Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
140 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

DeepTSF: Codeless machine learning operations for time series forecasting (2308.00709v2)

Published 28 Jul 2023 in cs.LG and cs.SE

Abstract: This paper presents DeepTSF, a comprehensive machine learning operations (MLOps) framework aiming to innovate time series forecasting through workflow automation and codeless modeling. DeepTSF automates key aspects of the ML lifecycle, making it an ideal tool for data scientists and MLops engineers engaged in ML and deep learning (DL)-based forecasting. DeepTSF empowers users with a robust and user-friendly solution, while it is designed to seamlessly integrate with existing data analysis workflows, providing enhanced productivity and compatibility. The framework offers a front-end user interface (UI) suitable for data scientists, as well as other higher-level stakeholders, enabling comprehensive understanding through insightful visualizations and evaluation metrics. DeepTSF also prioritizes security through identity management and access authorization mechanisms. The application of DeepTSF in real-life use cases of the I-NERGY project has already proven DeepTSF's efficacy in DL-based load forecasting, showcasing its significant added value in the electrical power and energy systems domain.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (66)
  1. TensorFlow: learning functions at scale, in: ICFP: 21st International Conference on Functional Programming, Association for Computing Machinery (ACM). pp. 1–1. doi:10.1145/2951913.2976746.
  2. Hands-On InfluxDB. Chapman and Hall/CRC. URL: https://www.taylorfrancis.com/chapters/edit/10.1201/9781315155579-20/hands-influxdb-khaleel-ahmad-masroor-ansari, doi:10.1201/9781315155579-20.
  3. Optuna: A Next-generation Hyperparameter Optimization Framework, in: Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Association for Computing Machinery. pp. 2623–2631. doi:10.1145/3292500.3330701.
  4. GluonTS: Probabilistic Time Series Models in Python. arXiv doi:https://doi.org/10.48550/arXiv.1906.05264.
  5. Beginning MLOps with MLFlow. Apress. doi:10.1007/978-1-4842-6549-9.
  6. Apache, 2023. Apache MXNet | A flexible and efficient library for deep learning. URL: https://mxnet.apache.org/versions/1.9.1/.
  7. Artemij Fedosejev, 2015. React.js Essentials.
  8. The theta model: a decomposition approach to forecasting. International Journal of Forecasting 16, 521–530. doi:10.1016/S0169-2070(00)00066-2.
  9. An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling. arXiv:1803.01271 .
  10. A Model of Inductive Bias Learning. Journal of Artificial Intelligence Research 12, 149–198. URL: https://jair.org/index.php/jair/article/view/10253, doi:10.1613/JAIR.731.
  11. BentoML, 2023. BentoML: Build, Ship, Scale AI Applications. URL: https://www.bentoml.com/.
  12. Algorithms for Hyper-Parameter Optimization. Advances in Neural Information Processing Systems 24. doi:https://dl.acm.org/doi/10.5555/2986459.2986743.
  13. Time Series Analysis: Forecasting and Control.
  14. Random forests. Machine Learning 45, 5–32. URL: https://link.springer.com/article/10.1023/A:1010933404324, doi:10.1023/A:1010933404324.
  15. Securing Web Services with Keycloak. Cloud Native Integration with Apache Camel , 77–115doi:https://doi.org/10.1007/978-1-4842-7211-4{_}3.
  16. N-HiTS: Neural Hierarchical Interpolation for Time Series Forecasting. arXiv doi:https://doi.org/10.48550/arXiv.2201.12886.
  17. Click, 2023. Click documentation. URL: https://click.palletsprojects.com/en/8.1.x/.
  18. Dagster, 2023. Dagster | Cloud-native orchestration of data pipelines. URL: https://dagster.io/.
  19. Darts, 2023. Darts documentation | Forecasting models. URL: https://unit8co.github.io/darts/generated_api/darts.models.forecasting.html.
  20. DeepTSF, 2023. DeepTSF Documentation. URL: https://github.com/I-NERGY/DeepTSF/blob/master/README.md.
  21. Docker, 2023. Docker: Accelerated, Containerized Application Development. URL: https://www.docker.com/.
  22. The Web SSO Standard OpenID Connect: In-depth Formal Security Analysis and Security Guidelines. Proceedings - IEEE Computer Security Foundations Symposium , 189–202doi:10.1109/CSF.2017.20.
  23. Exponential smoothing: The state of the art. Journal of Forecasting 4, 1–28. doi:10.1002/FOR.3980040103.
  24. GraphQL, 2023. GraphQL | A query language for your API. URL: https://graphql.org/.
  25. Darts: User-Friendly Modern Machine Learning for Time Series. Journal of Machine Learning Research 23, 1–6. doi:10.48550/arxiv.2110.03224.
  26. Long Short-Term Memory. Neural Computation doi:10.1162/neco.1997.9.8.1735.
  27. An Efficient Approach for Assessing Hyperparameter Importance, in: Xing, E.P., Jebara, T. (Eds.), Proceedings of the 31st International Conference on Machine Learning, PMLR, Bejing, China. pp. 754–762.
  28. Another look at measures of forecast accuracy. International Journal of Forecasting 22, 679–688. doi:10.1016/j.ijforecast.2006.03.001.
  29. Artificial Intelligence for Next Generation Energy Services Across Europe - The I-NERGY Project, in: ES 2021 : 19th International Conference e-Society 2021, Lisbon. pp. 61–68.
  30. LightGBM: A Highly Efficient Gradient Boosting Decision Tree, in: Advances in Neural Information Processing Systems 30. URL: https://github.com/Microsoft/LightGBM.
  31. ImageNet classification with deep convolutional neural networks. Communications of the ACM 60, 84–90. doi:10.1145/3065386.
  32. MLflow tracking server implementation. URL: https://github.com/I-NERGY/mlflow-tracking-server-nginx.
  33. Getting Started with FastAPI. High-Performance Web Apps with FastAPI , 29–64doi:10.1007/978-1-4842-9178-8{_}2.
  34. Deep Reinforcement Learning: An Overview. arXiV doi:arXiv:1701.07274.
  35. Temporal Fusion Transformers for Interpretable Multi-horizon Time Series Forecasting. International Journal of Forecasting 37, 1748–1764. URL: https://arxiv.org/abs/1912.09363v3, doi:10.1016/j.ijforecast.2021.03.012.
  36. A Unified Approach to Interpreting Model Predictions, in: Guyon, I., Luxburg, U.V., Bengio, S., Wallach, H., Fergus, R., Vishwanathan, S., Garnett, R. (Eds.), Advances in Neural Information Processing Systems, Curran Associates, Inc. URL: https://proceedings.neurips.cc/paper_files/paper/2017/file/8a20a8621978632d76c43dfd28b67767-Paper.pdf.
  37. Machine learning advances for time series forecasting. Journal of Economic Surveys 37, 76–111. doi:10.1111/JOES.12429.
  38. MinIO, 2022. MinIO | High Performance, Kubernetes Native Object Storage. URL: https://min.io/.
  39. MLflow, 2023a. Command-Line Interface — MLflow 1.25.1 documentation. URL: https://mlflow.org/docs/1.25.1/cli.html.
  40. MLflow, 2023b. MLflow documentation | Model Registry. URL: https://mlflow.org/docs/latest/model-registry.html.
  41. MongoDB, 2023. MongoDB: The Developer Data Platform | MongoDB. URL: https://www.mongodb.com/.
  42. Trend analysis of climate time series: A review of methods. Earth-Science Reviews 190, 310–322. doi:10.1016/J.EARSCIREV.2018.12.005.
  43. ONNX, 2023. ONNX | Home. URL: https://onnx.ai/.
  44. OpenResty, 2023. OpenResty | A dynamic web platform based on NGINX and LuaJIT. URL: https://openresty.org/en/.
  45. N-BEATS: Neural basis expansion analysis for interpretable time series forecasting. arXiv doi:https://doi.org/10.48550/arXiv.1905.10437.
  46. PyTorch: An Imperative Style, High-Performance Deep Learning Library. Advances in Neural Information Processing Systems 32.
  47. Scikit-learn: Machine Learning in Python. Journal of Machine Learning Research 12, 2825–2830.
  48. In Search of Deep Learning Architectures for Load Forecasting: A Comparative Analysis and the Impact of the Covid-19 Pandemic on Model Performance, in: 2022 13th International Conference on Information, Intelligence, Systems and Applications (IISA), IEEE. pp. 1–8. doi:10.1109/IISA56318.2022.9904363.
  49. A comparative assessment of deep learning models for day-ahead load forecasting: Investigating key accuracy drivers. arXiv:2302.12168 doi:10.48550/arxiv.2302.12168.
  50. Handling bad or missing smart meter data through advanced data imputation, in: 2016 IEEE Power & Energy Society Innovative Smart Grid Technologies Conference (ISGT), IEEE. pp. 1–5. doi:10.1109/ISGT.2016.7781213.
  51. PostgreSQL, 2022. PostgreSQL | The world’s most advanced open source database. URL: https://www.postgresql.org/.
  52. Prophet, 2023. Prophet | Forecasting at scale. URL: https://facebook.github.io/prophet/.
  53. PyTorch, 2023. PyTorch Forecasting. URL: https://pytorch-forecasting.readthedocs.io/en/stable/.
  54. Nginx: the high-performance web server and reverse proxy. Linux Journal URL: https://dl.acm.org/doi/10.5555/1412202.1412204, doi:10.5555/1412202.1412204.
  55. Seldon, 2023. Seldon Core - OSS Model Deployment. URL: https://www.seldon.io/solutions/open-source-projects/core.
  56. Financial time series forecasting with deep learning : A systematic literature review: 2005–2019. Applied Soft Computing 90, 106181. doi:10.1016/J.ASOC.2020.106181.
  57. SHAP, . SHAP documentation | Beeswarm plot. URL: https://shap.readthedocs.io/en/latest/example_notebooks/api_examples/plots/beeswarm.html.
  58. Shap, 2023. SHAP documentation. URL: https://shap.readthedocs.io/en/latest/.
  59. Product sales probabilistic forecasting: An empirical evaluation using the M5 competition data. International Journal of Production Economics 240, 108237. doi:10.1016/J.IJPE.2021.108237.
  60. Editorial: Special issue on time series analysis in the biological sciences. Journal of Time Series Analysis 33, 701–703. doi:10.1111/J.1467-9892.2012.00805.X.
  61. TensorFlow, 2023. TensorFlow Probability. URL: https://www.tensorflow.org/probability.
  62. Timescale, 2023. Timescale is PostgreSQL++ for time series and event data | Timescale. URL: https://www.timescale.com/.
  63. High-performance medicine: the convergence of human and artificial intelligence. Nature Medicine 2019 25:1 25, 44–56. doi:10.1038/s41591-018-0300-7.
  64. Attention Is All You Need, in: Advances in Neural Information Processing Systems, Neural information processing systems foundation. pp. 5999–6009.
  65. YAML, 2023. YAML Ain’t Markup Language (YAML™) revision 1.2.2. URL: https://yaml.org/spec/1.2.2/.
  66. Recent trends in deep learning based natural language processing. IEEE Computational Intelligence Magazine 13, 55–75. doi:10.1109/MCI.2018.2840738.
Citations (7)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now