy-args '{"horizon": 48}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NASDAQ/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NASDAQ.csv" --strategy-args '{"horizon": 60}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NASDAQ/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/NASDAQ_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NASDAQ.csv" --strategy-args '{"horizon": 24}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 24}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NASDAQ/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NASDAQ.csv" --strategy-args '{"horizon": 36}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 36}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NASDAQ/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NASDAQ.csv" --strategy-args '{"horizon": 48}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 48}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NASDAQ/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NASDAQ.csv" --strategy-args '{"horizon": 60}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 60}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NASDAQ/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/NASDAQ_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NASDAQ.csv" --strategy-args '{"horizon": 24}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 24}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NASDAQ/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NASDAQ.csv" --strategy-args '{"horizon": 36}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 36}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NASDAQ/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NASDAQ.csv" --strategy-args '{"horizon": 48}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 48}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NASDAQ/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NASDAQ.csv" --strategy-args '{"horizon": 60}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 60}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NASDAQ/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/NASDAQ_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NASDAQ.csv" --strategy-args '{"horizon": 24}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NASDAQ/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NASDAQ.csv" --strategy-args '{"horizon": 36}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NASDAQ/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NASDAQ.csv" --strategy-args '{"horizon": 48}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NASDAQ/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NASDAQ.csv" --strategy-args '{"horizon": 60}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NASDAQ/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/NN5_script/FITS.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NN5.csv" --strategy-args '{"horizon": 24}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 24, "loss": "MSE", "lr": 0.005, "norm": true, "patience": 20, "seq_len": 104}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NN5/FITS" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NN5.csv" --strategy-args '{"horizon": 36}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 36, "loss": "MSE", "lr": 0.005, "norm": true, "patience": 20, "seq_len": 104}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NN5/FITS" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NN5.csv" --strategy-args '{"horizon": 48}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 48, "loss": "MSE", "lr": 0.005, "norm": true, "patience": 20, "seq_len": 104}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NN5/FITS" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NN5.csv" --strategy-args '{"horizon": 60}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 60, "loss": "MSE", "lr": 0.005, "norm": true, "patience": 20, "seq_len": 104}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NN5/FITS" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/NN5_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NN5.csv" --strategy-args '{"horizon": 24}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NN5/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NN5.csv" --strategy-args '{"horizon": 36}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NN5/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NN5.csv" --strategy-args '{"horizon": 48}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NN5/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NN5.csv" --strategy-args '{"horizon": 60}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NN5/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/NN5_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NN5.csv" --strategy-args '{"horizon": 24}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NN5/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NN5.csv" --strategy-args '{"horizon": 36}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NN5/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NN5.csv" --strategy-args '{"horizon": 48}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NN5/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NN5.csv" --strategy-args '{"horizon": 60}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NN5/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/NN5_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NN5.csv" --strategy-args '{"horizon": 24}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 24}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NN5/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NN5.csv" --strategy-args '{"horizon": 36}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 36}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NN5/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NN5.csv" --strategy-args '{"horizon": 48}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 48}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NN5/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NN5.csv" --strategy-args '{"horizon": 60}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 60}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NN5/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/NN5_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NN5.csv" --strategy-args '{"horizon": 24}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NN5/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NN5.csv" --strategy-args '{"horizon": 36}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NN5/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NN5.csv" --strategy-args '{"horizon": 48}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NN5/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NN5.csv" --strategy-args '{"horizon": 60}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NN5/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/NYSE_script/FITS.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NYSE.csv" --strategy-args '{"horizon": 24}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 24, "loss": "MSE", "lr": 0.005, "norm": true, "patience": 20, "seq_len": 104}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NYSE/FITS" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NYSE.csv" --strategy-args '{"horizon": 36}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 36, "loss": "MSE", "lr": 0.005, "norm": true, "patience": 20, "seq_len": 104}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NYSE/FITS" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NYSE.csv" --strategy-args '{"horizon": 48}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 48, "loss": "MSE", "lr": 0.005, "norm": true, "patience": 20, "seq_len": 104}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NYSE/FITS" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NYSE.csv" --strategy-args '{"horizon": 60}' --model-name "fits.FITS" --model-hyper-params '{"H_order": 6, "batch_size": 64, "horizon": 60, "loss": "MSE", "lr": 0.005, "norm": true, "patience": 20, "seq_len": 104}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NYSE/FITS" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/NYSE_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NYSE.csv" --strategy-args '{"horizon": 24}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NYSE/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NYSE.csv" --strategy-args '{"horizon": 36}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NYSE/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NYSE.csv" --strategy-args '{"horizon": 48}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NYSE/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NYSE.csv" --strategy-args '{"horizon": 60}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NYSE/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/NYSE_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NYSE.csv" --strategy-args '{"horizon": 24}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NYSE/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NYSE.csv" --strategy-args '{"horizon": 36}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NYSE/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NYSE.csv" --strategy-args '{"horizon": 48}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NYSE/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NYSE.csv" --strategy-args '{"horizon": 60}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NYSE/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/NYSE_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NYSE.csv" --strategy-args '{"horizon": 24}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 24}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NYSE/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NYSE.csv" --strategy-args '{"horizon": 36}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 36}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NYSE/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NYSE.csv" --strategy-args '{"horizon": 48}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 48}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NYSE/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NYSE.csv" --strategy-args '{"horizon": 60}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 60}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NYSE/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/NYSE_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NYSE.csv" --strategy-args '{"horizon": 24}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NYSE/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NYSE.csv" --strategy-args '{"horizon": 36}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NYSE/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NYSE.csv" --strategy-args '{"horizon": 48}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NYSE/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "NYSE.csv" --strategy-args '{"horizon": 60}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "NYSE/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/PEMS-BAY_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS-BAY.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS-BAY/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS-BAY.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS-BAY/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS-BAY.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS-BAY/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS-BAY.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS-BAY/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/PEMS-BAY_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS-BAY.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS-BAY/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS-BAY.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS-BAY/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS-BAY.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS-BAY/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS-BAY.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS-BAY/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/PEMS-BAY_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS-BAY.csv" --strategy-args '{"horizon": 96}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS-BAY/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS-BAY.csv" --strategy-args '{"horizon": 192}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS-BAY/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS-BAY.csv" --strategy-args '{"horizon": 336}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS-BAY/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS-BAY.csv" --strategy-args '{"horizon": 720}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 960, "n_epochs": 10, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS-BAY/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/PEMS-BAY_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS-BAY.csv" --strategy-args '{"horizon": 96}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS-BAY/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS-BAY.csv" --strategy-args '{"horizon": 192}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS-BAY/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS-BAY.csv" --strategy-args '{"horizon": 336}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS-BAY/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS-BAY.csv" --strategy-args '{"horizon": 720}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS-BAY/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/PEMS04_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS04.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS04/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS04.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS04/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS04.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS04/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS04.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS04/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/PEMS04_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS04.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS04/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS04.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS04/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS04.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS04/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS04.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS04/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/PEMS04_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS04.csv" --strategy-args '{"horizon": 96}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS04/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS04.csv" --strategy-args '{"horizon": 192}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS04/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS04.csv" --strategy-args '{"horizon": 336}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS04/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS04.csv" --strategy-args '{"horizon": 720}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 960, "n_epochs": 10, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS04/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/PEMS04_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS04.csv" --strategy-args '{"horizon": 96}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS04/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS04.csv" --strategy-args '{"horizon": 192}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS04/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS04.csv" --strategy-args '{"horizon": 336}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS04/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS04.csv" --strategy-args '{"horizon": 720}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS04/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/PEMS08_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS08.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS08/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS08.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS08/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS08.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS08/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS08.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS08/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/PEMS08_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS08.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS08/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS08.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS08/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS08.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS08/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS08.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS08/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/PEMS08_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS08.csv" --strategy-args '{"horizon": 96}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS08/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS08.csv" --strategy-args '{"horizon": 192}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS08/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS08.csv" --strategy-args '{"horizon": 336}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS08/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS08.csv" --strategy-args '{"horizon": 720}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 960, "n_epochs": 10, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS08/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/PEMS08_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS08.csv" --strategy-args '{"horizon": 96}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS08/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS08.csv" --strategy-args '{"horizon": 192}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS08/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS08.csv" --strategy-args '{"horizon": 336}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS08/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "PEMS08.csv" --strategy-args '{"horizon": 720}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "PEMS08/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Solar_script/Amplifier.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Solar.csv" --strategy-args '{"horizon": 192}' --model-name "amplifier.Amplifier" --model-hyper-params '{"SCI": 0, "batch_size": 32, "hidden_size": 512, "horizon": 192, "lr": 0.02, "norm": true, "seq_len": 512}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Solar/Amplifier" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Solar.csv" --strategy-args '{"horizon": 336}' --model-name "amplifier.Amplifier" --model-hyper-params '{"SCI": 0, "batch_size": 32, "hidden_size": 64, "horizon": 336, "lr": 0.0005, "norm": true, "seq_len": 512}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Solar/Amplifier" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Solar.csv" --strategy-args '{"horizon": 720}' --model-name "amplifier.Amplifier" --model-hyper-params '{"SCI": 0, "batch_size": 32, "hidden_size": 64, "horizon": 720, "lr": 0.0005, "norm": true, "seq_len": 512}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Solar/Amplifier" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Solar.csv" --strategy-args '{"horizon": 96}' --model-name "amplifier.Amplifier" --model-hyper-params '{"SCI": 0, "batch_size": 32, "hidden_size": 64, "horizon": 96, "lr": 0.02, "norm": true, "seq_len": 512}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Solar/Amplifier" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Solar_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Solar.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Solar/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Solar.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Solar/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Solar.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Solar/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Solar.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Solar/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Solar_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Solar.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Solar/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Solar.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Solar/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Solar.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Solar/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Solar.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Solar/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Solar_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Solar.csv" --strategy-args '{"horizon": 96}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Solar/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Solar.csv" --strategy-args '{"horizon": 192}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Solar/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Solar.csv" --strategy-args '{"horizon": 336}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Solar/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Solar.csv" --strategy-args '{"horizon": 720}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 960, "n_epochs": 100, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Solar/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Solar_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Solar.csv" --strategy-args '{"horizon": 96}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Solar/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Solar.csv" --strategy-args '{"horizon": 192}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Solar/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Solar.csv" --strategy-args '{"horizon": 336}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Solar/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Solar.csv" --strategy-args '{"horizon": 720}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Solar/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Traffic_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Traffic.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Traffic/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Traffic.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Traffic/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Traffic.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Traffic/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Traffic.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Traffic/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Traffic_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Traffic.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Traffic/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Traffic.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Traffic/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Traffic.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Traffic/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Traffic.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Traffic/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Traffic_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Traffic.csv" --strategy-args '{"horizon": 96}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Traffic/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Traffic.csv" --strategy-args '{"horizon": 192}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Traffic/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Traffic.csv" --strategy-args '{"horizon": 336}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Traffic/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Traffic.csv" --strategy-args '{"horizon": 720}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 960, "n_epochs": 10, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Traffic/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Traffic_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Traffic.csv" --strategy-args '{"horizon": 96}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Traffic/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Traffic.csv" --strategy-args '{"horizon": 192}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Traffic/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Traffic.csv" --strategy-args '{"horizon": 336}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Traffic/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Traffic.csv" --strategy-args '{"horizon": 720}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Traffic/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Weather_script/PatchMLP.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Weather.csv" --strategy-args '{"horizon": 96}' --model-name "patchmlp.PatchMLP" --model-hyper-params '{"data": "custom", "seq_len": 512, "horizon": 96, "enc_in": 21}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Weather/PatchMLP" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Weather.csv" --strategy-args '{"horizon": 192}' --model-name "patchmlp.PatchMLP" --model-hyper-params '{"data": "custom", "seq_len": 512, "horizon": 192, "enc_in": 21}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Weather/PatchMLP" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Weather.csv" --strategy-args '{"horizon": 336}' --model-name "patchmlp.PatchMLP" --model-hyper-params '{"data": "custom", "seq_len": 512, "horizon": 336, "enc_in": 21}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Weather/PatchMLP" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Weather.csv" --strategy-args '{"horizon": 720}' --model-name "patchmlp.PatchMLP" --model-hyper-params '{"data": "custom", "seq_len": 512, "horizon": 720, "enc_in": 21}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Weather/PatchMLP" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Weather_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Weather.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Weather/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Weather.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Weather/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Weather.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Weather/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Weather.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Weather/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Weather_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Weather.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Weather/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Weather.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Weather/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Weather.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Weather/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Weather.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Weather/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Weather_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Weather.csv" --strategy-args '{"horizon": 96}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Weather/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Weather.csv" --strategy-args '{"horizon": 192}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Weather/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Weather.csv" --strategy-args '{"horizon": 336}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 10, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Weather/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Weather.csv" --strategy-args '{"horizon": 720}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 960, "n_epochs": 10, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Weather/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Weather_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Weather.csv" --strategy-args '{"horizon": 96}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Weather/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Weather.csv" --strategy-args '{"horizon": 192}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Weather/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Weather.csv" --strategy-args '{"horizon": 336}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Weather/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Weather.csv" --strategy-args '{"horizon": 720}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Weather/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Wike2000_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wike2000.csv" --strategy-args '{"horizon": 24}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wike2000/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wike2000.csv" --strategy-args '{"horizon": 36}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wike2000/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wike2000.csv" --strategy-args '{"horizon": 48}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wike2000/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wike2000.csv" --strategy-args '{"horizon": 60}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wike2000/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Wike2000_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wike2000.csv" --strategy-args '{"horizon": 24}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wike2000/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wike2000.csv" --strategy-args '{"horizon": 36}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wike2000/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wike2000.csv" --strategy-args '{"horizon": 48}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wike2000/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wike2000.csv" --strategy-args '{"horizon": 60}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wike2000/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Wike2000_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wike2000.csv" --strategy-args '{"horizon": 24}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 24}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wike2000/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wike2000.csv" --strategy-args '{"horizon": 36}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 36}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wike2000/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wike2000.csv" --strategy-args '{"horizon": 48}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 48}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wike2000/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wike2000.csv" --strategy-args '{"horizon": 60}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 104, "n_epochs": 100, "norm": true, "output_chunk_length": 60}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wike2000/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Wike2000_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wike2000.csv" --strategy-args '{"horizon": 24}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wike2000/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wike2000.csv" --strategy-args '{"horizon": 36}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wike2000/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wike2000.csv" --strategy-args '{"horizon": 48}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wike2000/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wike2000.csv" --strategy-args '{"horizon": 60}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wike2000/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Wind_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wind.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wind/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wind.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wind/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wind.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wind/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wind.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wind/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Wind_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wind.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wind/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wind.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wind/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wind.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wind/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wind.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 1}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wind/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Wind_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wind.csv" --strategy-args '{"horizon": 96}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wind/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wind.csv" --strategy-args '{"horizon": 192}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wind/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wind.csv" --strategy-args '{"horizon": 336}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wind/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wind.csv" --strategy-args '{"horizon": 720}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 960, "n_epochs": 100, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wind/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/Wind_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wind.csv" --strategy-args '{"horizon": 96}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wind/VAR_model" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wind.csv" --strategy-args '{"horizon": 192}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wind/VAR_model" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wind.csv" --strategy-args '{"horizon": 336}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wind/VAR_model" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "Wind.csv" --strategy-args '{"horizon": 720}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "Wind/VAR_model" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ZafNoo_script/RNNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ZafNoo.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ZafNoo/RNNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ZafNoo.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ZafNoo/RNNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ZafNoo.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ZafNoo/RNNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ZafNoo.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RNNModel" --model-hyper-params '{"input_chunk_length": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ZafNoo/RNNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ZafNoo_script/RegressionModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ZafNoo.csv" --strategy-args '{"horizon": 96}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ZafNoo/RegressionModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ZafNoo.csv" --strategy-args '{"horizon": 192}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ZafNoo/RegressionModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ZafNoo.csv" --strategy-args '{"horizon": 336}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ZafNoo/RegressionModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ZafNoo.csv" --strategy-args '{"horizon": 720}' --model-name "darts.RegressionModel" --model-hyper-params '{"lags": 96, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ZafNoo/RegressionModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ZafNoo_script/TCNModel.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ZafNoo.csv" --strategy-args '{"horizon": 96}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 96}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ZafNoo/TCNModel" 2 | 3 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ZafNoo.csv" --strategy-args '{"horizon": 192}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 192}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ZafNoo/TCNModel" 4 | 5 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ZafNoo.csv" --strategy-args '{"horizon": 336}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 512, "n_epochs": 100, "norm": true, "output_chunk_length": 336}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ZafNoo/TCNModel" 6 | 7 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ZafNoo.csv" --strategy-args '{"horizon": 720}' --model-name "darts.TCNModel" --model-hyper-params '{"input_chunk_length": 960, "n_epochs": 100, "norm": true, "output_chunk_length": 720}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ZafNoo/TCNModel" 8 | 9 | -------------------------------------------------------------------------------- /scripts/multivariate_forecast/ZafNoo_script/VAR_model.sh: -------------------------------------------------------------------------------- 1 | python ./scripts/run_benchmark.py --config-path "rolling_forecast_config.json" --data-name-list "ZafNoo.csv" --strategy-args '{"horizon": 96}' --model-name "self_impl.VAR_model" --model-hyper-params '{}' --gpus 0 --num-workers 1 --timeout 60000 --save-path "ZafNoo/VAR_model" 2 | 3 | -------------------------------------------------------------------------------- /ts_benchmark/__init__.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/__init__.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | ADAPTER = { 3 | "darts_deep_model_adapter": "ts_benchmark.baselines.darts.darts_deep_model_adapter", 4 | "darts_statistical_model_adapter": "ts_benchmark.baselines.darts.darts_statistical_model_adapter", 5 | "darts_regression_model_adapter": "ts_benchmark.baselines.darts.darts_regression_model_adapter", 6 | "transformer_adapter": "ts_benchmark.baselines.time_series_library.adapters_for_transformers.transformer_adapter", 7 | } 8 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/amplifier/__init__.py: -------------------------------------------------------------------------------- 1 | __all__ = [ 2 | "Amplifier" 3 | ] 4 | 5 | from ts_benchmark.baselines.amplifier.amplifier import Amplifier 6 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/amplifier/layers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/ts_benchmark/baselines/amplifier/layers/__init__.py -------------------------------------------------------------------------------- /ts_benchmark/baselines/amplifier/models/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/ts_benchmark/baselines/amplifier/models/__init__.py -------------------------------------------------------------------------------- /ts_benchmark/baselines/duet/__init__.py: -------------------------------------------------------------------------------- 1 | __all__ = [ 2 | "DUET" 3 | ] 4 | 5 | from ts_benchmark.baselines.duet.duet import DUET 6 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/duet/layers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/ts_benchmark/baselines/duet/layers/__init__.py -------------------------------------------------------------------------------- /ts_benchmark/baselines/duet/layers/distributional_router_encoder.py: -------------------------------------------------------------------------------- 1 | import torch 2 | import torch.nn as nn 3 | 4 | 5 | class encoder(nn.Module): 6 | def __init__(self, config): 7 | super(encoder, self).__init__() 8 | input_size = config.seq_len 9 | num_experts = config.num_experts 10 | encoder_hidden_size = config.hidden_size 11 | 12 | self.distribution_fit = nn.Sequential(nn.Linear(input_size, encoder_hidden_size, bias=False), nn.ReLU(), 13 | nn.Linear(encoder_hidden_size, num_experts, bias=False)) 14 | 15 | def forward(self, x): 16 | mean = torch.mean(x, dim=-1) 17 | out = self.distribution_fit(mean) 18 | return out 19 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/duet/models/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/ts_benchmark/baselines/duet/models/__init__.py -------------------------------------------------------------------------------- /ts_benchmark/baselines/duet/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/ts_benchmark/baselines/duet/utils/__init__.py -------------------------------------------------------------------------------- /ts_benchmark/baselines/duet/utils/masking.py: -------------------------------------------------------------------------------- 1 | import torch 2 | 3 | 4 | class TriangularCausalMask(): 5 | def __init__(self, B, L, device="cpu"): 6 | mask_shape = [B, 1, L, L] 7 | with torch.no_grad(): 8 | self._mask = torch.triu(torch.ones(mask_shape, dtype=torch.bool), diagonal=1).to(device) 9 | 10 | @property 11 | def mask(self): 12 | return self._mask 13 | 14 | 15 | class ProbMask(): 16 | def __init__(self, B, H, L, index, scores, device="cpu"): 17 | _mask = torch.ones(L, scores.shape[-1], dtype=torch.bool).to(device).triu(1) 18 | _mask_ex = _mask[None, None, :].expand(B, H, L, scores.shape[-1]) 19 | indicator = _mask_ex[torch.arange(B)[:, None, None], 20 | torch.arange(H)[None, :, None], 21 | index, :].to(device) 22 | self._mask = indicator.view(scores.shape).to(device) 23 | 24 | @property 25 | def mask(self): 26 | return self._mask 27 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/fits/__init__.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | __all__ = [ 3 | "FITS", 4 | ] 5 | 6 | 7 | from ts_benchmark.baselines.fits.fits import FITS 8 | 9 | 10 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/hdmixer/__init__.py: -------------------------------------------------------------------------------- 1 | __all__ = [ 2 | "HDMixer" 3 | ] 4 | 5 | from ts_benchmark.baselines.hdmixer.hdmixer import HDMixer 6 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/patchmlp/__init__.py: -------------------------------------------------------------------------------- 1 | __all__ = ["PatchMLP"] 2 | 3 | from ts_benchmark.baselines.patchmlp.patchmlp import PatchMLP 4 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/patchmlp/layers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/ts_benchmark/baselines/patchmlp/layers/__init__.py -------------------------------------------------------------------------------- /ts_benchmark/baselines/patchmlp/models/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/ts_benchmark/baselines/patchmlp/models/__init__.py -------------------------------------------------------------------------------- /ts_benchmark/baselines/patchmlp/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/ts_benchmark/baselines/patchmlp/utils/__init__.py -------------------------------------------------------------------------------- /ts_benchmark/baselines/patchmlp/utils/masking.py: -------------------------------------------------------------------------------- 1 | import torch 2 | 3 | 4 | class TriangularCausalMask: 5 | def __init__(self, B, L, device="cpu"): 6 | mask_shape = [B, 1, L, L] 7 | with torch.no_grad(): 8 | self._mask = torch.triu( 9 | torch.ones(mask_shape, dtype=torch.bool), diagonal=1 10 | ).to(device) 11 | 12 | @property 13 | def mask(self): 14 | return self._mask 15 | 16 | 17 | class ProbMask: 18 | def __init__(self, B, H, L, index, scores, device="cpu"): 19 | _mask = torch.ones(L, scores.shape[-1], dtype=torch.bool).to(device).triu(1) 20 | _mask_ex = _mask[None, None, :].expand(B, H, L, scores.shape[-1]) 21 | indicator = _mask_ex[ 22 | torch.arange(B)[:, None, None], torch.arange(H)[None, :, None], index, : 23 | ].to(device) 24 | self._mask = indicator.view(scores.shape).to(device) 25 | 26 | @property 27 | def mask(self): 28 | return self._mask 29 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/patchmlp/utils/metrics.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | 3 | 4 | def RSE(pred, true): 5 | return np.sqrt(np.sum((true - pred) ** 2)) / np.sqrt( 6 | np.sum((true - true.mean()) ** 2) 7 | ) 8 | 9 | 10 | def CORR(pred, true): 11 | u = ((true - true.mean(0)) * (pred - pred.mean(0))).sum(0) 12 | d = np.sqrt(((true - true.mean(0)) ** 2 * (pred - pred.mean(0)) ** 2).sum(0)) 13 | return (u / d).mean(-1) 14 | 15 | 16 | def MAE(pred, true): 17 | return np.mean(np.abs(pred - true)) 18 | 19 | 20 | def MSE(pred, true): 21 | return np.mean((pred - true) ** 2) 22 | 23 | 24 | def RMSE(pred, true): 25 | return np.sqrt(MSE(pred, true)) 26 | 27 | 28 | def MAPE(pred, true): 29 | return np.mean(np.abs((pred - true) / true)) 30 | 31 | 32 | def MSPE(pred, true): 33 | return np.mean(np.square((pred - true) / true)) 34 | 35 | 36 | def metric(pred, true): 37 | mae = MAE(pred, true) 38 | mse = MSE(pred, true) 39 | rmse = RMSE(pred, true) 40 | mape = MAPE(pred, true) 41 | mspe = MSPE(pred, true) 42 | 43 | return mae, mse, rmse, mape, mspe 44 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/pathformer/__init__.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | __all__ = [ 3 | "Pathformer", 4 | ] 5 | 6 | 7 | from ts_benchmark.baselines.pathformer.pathformer import Pathformer 8 | 9 | 10 | 11 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/pathformer/layers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/ts_benchmark/baselines/pathformer/layers/__init__.py -------------------------------------------------------------------------------- /ts_benchmark/baselines/pathformer/models/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/ts_benchmark/baselines/pathformer/models/__init__.py -------------------------------------------------------------------------------- /ts_benchmark/baselines/pathformer/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/ts_benchmark/baselines/pathformer/utils/__init__.py -------------------------------------------------------------------------------- /ts_benchmark/baselines/pathformer/utils/masking.py: -------------------------------------------------------------------------------- 1 | import torch 2 | 3 | 4 | class TriangularCausalMask: 5 | def __init__(self, B, L, device="cpu"): 6 | mask_shape = [B, 1, L, L] 7 | with torch.no_grad(): 8 | self._mask = torch.triu( 9 | torch.ones(mask_shape, dtype=torch.bool), diagonal=1 10 | ).to(device) 11 | 12 | @property 13 | def mask(self): 14 | return self._mask 15 | 16 | 17 | class ProbMask: 18 | def __init__(self, B, H, L, index, scores, device="cpu"): 19 | _mask = torch.ones(L, scores.shape[-1], dtype=torch.bool).to(device).triu(1) 20 | _mask_ex = _mask[None, None, :].expand(B, H, L, scores.shape[-1]) 21 | indicator = _mask_ex[ 22 | torch.arange(B)[:, None, None], torch.arange(H)[None, :, None], index, : 23 | ].to(device) 24 | self._mask = indicator.view(scores.shape).to(device) 25 | 26 | @property 27 | def mask(self): 28 | return self._mask 29 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/pathformer/utils/metrics.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | 3 | 4 | def RSE(pred, true): 5 | return np.sqrt(np.sum((true - pred) ** 2)) / np.sqrt( 6 | np.sum((true - true.mean()) ** 2) 7 | ) 8 | 9 | 10 | def CORR(pred, true): 11 | u = ((true - true.mean(0)) * (pred - pred.mean(0))).sum(0) 12 | d = np.sqrt(((true - true.mean(0)) ** 2 * (pred - pred.mean(0)) ** 2).sum(0)) 13 | d += 1e-12 14 | return 0.01 * (u / d).mean(-1) 15 | 16 | 17 | def MAE(pred, true): 18 | return np.mean(np.abs(pred - true)) 19 | 20 | 21 | def MSE(pred, true): 22 | return np.mean((pred - true) ** 2) 23 | 24 | 25 | def RMSE(pred, true): 26 | return np.sqrt(MSE(pred, true)) 27 | 28 | 29 | def MAPE(pred, true): 30 | return np.mean(np.abs((pred - true) / true)) 31 | 32 | 33 | def MSPE(pred, true): 34 | return np.mean(np.square((pred - true) / true)) 35 | 36 | 37 | def metric(pred, true): 38 | mae = MAE(pred, true) 39 | mse = MSE(pred, true) 40 | rmse = RMSE(pred, true) 41 | mape = MAPE(pred, true) 42 | mspe = MSPE(pred, true) 43 | rse = RSE(pred, true) 44 | corr = CORR(pred, true) 45 | 46 | return mae, mse, rmse, mape, mspe, rse, corr 47 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/pdf/__init__.py: -------------------------------------------------------------------------------- 1 | ''' 2 | * @author: EmpyreanMoon 3 | * 4 | * @create: 2024-10-05 00:59 5 | * 6 | * @description: 7 | ''' 8 | __all__ = [ 9 | "PDF", 10 | ] 11 | 12 | 13 | from ts_benchmark.baselines.pdf.PDF import PDF 14 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/pdf/layers/__init__.py: -------------------------------------------------------------------------------- 1 | ''' 2 | * @author: EmpyreanMoon 3 | * 4 | * @create: 2024-10-05 00:59 5 | * 6 | * @description: 7 | ''' 8 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/pdf/models/__init__.py: -------------------------------------------------------------------------------- 1 | ''' 2 | * @author: EmpyreanMoon 3 | * 4 | * @create: 2024-10-05 00:59 5 | * 6 | * @description: 7 | ''' 8 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/pdf/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/ts_benchmark/baselines/pdf/utils/__init__.py -------------------------------------------------------------------------------- /ts_benchmark/baselines/pdf/utils/masking.py: -------------------------------------------------------------------------------- 1 | import torch 2 | 3 | 4 | class TriangularCausalMask: 5 | def __init__(self, B, L, device="cpu"): 6 | mask_shape = [B, 1, L, L] 7 | with torch.no_grad(): 8 | self._mask = torch.triu( 9 | torch.ones(mask_shape, dtype=torch.bool), diagonal=1 10 | ).to(device) 11 | 12 | @property 13 | def mask(self): 14 | return self._mask 15 | 16 | 17 | class ProbMask: 18 | def __init__(self, B, H, L, index, scores, device="cpu"): 19 | _mask = torch.ones(L, scores.shape[-1], dtype=torch.bool).to(device).triu(1) 20 | _mask_ex = _mask[None, None, :].expand(B, H, L, scores.shape[-1]) 21 | indicator = _mask_ex[ 22 | torch.arange(B)[:, None, None], torch.arange(H)[None, :, None], index, : 23 | ].to(device) 24 | self._mask = indicator.view(scores.shape).to(device) 25 | 26 | @property 27 | def mask(self): 28 | return self._mask 29 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/pdf/utils/metrics.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | 3 | 4 | def RSE(pred, true): 5 | return np.sqrt(np.sum((true - pred) ** 2)) / np.sqrt( 6 | np.sum((true - true.mean()) ** 2) 7 | ) 8 | 9 | 10 | def CORR(pred, true): 11 | u = ((true - true.mean(0)) * (pred - pred.mean(0))).sum(0) 12 | d = np.sqrt(((true - true.mean(0)) ** 2 * (pred - pred.mean(0)) ** 2).sum(0)) 13 | d += 1e-12 14 | return 0.01 * (u / d).mean(-1) 15 | 16 | 17 | def MAE(pred, true): 18 | return np.mean(np.abs(pred - true)) 19 | 20 | 21 | def MSE(pred, true): 22 | return np.mean((pred - true) ** 2) 23 | 24 | 25 | def RMSE(pred, true): 26 | return np.sqrt(MSE(pred, true)) 27 | 28 | 29 | def MAPE(pred, true): 30 | return np.mean(np.abs((pred - true) / true)) 31 | 32 | 33 | def MSPE(pred, true): 34 | return np.mean(np.square((pred - true) / true)) 35 | 36 | 37 | def metric(pred, true): 38 | mae = MAE(pred, true) 39 | mse = MSE(pred, true) 40 | rmse = RMSE(pred, true) 41 | mape = MAPE(pred, true) 42 | mspe = MSPE(pred, true) 43 | rse = RSE(pred, true) 44 | corr = CORR(pred, true) 45 | 46 | return mae, mse, rmse, mape, mspe, rse, corr 47 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/self_impl/__init__.py: -------------------------------------------------------------------------------- 1 | __all__ = [ 2 | "VAR_model", 3 | ] 4 | 5 | 6 | from ts_benchmark.baselines.self_impl.VAR.VAR import VAR_model 7 | 8 | 9 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/tag_csv/darts_tag.csv: -------------------------------------------------------------------------------- 1 | Methods_name,Allow_validation 2 | ARIMA,FALSE 3 | DLinearModel,TRUE 4 | NBEATSModel,TRUE 5 | NLinearModel,TRUE 6 | RNNModel,TRUE 7 | TCNModel,TRUE 8 | AutoARIMA,FALSE 9 | StatsForecastAutoARIMA,FALSE 10 | ExponentialSmoothing,FALSE 11 | StatsForecastAutoETS,FALSE 12 | StatsForecastAutoCES,FALSE 13 | StatsForecastAutoTheta,FALSE 14 | FourTheta,FALSE 15 | FFT,FALSE 16 | KalmanForecaster,FALSE 17 | Croston,FALSE 18 | RegressionModel,TRUE 19 | RandomForest,FALSE 20 | LinearRegressionModel,TRUE 21 | LightGBMModel,TRUE 22 | CatBoostModel,TRUE 23 | XGBModel,TRUE 24 | BlockRNNModel,TRUE 25 | NHiTSModel,TRUE 26 | TransformerModel,TRUE 27 | TFTModel,TRUE 28 | TiDEModel,TRUE 29 | NaiveDrift,FALSE 30 | VARIMA,FALSE 31 | NaiveMean,FALSE 32 | NaiveSeasonal,FALSE 33 | NaiveMovingAverage,FALSE -------------------------------------------------------------------------------- /ts_benchmark/baselines/tag_csv/other_methods_tag.csv: -------------------------------------------------------------------------------- 1 | Methods_name,Allow_validation 2 | VAR_model,TRUE -------------------------------------------------------------------------------- /ts_benchmark/baselines/tag_csv/tslib_tag.csv: -------------------------------------------------------------------------------- 1 | Methods_name,Allow_validation 2 | Autoformer,TRUE 3 | Crossformer,TRUE 4 | DLinear,TRUE 5 | ETSformer,TRUE 6 | FEDformer,TRUE 7 | FiLM,TRUE 8 | Informer,TRUE 9 | iTransformer,TRUE 10 | Koopa,TRUE 11 | LightTS,TRUE 12 | Linear,TRUE 13 | MICN,TRUE 14 | NLinear,TRUE 15 | Nonstationary_Transformer,TRUE 16 | PatchTST,TRUE 17 | Pyraformer,TRUE 18 | Reformer,TRUE 19 | TimesNet,TRUE 20 | Transformer,TRUE 21 | Triformer,TRUE -------------------------------------------------------------------------------- /ts_benchmark/baselines/time_series_library/layers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/ts_benchmark/baselines/time_series_library/layers/__init__.py -------------------------------------------------------------------------------- /ts_benchmark/baselines/time_series_library/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/ts_benchmark/baselines/time_series_library/utils/__init__.py -------------------------------------------------------------------------------- /ts_benchmark/baselines/time_series_library/utils/masking.py: -------------------------------------------------------------------------------- 1 | import torch 2 | 3 | 4 | class TriangularCausalMask(): 5 | def __init__(self, B, L, device="cpu"): 6 | mask_shape = [B, 1, L, L] 7 | with torch.no_grad(): 8 | self._mask = torch.triu(torch.ones(mask_shape, dtype=torch.bool), diagonal=1).to(device) 9 | 10 | @property 11 | def mask(self): 12 | return self._mask 13 | 14 | 15 | class ProbMask(): 16 | def __init__(self, B, H, L, index, scores, device="cpu"): 17 | _mask = torch.ones(L, scores.shape[-1], dtype=torch.bool).to(device).triu(1) 18 | _mask_ex = _mask[None, None, :].expand(B, H, L, scores.shape[-1]) 19 | indicator = _mask_ex[torch.arange(B)[:, None, None], 20 | torch.arange(H)[None, :, None], 21 | index, :].to(device) 22 | self._mask = indicator.view(scores.shape).to(device) 23 | 24 | @property 25 | def mask(self): 26 | return self._mask 27 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/timekan/__init__.py: -------------------------------------------------------------------------------- 1 | __all__ = ["TimeKAN"] 2 | 3 | from ts_benchmark.baselines.timekan.timekan import TimeKAN 4 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/timekan/layers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/ts_benchmark/baselines/timekan/layers/__init__.py -------------------------------------------------------------------------------- /ts_benchmark/baselines/timekan/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/ts_benchmark/baselines/timekan/utils/__init__.py -------------------------------------------------------------------------------- /ts_benchmark/baselines/timekan/utils/masking.py: -------------------------------------------------------------------------------- 1 | import torch 2 | 3 | 4 | class TriangularCausalMask(): 5 | def __init__(self, B, L, device="cpu"): 6 | mask_shape = [B, 1, L, L] 7 | with torch.no_grad(): 8 | self._mask = torch.triu(torch.ones(mask_shape, dtype=torch.bool), diagonal=1).to(device) 9 | 10 | @property 11 | def mask(self): 12 | return self._mask 13 | 14 | 15 | class ProbMask(): 16 | def __init__(self, B, H, L, index, scores, device="cpu"): 17 | _mask = torch.ones(L, scores.shape[-1], dtype=torch.bool).to(device).triu(1) 18 | _mask_ex = _mask[None, None, :].expand(B, H, L, scores.shape[-1]) 19 | indicator = _mask_ex[torch.arange(B)[:, None, None], 20 | torch.arange(H)[None, :, None], 21 | index, :].to(device) 22 | self._mask = indicator.view(scores.shape).to(device) 23 | 24 | @property 25 | def mask(self): 26 | return self._mask 27 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/timekan/utils/metrics.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | 3 | 4 | def RSE(pred, true): 5 | return np.sqrt(np.sum((true - pred) ** 2)) / np.sqrt(np.sum((true - true.mean()) ** 2)) 6 | 7 | 8 | def CORR(pred, true): 9 | u = ((true - true.mean(0)) * (pred - pred.mean(0))).sum(0) 10 | d = np.sqrt(((true - true.mean(0)) ** 2 * (pred - pred.mean(0)) ** 2).sum(0)) 11 | return (u / d).mean(-1) 12 | 13 | 14 | def MAE(pred, true): 15 | return np.mean(np.abs(pred - true)) 16 | 17 | 18 | def MSE(pred, true): 19 | return np.mean((pred - true) ** 2) 20 | 21 | 22 | def RMSE(pred, true): 23 | return np.sqrt(MSE(pred, true)) 24 | 25 | 26 | def MAPE(pred, true): 27 | mape = np.abs((pred - true) / true) 28 | mape = np.where(mape > 5, 0, mape) 29 | return np.mean(mape) 30 | 31 | 32 | def MSPE(pred, true): 33 | return np.mean(np.square((pred - true) / true)) 34 | 35 | 36 | def metric(pred, true): 37 | mae = MAE(pred, true) 38 | mse = MSE(pred, true) 39 | rmse = RMSE(pred, true) 40 | mape = MAPE(pred, true) 41 | mspe = MSPE(pred, true) 42 | 43 | return mae, mse, rmse, mape, mspe 44 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/xpatch/__init__.py: -------------------------------------------------------------------------------- 1 | __all__ = [ 2 | "xPatch" 3 | ] 4 | 5 | from ts_benchmark.baselines.xpatch.xpatch import xPatch 6 | -------------------------------------------------------------------------------- /ts_benchmark/baselines/xpatch/layers/decomp.py: -------------------------------------------------------------------------------- 1 | import torch 2 | from torch import nn 3 | 4 | from ..layers.ema import EMA 5 | from ..layers.dema import DEMA 6 | 7 | class DECOMP(nn.Module): 8 | """ 9 | Series decomposition block 10 | """ 11 | def __init__(self, ma_type, alpha, beta): 12 | super(DECOMP, self).__init__() 13 | if ma_type == 'ema': 14 | self.ma = EMA(alpha) 15 | elif ma_type == 'dema': 16 | self.ma = DEMA(alpha, beta) 17 | 18 | def forward(self, x): 19 | moving_average = self.ma(x) 20 | res = x - moving_average 21 | return res, moving_average -------------------------------------------------------------------------------- /ts_benchmark/baselines/xpatch/layers/dema.py: -------------------------------------------------------------------------------- 1 | import torch 2 | from torch import nn 3 | 4 | class DEMA(nn.Module): 5 | """ 6 | Double Exponential Moving Average (DEMA) block to highlight the trend of time series 7 | """ 8 | def __init__(self, alpha, beta): 9 | super(DEMA, self).__init__() 10 | # self.alpha = nn.Parameter(alpha) # Learnable alpha 11 | # self.beta = nn.Parameter(beta) # Learnable beta 12 | self.alpha = alpha.to(device='cuda') 13 | self.beta = beta.to(device='cuda') 14 | 15 | def forward(self, x): 16 | # self.alpha.data.clamp_(0, 1) # Clamp learnable alpha to [0, 1] 17 | # self.beta.data.clamp_(0, 1) # Clamp learnable beta to [0, 1] 18 | s_prev = x[:, 0, :] 19 | b = x[:, 1, :] - s_prev 20 | res = [s_prev.unsqueeze(1)] 21 | for t in range(1, x.shape[1]): 22 | xt = x[:, t, :] 23 | s = self.alpha * xt + (1 - self.alpha) * (s_prev + b) 24 | b = self.beta * (s - s_prev) + (1 - self.beta) * b 25 | s_prev = s 26 | res.append(s.unsqueeze(1)) 27 | return torch.cat(res, dim=1) -------------------------------------------------------------------------------- /ts_benchmark/baselines/xpatch/utils/metrics.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | 3 | 4 | def RSE(pred, true): 5 | return np.sqrt(np.sum((true - pred) ** 2)) / np.sqrt(np.sum((true - true.mean()) ** 2)) 6 | 7 | 8 | def CORR(pred, true): 9 | u = ((true - true.mean(0)) * (pred - pred.mean(0))).sum(0) 10 | d = np.sqrt(((true - true.mean(0)) ** 2 * (pred - pred.mean(0)) ** 2).sum(0)) 11 | d += 1e-12 12 | return 0.01*(u / d).mean(-1) 13 | 14 | 15 | def MAE(pred, true): 16 | return np.mean(np.abs(pred - true)) 17 | 18 | 19 | def MSE(pred, true): 20 | return np.mean((pred - true) ** 2) 21 | 22 | 23 | def RMSE(pred, true): 24 | return np.sqrt(MSE(pred, true)) 25 | 26 | 27 | def MAPE(pred, true): 28 | return np.mean(np.abs((pred - true) / true)) 29 | 30 | 31 | def MSPE(pred, true): 32 | return np.mean(np.square((pred - true) / true)) 33 | 34 | 35 | def metric(pred, true): 36 | mae = MAE(pred, true) 37 | mse = MSE(pred, true) 38 | # rmse = RMSE(pred, true) 39 | # mape = MAPE(pred, true) 40 | # mspe = MSPE(pred, true) 41 | # rse = RSE(pred, true) 42 | # corr = CORR(pred, true) 43 | 44 | return mae, mse 45 | # return mae, mse, rmse, mape, mspe, rse, corr -------------------------------------------------------------------------------- /ts_benchmark/common/__init__.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | -------------------------------------------------------------------------------- /ts_benchmark/common/constant.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | import os 3 | 4 | # Get the root path where the code file is located 5 | ROOT_PATH = os.path.abspath(os.path.join(__file__, "..", "..", "..")) 6 | 7 | # Build the path to the dataset folder 8 | FORECASTING_DATASET_PATH = os.path.join(ROOT_PATH, "dataset", "forecasting") 9 | 10 | # Profile Path 11 | CONFIG_PATH = os.path.join(ROOT_PATH, "config") 12 | 13 | # third-party library path 14 | THIRD_PARTY_PATH = os.path.join(ROOT_PATH, "ts_benchmark", "baselines", "third_party") 15 | -------------------------------------------------------------------------------- /ts_benchmark/data/__init__.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | """ 3 | Utilities for data managing, loading and serving 4 | """ -------------------------------------------------------------------------------- /ts_benchmark/data/data_pool.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | from typing import NoReturn 3 | 4 | from ts_benchmark.data.data_pool_impl_base import DataPoolImpl 5 | from ts_benchmark.utils.design_pattern import Singleton 6 | 7 | 8 | class DataPool(metaclass=Singleton): 9 | """ 10 | The global interface of data pools 11 | """ 12 | 13 | def __init__(self): 14 | self.pool = None 15 | 16 | def set_pool(self, pool: DataPoolImpl) -> NoReturn: 17 | """ 18 | Set the global data pool object 19 | 20 | :param pool: a DataPoolImpl object. 21 | """ 22 | self.pool = pool 23 | 24 | def get_pool(self) -> DataPoolImpl: 25 | """ 26 | Get the global data pool object 27 | """ 28 | return self.pool 29 | -------------------------------------------------------------------------------- /ts_benchmark/data/data_pool_impl_base.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | import abc 3 | from typing import Optional 4 | 5 | import pandas as pd 6 | 7 | 8 | class DataPoolImpl(metaclass=abc.ABCMeta): 9 | """ 10 | Base class for data pool implementations 11 | 12 | This class acts as a data client in each worker that feeds data to the pipeline. 13 | Techniques such as local caching may be adopted to improve performance. 14 | """ 15 | 16 | @abc.abstractmethod 17 | def get_series(self, name: str) -> Optional[pd.DataFrame]: 18 | """ 19 | Gets time series data by name 20 | 21 | The returned DataFrame follows the OTB protocol. 22 | 23 | :param name: The name of the series to get. 24 | :return: Time series data in DataFrame format. If the time series is not available, 25 | return None. 26 | """ 27 | 28 | @abc.abstractmethod 29 | def get_series_meta_info(self, name: str) -> Optional[pd.Series]: 30 | """ 31 | Gets the meta information of time series by name 32 | 33 | We do not return the meta information of unexisting series even if 34 | the meta information itself is available. 35 | 36 | :param name: The name of the series to get. 37 | :return: Meta information data in Series format. If the meta information or the 38 | corresponding series is not available, return None. 39 | """ 40 | -------------------------------------------------------------------------------- /ts_benchmark/data/data_server_base.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | import abc 3 | from typing import NoReturn 4 | 5 | from ts_benchmark.data.data_source import DataSource 6 | 7 | 8 | class DataServer(metaclass=abc.ABCMeta): 9 | """ 10 | Base class for data servers 11 | 12 | Data servers are responsible for sharing data to the workers 13 | through shared storage, message queue, etc. 14 | """ 15 | 16 | @abc.abstractmethod 17 | def start_async(self) -> NoReturn: 18 | """ 19 | Start the data server in non-blocking mode 20 | """ 21 | -------------------------------------------------------------------------------- /ts_benchmark/data/suites/__init__.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | """ 3 | Suites of data server/pool and other components 4 | """ 5 | -------------------------------------------------------------------------------- /ts_benchmark/evaluation/__init__.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | 3 | -------------------------------------------------------------------------------- /ts_benchmark/evaluation/metrics/__init__.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | from ts_benchmark.evaluation.metrics import regression_metrics 3 | 4 | REGRESSION_METRICS = { 5 | k: getattr(regression_metrics, k) for k in regression_metrics.__all__ 6 | } 7 | METRICS = { 8 | **REGRESSION_METRICS, 9 | } -------------------------------------------------------------------------------- /ts_benchmark/evaluation/strategy/__init__.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | from ts_benchmark.evaluation.strategy.fixed_forecast import FixedForecast 3 | from ts_benchmark.evaluation.strategy.rolling_forecast import RollingForecast 4 | 5 | STRATEGY = { 6 | "fixed_forecast": FixedForecast, 7 | "rolling_forecast": RollingForecast, 8 | } -------------------------------------------------------------------------------- /ts_benchmark/evaluation/strategy/constants.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | 3 | from __future__ import absolute_import 4 | 5 | from typing import List 6 | 7 | 8 | class FieldNames: 9 | 10 | MODEL_NAME = "model_name" 11 | FILE_NAME = "file_name" 12 | MODEL_PARAMS = "model_params" 13 | STRATEGY_ARGS = "strategy_args" 14 | FIT_TIME = "fit_time" 15 | INFERENCE_TIME = "inference_time" 16 | ACTUAL_DATA = "actual_data" 17 | INFERENCE_DATA = "inference_data" 18 | LOG_INFO = "log_info" 19 | 20 | @classmethod 21 | def all_fields(cls) -> List[str]: 22 | return [ 23 | cls.MODEL_NAME, 24 | cls.FILE_NAME, 25 | cls.MODEL_PARAMS, 26 | cls.STRATEGY_ARGS, 27 | cls.FIT_TIME, 28 | cls.INFERENCE_TIME, 29 | cls.ACTUAL_DATA, 30 | cls.INFERENCE_DATA, 31 | cls.LOG_INFO, 32 | ] 33 | -------------------------------------------------------------------------------- /ts_benchmark/models/__init__.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | 3 | __all__ = [ 4 | "ModelFactory", 5 | "ModelBase", 6 | "get_models", 7 | ] 8 | 9 | from ts_benchmark.models.model_base import ModelBase 10 | from ts_benchmark.models.model_loader import ModelFactory, get_models 11 | -------------------------------------------------------------------------------- /ts_benchmark/report/__init__.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | 3 | from __future__ import absolute_import 4 | 5 | from ts_benchmark.report import report_dash, report_csv 6 | 7 | 8 | def report(report_config: dict, report_method: str = "csv") -> None: 9 | if report_method == "dash": 10 | report_dash.report(report_config) 11 | elif report_method == "csv": 12 | report_csv.report(report_config) 13 | else: 14 | raise ValueError(f"Unknown report method {report_method}") 15 | -------------------------------------------------------------------------------- /ts_benchmark/report/report_dash/__init__.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | 3 | from __future__ import absolute_import 4 | 5 | __all__ = ["report"] 6 | 7 | from ts_benchmark.report.report_dash.app import report 8 | -------------------------------------------------------------------------------- /ts_benchmark/report/report_dash/memory.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | 3 | from __future__ import absolute_import 4 | 5 | 6 | # storing global read-only data, which should only be set during main app initialization 7 | READONLY_MEMORY = {} 8 | -------------------------------------------------------------------------------- /ts_benchmark/report/report_dash/pages/leaderboard.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | 3 | from __future__ import absolute_import 4 | 5 | import dash 6 | from dash import dash_table 7 | 8 | from ts_benchmark.report.report_dash.layout import generate_layout, format_data_columns 9 | from ts_benchmark.report.report_dash.memory import READONLY_MEMORY 10 | 11 | dash.register_page(__name__) 12 | 13 | leaderboard_df = READONLY_MEMORY["leaderboard_df"] 14 | 15 | table = dash_table.DataTable( 16 | id="leaderboard", 17 | columns=format_data_columns(leaderboard_df), 18 | data=leaderboard_df.to_dict("records"), 19 | filter_action="native", 20 | sort_action="native", 21 | sort_mode="multi", 22 | column_selectable="single", 23 | row_selectable="multi", 24 | selected_columns=[], 25 | selected_rows=[], 26 | page_action="native", 27 | page_current=0, 28 | page_size=100, 29 | style_data={ 30 | "whiteSpace": "normal", 31 | "height": "auto", 32 | "lineHeight": "15px", 33 | "maxWidth": "200px", 34 | "wordWrap": "break-word", 35 | }, 36 | ) 37 | 38 | layout = generate_layout([table]) 39 | -------------------------------------------------------------------------------- /ts_benchmark/report/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/decisionintelligence/TFB/28984f821aa4b2ddc7fa5cfd90fba088e9ad6273/ts_benchmark/report/utils/__init__.py -------------------------------------------------------------------------------- /ts_benchmark/utils/__init__.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | -------------------------------------------------------------------------------- /ts_benchmark/utils/design_pattern.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | class Singleton(type): 3 | """ 4 | Used to construct singleton classes through the method of meta classes 5 | """ 6 | 7 | _instance_dict = {} 8 | 9 | def __call__(cls, *args, **kwargs): 10 | if cls not in cls._instance_dict: 11 | cls._instance_dict[cls] = super(Singleton, cls).__call__(*args, **kwargs) 12 | return cls._instance_dict[cls] 13 | -------------------------------------------------------------------------------- /ts_benchmark/utils/get_file_name.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | import os 3 | import socket 4 | import time 5 | 6 | 7 | def get_unique_file_suffix(): 8 | """ 9 | Generate a log file name suffix that includes the following information: 10 | 11 | - Hostname 12 | - The current timestamp, in seconds, is the number of seconds since the Unix era 13 | - PID (process identifier) of the process 14 | 15 | Return: 16 | str: The name of the generated log file, in the format '.timestamp.hostname.pid.csv' 17 | 18 | For example, if the host name is' myhost ', the current timestamp is 1631655702, and the current process ID is 12345 19 | The returned file name may be '.1631655702.myhost.12345.csv'. 20 | """ 21 | # Get Host Name 22 | hostname = socket.gethostname() 23 | 24 | # Get current timestamp (seconds since Unix era) 25 | timestamp = int(time.time()) 26 | 27 | # Obtain the PID (process identifier) of the process 28 | pid = os.getpid() 29 | 30 | # Build file name 31 | log_filename = f".{timestamp}.{hostname}.{pid}.csv" 32 | return log_filename 33 | -------------------------------------------------------------------------------- /ts_benchmark/utils/parallel/base.py: -------------------------------------------------------------------------------- 1 | # -*- coding: utf-8 -*- 2 | 3 | from __future__ import absolute_import 4 | 5 | import abc 6 | from typing import Any, NoReturn 7 | 8 | 9 | class TaskResult(metaclass=abc.ABCMeta): 10 | 11 | @abc.abstractmethod 12 | def result(self) -> Any: 13 | """ 14 | block until result is available 15 | """ 16 | 17 | @abc.abstractmethod 18 | def put(self, value: Any) -> NoReturn: 19 | """ 20 | set value of the result 21 | """ 22 | 23 | 24 | class SharedStorage(metaclass=abc.ABCMeta): 25 | 26 | @abc.abstractmethod 27 | def put(self, name: str, value: Any) -> NoReturn: 28 | """ 29 | store variable into storage 30 | """ 31 | 32 | @abc.abstractmethod 33 | def get(self, name: str, default_value: Any = None) -> Any: 34 | """ 35 | get variable from storage 36 | """ 37 | -------------------------------------------------------------------------------- /ts_benchmark/utils/random_utils.py: -------------------------------------------------------------------------------- 1 | import os 2 | import random 3 | from typing import Optional, NoReturn 4 | 5 | import numpy as np 6 | import torch 7 | 8 | 9 | def fix_random_seed(seed: Optional[int] = 2021) -> NoReturn: 10 | if seed is None: 11 | return 12 | 13 | random.seed(seed) 14 | torch.manual_seed(seed) 15 | np.random.seed(seed) 16 | 17 | 18 | def fix_all_random_seed(seed: Optional[int] = 2021) -> NoReturn: 19 | if seed is None: 20 | return 21 | 22 | torch.manual_seed(seed) 23 | torch.cuda.manual_seed_all(seed) 24 | np.random.seed(seed) 25 | random.seed(seed) 26 | torch.cuda.manual_seed(seed) 27 | torch.backends.cudnn.deterministic = True 28 | torch.backends.cudnn.benchmark = False 29 | torch.backends.cudnn.enabled = False 30 | 31 | os.environ['PYTHONHASHSEED'] = str(1) 32 | --------------------------------------------------------------------------------