├── .gitignore ├── CODE_OF_CONDUCT.md ├── CONTRIBUTING.md ├── LICENSE ├── Makefile ├── README.md ├── cfn-templates └── sagemaker-domain.yaml ├── img ├── canvas-add-to-model-registry.png ├── canvas-analyze-quick-build.png ├── canvas-analyze-standard-build.png ├── canvas-build-comparison.png ├── canvas-configure-model-config.png ├── canvas-create-data-flow.png ├── canvas-create-model.png ├── canvas-datasets.png ├── canvas-download-performance-report.png ├── canvas-endpoint-resource-error.png ├── canvas-model-deploy.png ├── canvas-model-leaderboard.png ├── canvas-model-new-version.png ├── canvas-model-registry-details.png ├── canvas-predictions.png ├── canvas-start-quick-build.png ├── chronos-main-figure.png ├── chronos.png ├── git-repo-qr-code.png ├── quicksight_filter_item_101_store_001.png ├── smd-select.png └── workshop-qr-code.svg ├── notebooks ├── additional │ ├── gluonts_pipeline │ │ ├── inference.py │ │ ├── preprocess.py │ │ ├── register.py │ │ ├── train.py │ │ └── train_step.py │ ├── lab1a_gluonts.ipynb │ └── lab2a_tide.ipynb ├── lab1_sagemaker_canvas.ipynb ├── lab2_sagemaker_autopilot_api.ipynb ├── lab3_sagemaker_deepar.ipynb ├── lab4_chronos.ipynb ├── lab5_autogluon.ipynb └── lab6_results.ipynb ├── scripts └── lcc-script.sh └── test └── model-performance ├── autogluon-2H-2-17533-20240919-145927.csv ├── autogluon-2h-370-17533-20240927-145601.csv ├── autogluon-Chronos[base]-2h-370-17533-20240930-114821.csv ├── autogluon-Chronos[base]-2h-370-17533-bt4-20240930-122110.csv ├── autogluon-Chronos[mini]-2H-2-17533-20240919-181432.csv ├── autogluon-Chronos[mini]-2H-2-17533-20240919-182135.csv ├── autogluon-Chronos[mini]-2H-2-17533-bt4-20240919-182239.csv ├── autopilot-2H-370-17533-20240919-151404.csv ├── autopilot-full-2h-370-17533-20241001-091815.csv ├── canvas-1D-2-1462-20240927-150101.csv ├── canvas-1D-2-35065-20240920-091227.csv ├── canvas-1D-370-35065-20240920-091316.csv ├── canvas-1D-370-35065-20240927-105539.csv ├── canvas-1H-2-1462-20240927-191325.csv ├── canvas-1H-370-35065-20240920-091430.csv ├── canvas-1H-370-35065-20240927-134914.csv ├── chronos-2H-370-17533-bt4-20240919-130527.csv ├── chronos-2H-370-17533-off10-20240919-130527.csv ├── deepar-2H-370-17533-20240919-202341.csv ├── deepar-2h-2-17533-20240927-105431.csv ├── gluonts-1h-10-8736-bt4-20241106-080455.csv ├── gluonts-1h-10-8736-bt4-20241106-181402.csv └── gluonts-1h-10-8736-bt4-20241107-214657.csv /.gitignore: -------------------------------------------------------------------------------- 1 | # Byte-compiled / optimized / DLL files 2 | __pycache__/ 3 | *.py[cod] 4 | *$py.class 5 | 6 | # C extensions 7 | *.so 8 | 9 | # Distribution / packaging 10 | .Python 11 | build/ 12 | develop-eggs/ 13 | dist/ 14 | downloads/ 15 | eggs/ 16 | .eggs/ 17 | lib/ 18 | lib64/ 19 | parts/ 20 | sdist/ 21 | var/ 22 | wheels/ 23 | pip-wheel-metadata/ 24 | share/python-wheels/ 25 | *.egg-info/ 26 | .installed.cfg 27 | *.egg 28 | MANIFEST 29 | 30 | # PyInstaller 31 | # Usually these files are written by a python script from a template 32 | # before PyInstaller builds the exe, so as to inject date/other infos into it. 33 | *.manifest 34 | *.spec 35 | 36 | # Installer logs 37 | pip-log.txt 38 | pip-delete-this-directory.txt 39 | 40 | # Unit test / coverage reports 41 | htmlcov/ 42 | .tox/ 43 | .nox/ 44 | .coverage 45 | .coverage.* 46 | .cache 47 | nosetests.xml 48 | coverage.xml 49 | *.cover 50 | *.py,cover 51 | .hypothesis/ 52 | .pytest_cache/ 53 | 54 | # Translations 55 | *.mo 56 | *.pot 57 | 58 | # Django stuff: 59 | *.log 60 | local_settings.py 61 | db.sqlite3 62 | db.sqlite3-journal 63 | 64 | # Flask stuff: 65 | instance/ 66 | .webassets-cache 67 | 68 | # Scrapy stuff: 69 | .scrapy 70 | 71 | # Sphinx documentation 72 | docs/_build/ 73 | 74 | # PyBuilder 75 | target/ 76 | 77 | # Jupyter Notebook 78 | .ipynb_checkpoints 79 | 80 | # IPython 81 | profile_default/ 82 | ipython_config.py 83 | 84 | # pyenv 85 | .python-version 86 | 87 | # pipenv 88 | # According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control. 89 | # However, in case of collaboration, if having platform-specific dependencies or dependencies 90 | # having no cross-platform support, pipenv may install dependencies that don't work, or not 91 | # install all needed dependencies. 92 | #Pipfile.lock 93 | 94 | # PEP 582; used by e.g. github.com/David-OConnor/pyflow 95 | __pypackages__/ 96 | 97 | # Celery stuff 98 | celerybeat-schedule 99 | celerybeat.pid 100 | 101 | # SageMath parsed files 102 | *.sage.py 103 | 104 | # Environments 105 | .env 106 | .venv 107 | env/ 108 | venv/ 109 | ENV/ 110 | env.bak/ 111 | venv.bak/ 112 | 113 | # Spyder project settings 114 | .spyderproject 115 | .spyproject 116 | 117 | # Rope project settings 118 | .ropeproject 119 | 120 | # mkdocs documentation 121 | /site 122 | 123 | # mypy 124 | .mypy_cache/ 125 | .dmypy.json 126 | dmypy.json 127 | 128 | # Pyre type checker 129 | .pyre/ 130 | 131 | # Project-specific 132 | deploy 133 | 134 | # Build folder 135 | 136 | */build/* 137 | 138 | # SAM 139 | .aws-sam/* 140 | samconfig.* 141 | .DS_Store 142 | .environment 143 | .not-used-snippets 144 | 145 | .gp2/* 146 | *.pdf 147 | *snapshot.json 148 | .test* 149 | *.zip -------------------------------------------------------------------------------- /CODE_OF_CONDUCT.md: -------------------------------------------------------------------------------- 1 | ## Code of Conduct 2 | This project has adopted the [Amazon Open Source Code of Conduct](https://aws.github.io/code-of-conduct). 3 | For more information see the [Code of Conduct FAQ](https://aws.github.io/code-of-conduct-faq) or contact 4 | opensource-codeofconduct@amazon.com with any additional questions or comments. -------------------------------------------------------------------------------- /CONTRIBUTING.md: -------------------------------------------------------------------------------- 1 | # Contributing Guidelines 2 | 3 | Thank you for your interest in contributing to our project. Whether it's a bug report, new feature, correction, or additional 4 | documentation, we greatly value feedback and contributions from our community. 5 | 6 | Please read through this document before submitting any issues or pull requests to ensure we have all the necessary 7 | information to effectively respond to your bug report or contribution. 8 | 9 | 10 | ## Reporting Bugs/Feature Requests 11 | 12 | We welcome you to use the GitHub issue tracker to report bugs or suggest features. 13 | 14 | When filing an issue, please check existing open, or recently closed, issues to make sure somebody else hasn't already 15 | reported the issue. Please try to include as much information as you can. Details like these are incredibly useful: 16 | 17 | * A reproducible test case or series of steps 18 | * The version of our code being used 19 | * Any modifications you've made relevant to the bug 20 | * Anything unusual about your environment or deployment 21 | 22 | 23 | ## Contributing via Pull Requests 24 | Contributions via pull requests are much appreciated. Before sending us a pull request, please ensure that: 25 | 26 | 1. You are working against the latest source on the *main* branch. 27 | 2. You check existing open, and recently merged, pull requests to make sure someone else hasn't addressed the problem already. 28 | 3. You open an issue to discuss any significant work - we would hate for your time to be wasted. 29 | 30 | To send us a pull request, please: 31 | 32 | 1. Fork the repository. 33 | 2. Modify the source; please focus on the specific change you are contributing. If you also reformat all the code, it will be hard for us to focus on your change. 34 | 3. Ensure local tests pass. 35 | 4. Commit to your fork using clear commit messages. 36 | 5. Send us a pull request, answering any default questions in the pull request interface. 37 | 6. Pay attention to any automated CI failures reported in the pull request, and stay involved in the conversation. 38 | 39 | GitHub provides additional document on [forking a repository](https://help.github.com/articles/fork-a-repo/) and 40 | [creating a pull request](https://help.github.com/articles/creating-a-pull-request/). 41 | 42 | 43 | ## Finding contributions to work on 44 | Looking at the existing issues is a great way to find something to contribute on. As our projects, by default, use the default GitHub issue labels (enhancement/bug/duplicate/help wanted/invalid/question/wontfix), looking at any 'help wanted' issues is a great place to start. 45 | 46 | 47 | ## Code of Conduct 48 | This project has adopted the [Amazon Open Source Code of Conduct](https://aws.github.io/code-of-conduct). 49 | For more information see the [Code of Conduct FAQ](https://aws.github.io/code-of-conduct-faq) or contact 50 | opensource-codeofconduct@amazon.com with any additional questions or comments. 51 | 52 | 53 | ## Security issue notifications 54 | If you discover a potential security issue in this project we ask that you notify AWS/Amazon Security via our [vulnerability reporting page](http://aws.amazon.com/security/vulnerability-reporting/). Please do **not** create a public github issue. 55 | 56 | 57 | ## Licensing 58 | 59 | See the [LICENSE](LICENSE) file for our project's licensing. We will ask you to confirm the licensing of your contribution. -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- 1 | MIT No Attribution 2 | 3 | Copyright Amazon.com, Inc. or its affiliates. All Rights Reserved. 4 | 5 | Permission is hereby granted, free of charge, to any person obtaining a copy of 6 | this software and associated documentation files (the "Software"), to deal in 7 | the Software without restriction, including without limitation the rights to 8 | use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of 9 | the Software, and to permit persons to whom the Software is furnished to do so. 10 | 11 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 12 | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS 13 | FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR 14 | COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER 15 | IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN 16 | CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. -------------------------------------------------------------------------------- /Makefile: -------------------------------------------------------------------------------- 1 | deploy-domain: 2 | aws cloudformation deploy \ 3 | --template-file cfn-templates/sagemaker-domain.yaml \ 4 | --stack-name sm-domain-ts-workshop \ 5 | --capabilities CAPABILITY_IAM CAPABILITY_NAMED_IAM \ 6 | --parameter-overrides \ 7 | DomainNamePrefix='sm-domain-time-series' 8 | 9 | destroy-domain: 10 | aws cloudformation delete-stack \ 11 | --stack-name sm-domain-ts-workshop && \ 12 | aws cloudformation wait stack-delete-complete \ 13 | --stack-name sm-domain-ts-workshop 14 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # Modern Time Series Forecasting on AWS 2 | 3 | ## Overview 4 | This workshop demonstrates how to use AWS services to implement time series forecasting. It covers the following examples and AWS services: 5 | 1. Amazon SageMaker Canvas 6 | 2. Amazon SageMaker Autopilot API 7 | 3. Amazon SageMaker DeepAR 8 | 4. Chronos 9 | 5. AutoGluon 10 | 11 | Additional notebooks cover forecasting with GluonTS, a custom algorithm on SageMaker, and Amazon QuickSight. 12 | 13 | The [workshop](https://catalog.us-east-1.prod.workshops.aws/workshops/76a419ba-6303-4e7c-ac02-47112ed7cb3f/en-US) is available in AWS workshop catalog. You can run this workshop on an AWS-led event or in your own AWS account. 14 | 15 | ## How to use this workshop 16 | To use this workshop, you need an Amazon SageMaker domain. All workshop content is in Jupyter notebooks running on Amazon SageMaker. To get started, follow the instructions in the **Getting started** section. To clean up resources, follow the instructions in the **Clean-up** section. You can execute the notebooks in any order, and you don't need to switch between the notebooks and the workshop web page. 17 | 18 | ## Required resources 19 | 20 | **Ignore this section if you're using an AWS-provided account as a part of an AWS-led workshop.** 21 | 22 | In order to be able to run notebooks and complete workshop labs you need access to the following resources in your AWS account. You can check quotas for all following resources in AWS console in [Service Quotas](https://us-east-1.console.aws.amazon.com/servicequotas/home/services/sagemaker/quotas) console. 23 | 24 | **Studio JupyterLab app** 25 | Minimal required instance type is `ml.m5.2xlarge`. We recommend to use `ml.m5.4xlarge` as an instance to run all notebooks. If you have access to GPU-instances like `ml.g5.4xlarge` or `ml.g6.4xlarge`, use these instance to run the notebooks. 26 | 27 | To experiment with the full dataset with 370 time series in the lab 5 AutoGluon you need a GPU instance for the notebook - `ml.g5.4xlarge`/`ml.g6.4xlarge` or `ml.g5.8xlarge`/`ml.g6.8xlarge`. 28 | 29 | - Check quota for [`ml.m5.2xlarge`](https://us-east-1.console.aws.amazon.com/servicequotas/home/services/sagemaker/quotas/L-7C9662F1) 30 | - Check quota for [`ml.m5.4xlarge`](https://us-east-1.console.aws.amazon.com/servicequotas/home/services/sagemaker/quotas/L-2CA31BFA) 31 | - Check quota for [`ml.g5.4xlarge`](https://us-east-1.console.aws.amazon.com/servicequotas/home/services/sagemaker/quotas/L-81940D85) 32 | - Check quota for [`ml.g6.4xlarge`](https://us-east-1.console.aws.amazon.com/servicequotas/home/services/sagemaker/quotas/L-692B8304) 33 | - Check quota for [`ml.g5.8xlarge`](https://us-east-1.console.aws.amazon.com/servicequotas/home/services/sagemaker/quotas/L-19B6BAFC) 34 | - Check quota for [`ml.g6.8xlarge`](https://us-east-1.console.aws.amazon.com/servicequotas/home/services/sagemaker/quotas/L-804C2AFF) 35 | 36 | 37 | **Number of concurrent AutoML Jobs** 38 | To follow the optimal flow of the workshop, you need to run at least three AutoML jobs in parallel. We recommend to have a quota set to six or more concurrent jobs. 39 | 40 | - Check quota for [maximum number of concurrent AutoML Jobs](https://us-east-1.console.aws.amazon.com/servicequotas/home/services/sagemaker/quotas/L-CFC2D5B6) 41 | 42 | **Training jobs** 43 | To run a training job for DeepAR algorithm you need a `ml.c5.4xlarge` compute instance 44 | 45 | - Check quota for [`ml.c5.4xlarge`](https://us-east-1.console.aws.amazon.com/servicequotas/home/services/sagemaker/quotas/L-E7898792) 46 | 47 | **SageMaker real-time inference endpoints** 48 | DeepAR, Chronos, and AutoGluon notebooks deploy SageMaker real-time inference endpoints to test models. You need access to the following compute instances for endpoint use: 49 | - Minimal for Autopilot and DeepAR endpoints: check [`ml.m5.xlarge`](https://us-east-1.console.aws.amazon.com/servicequotas/home/services/sagemaker/quotas/L-2F737F8D) 50 | - Recommended for Autopilot and DeepAR endpoints: check [`ml.m5.4xlarge`](https://us-east-1.console.aws.amazon.com/servicequotas/home/services/sagemaker/quotas/L-E2649D46) 51 | - Minimal for Chronos Small endpoint: check [`ml.g5.xlarge`](https://us-east-1.console.aws.amazon.com/servicequotas/home/services/sagemaker/quotas/L-1928E07B) 52 | - Optional for Chronos Base: check [`ml.g5.2xlarge`](https://us-east-1.console.aws.amazon.com/servicequotas/home/services/sagemaker/quotas/L-9614C779) 53 | - Optional for Chronos Large: check [`ml.g5.4xlarge`](https://us-east-1.console.aws.amazon.com/servicequotas/home/services/sagemaker/quotas/L-C1B9A48D) 54 | 55 | ## Workshop flow 56 | The notebooks from Lab 1 to Lab 5 are self-sufficient. You can run them in any order. If you're unfamiliar with time series forecasting, we recommend starting with the Lab 1 notebook and continuing from there. Alternatively, you can run only the notebooks that interest you, such as `lab4_chronos` or `lab5_autogluon`. 57 | 58 | The model training in Labs 1, 2, and 3 takes 15-40 minutes, depending on the algorithm. You don't need to wait for the training to complete before moving on to the next notebook. You can come back to the previous notebook once the training is done. 59 | 60 | Executing all five notebooks will take 2-3 hours. If you're new to time series forecasting, Jupyter notebooks, or Python, it may take longer. 61 | 62 | ## Workshop costs 63 | The notebooks in this workshop create cost-generating resources in your account. Make sure you always delete created SageMaker inference endpoints, log out of Canvas, and stop JupyterLab spaces if you don't use them. 64 | 65 | If running all notebooks with all sections, including optional sections and training three models using **Standard** builds in Canvas, the estimated cost is approximately 90-100 USD. 66 | 67 | Please note that your actual costs may vary depending on the duration of the workshop, the number of inference endpoints created, and the time the endpoints remain in service. 68 | 69 | To optimize costs, follow these recommendations: 70 | 1. Run only **Quick** builds in Canvas to minimize costs. Note that in this case you cannot download model performance JSON files 71 | 2. Use only a sample from the full dataset to train models and run all experiments. Each notebook contains code to create a small dataset with a sample from the time series 72 | 3. Promptly delete SageMaker inference endpoints after use 73 | 4. Use `ml.m5.xlarge` instance for JupyterLab app to balance performance and cost 74 | 5. Limit Chronos experiments to one endpoint and a sample of the time series in the notebook `lab4_chronos` 75 | 76 | ## Getting started 77 | If you'd lke to create a new domain, you have two options: 78 | 1. Use the provided AWS CloudFormation [template](./cfn-templates/sagemaker-domain.yaml) that creates a SageMaker domain, a user profile, and adds the IAM roles required for executing the provided notebooks - this is the recommended approach 79 | 1. Follow the onboarding [instructions](https://docs.aws.amazon.com/sagemaker/latest/dg/gs-studio-onboard.html) in the Developer Guide and create a new domain and a user profile via AWS Console 80 | 81 | ## Datasets 82 | 83 | All examples and notebooks in this workshop using the same real-world dataset. It makes possible to compare performance and model metrics across different approaches. 84 | 85 | You use the [electricity dataset](https://archive.ics.uci.edu/ml/datasets/ElectricityLoadDiagrams20112014) from the repository of the University of California, Irvine: 86 | > Trindade, Artur. (2015). ElectricityLoadDiagrams20112014. UCI Machine Learning Repository. https://doi.org/10.24432/C58C86. 87 | 88 | 89 | ## Example 1: Amazon SageMaker Canvas 90 | 91 | Open the [lab 1 notebook](./notebooks/lab1_sagemaker_canvas.ipynb) and follow the instructions. 92 | 93 | Additional SageMaker Canvas links: 94 | - [Time Series Forecasts in Amazon SageMaker Canvas](https://docs.aws.amazon.com/sagemaker/latest/dg/canvas-time-series.html) 95 | - [Canvas Workshop - time series forecast lab](https://catalog.workshops.aws/canvas-immersion-day/en-US/1-use-cases/3-retail) 96 | - [Time-Series Forecasting Using Amazon SageMaker Canvas](https://catalog.us-east-1.prod.workshops.aws/workshops/866925a4-cb5f-4a3d-9cd7-80edc0aa5f0c/en-US/4-0sagemakercanvas) 97 | 98 | 99 | ## Example 2: Amazon SageMaker Autopilot API 100 | 101 | Open the [lab 2 notebook](./notebooks/lab2_sagemaker_autopilot_api.ipynb) and follow the instructions. 102 | 103 | Note: previous Autopilot UX in Studio Classic merged with Canvas as of re:Invent 2023. All AutoML functionality is moved to Canvas as of now. 104 | 105 | Additional SageMaker Autopilot API links: 106 | - [Amazon SageMaker Autopilot](https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-automate-model-development.html) 107 | - [Time series forecasting algorithms in SageMaker](https://docs.aws.amazon.com/sagemaker/latest/dg/timeseries-forecasting-algorithms.html) 108 | - Example notebook [Time series Forecasting with Amazon SageMaker Autopilot](https://github.com/aws/amazon-sagemaker-examples/blob/main/autopilot/autopilot_time_series.ipynb) 109 | - [Lab 2 - Demand Forecasting with SageMaker Autopilot API](https://catalog.us-east-1.prod.workshops.aws/workshops/caef4710-3721-4957-a2ce-33799920ef72/en-US/40-sagemakerautopilot) 110 | - [Time series forecasting with Amazon SageMaker AutoML](https://aws.amazon.com/it/blogs/machine-learning/time-series-forecasting-with-amazon-sagemaker-automl/) 111 | 112 | ## Example 3: Amazon SageMaker DeepAR 113 | 114 | Open the [lab 3 notebook](./notebooks/lab3_sagemaker_deepar.ipynb) and follow the instructions. 115 | 116 | Additional DeepAR links: 117 | - [Use the SageMaker DeepAR forecasting algorithm](https://docs.aws.amazon.com/sagemaker/latest/dg/deepar.html) 118 | - [Deep AR Forecasting](https://sagemaker.readthedocs.io/en/stable/algorithms/time_series/deep_ar.html) 119 | - [Example notebook](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/deepar_electricity/DeepAR-Electricity.ipynb) 120 | - [Deep Demand Forecasting with Amazon SageMaker notebook](https://github.com/awslabs/sagemaker-deep-demand-forecast/blob/mainline/src/deep-demand-forecast.ipynb) 121 | - [DeepAR: Probabilistic Forecasting with Autoregressive Recurrent Networks](https://arxiv.org/abs/1704.04110) 122 | - [Predictive Analytics with Time-series Machine Learning on Amazon Timestream](https://aws.amazon.com/blogs/database/predictive-analytics-with-time-series-machine-learning-on-amazon-timestream/) 123 | - [Bike-Share Demand Forecasting 2b: SageMaker DeepAR Algorithm](https://github.com/aws-samples/time-series-forecasting-on-aws/blob/main/2b_SageMaker_Built-In_DeepAR.ipynb) 124 | 125 | ## Example 4: Chronos 126 | 127 | Open the [lab 4 notebook](./notebooks/lab4_chronos.ipynb) and follow the instructions. 128 | 129 | Links to more Chronos content: 130 | - [Chronos models on Huggingface](https://huggingface.co/amazon/chronos-t5-large) 131 | - [Chronos GitHub](https://github.com/amazon-science/chronos-forecasting) 132 | - [Lot of Chronos-related content on Chronos GitHub](https://github.com/amazon-science/chronos-forecasting?tab=readme-ov-file#-coverage) 133 | - [Chronos: Learning the Language of Time Series](https://arxiv.org/html/2403.07815v1) 134 | - [Adapting language model architectures for time series forecasting](https://www.amazon.science/blog/adapting-language-model-architectures-for-time-series-forecasting) 135 | - [Evaluating Chronos models](https://github.com/amazon-science/chronos-forecasting/blob/main/scripts/README.md#evaluating-chronos-models) 136 | - [Chronos-related content on Chronos GitHub](https://github.com/amazon-science/chronos-forecasting?tab=readme-ov-file#-coverage) 137 | - [Fast and accurate zero-shot forecasting with Chronos-Bolt and AutoGluon](https://aws.amazon.com/blogs/machine-learning/fast-and-accurate-zero-shot-forecasting-with-chronos-bolt-and-autogluon/) 138 | 139 | 140 | ## Example 5: AutoGluon 141 | 142 | Open the [lab 5 notebook](./notebooks/lab5_autogluon.ipynb) and follow the instructions. 143 | 144 | Links to AutoGluon content: 145 | - AutoGluon time series 146 | - [AutoGluon time series forecasting](https://auto.gluon.ai/stable/tutorials/timeseries/index.html) 147 | - AutoGluon Chronos 148 | - [AutoGluon forecasting with Chronos](https://auto.gluon.ai/stable/tutorials/timeseries/forecasting-chronos.html) 149 | - [Forecasting with Chronos notebook Colab](https://colab.research.google.com/github/autogluon/autogluon/blob/stable/docs/tutorials/timeseries/forecasting-chronos.ipynb) 150 | - [AutoGluon Cloud](https://auto.gluon.ai/cloud/dev/tutorials/autogluon-cloud.html) 151 | - [AutoGluon Assistant](https://github.com/autogluon/autogluon-assistant) 152 | 153 | ## Additional examples 154 | The additional notebooks in the folder `notebooks/additional` cover more approaches you can use for time series forecasting. These notebooks demonstrate: 155 | 1. GluonTS 156 | 2. Custom algorithms on SageMaker 157 | 3. Amazon QuickSight forecast 158 | 159 | ### Example 1A: GluonTS 160 | 161 | Navigate to the `additional` folder inside the `notebooks` folder. Open the [lab 1A notebook](./notebooks/additional/lab1a_gluonts.ipynb) and follow the instructions. 162 | 163 | The notebook `additional\lab1a_gluonts` also contains an end-to-end example of productization of a time series forecasting workflow. The lab demonstrates how to create a reproducible SageMaker pipeline with data processing, model training, model evaluation, model registration in the model registry, and model deployment to a SageMaker endpoint. The notebook uses [GluonTS implementation](https://github.com/awslabs/gluonts/blob/dev/src/gluonts/torch/model/tft/estimator.py) of Temporal Fusion Transformer forecast and SageMaker Python SDK PyTorch framework together with SageMaker built-in [Deep Learning Containers (DLC)](https://github.com/aws/deep-learning-containers). 164 | 165 | Links to GluonTS content: 166 | - [GluonTS: Probabilistic and Neural Time Series Modeling in Python](https://www.jmlr.org/papers/volume21/19-820/19-820.pdf): paper 167 | - [GluonTS - Probabilistic Time Series Modeling in Python](https://github.com/awslabs/gluonts): GitHub repository 168 | - [Creating neural time series models with Gluon Time Series](https://aws.amazon.com/blogs/machine-learning/creating-neural-time-series-models-with-gluon-time-series/) 169 | - [Deep demand forecast with Amazon SageMaker](https://github.com/awslabs/sagemaker-deep-demand-forecast) 170 | 171 | ### Example 2A: Amazon SageMaker custom algorithm 172 | 173 | This example is under development. 174 | 175 | Refer to the following resources to see how you can run custom algorithms on SageMaker: 176 | - [Robust time series forecasting with MLOps on Amazon SageMaker](https://aws.amazon.com/blogs/machine-learning/robust-time-series-forecasting-with-mlops-on-amazon-sagemaker/) 177 | - [Deep demand forecast with Amazon SageMaker](https://github.com/awslabs/sagemaker-deep-demand-forecast) 178 | - [Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting (AAAI'21 Best Paper)](https://github.com/aws-samples/time-series-forecasting-on-aws/blob/main/3_SagaMaker_Custom_algorithm_Informer.ipynb) 179 | - [TiDE](https://arxiv.org/pdf/2304.08424.pdf) 180 | 181 | 182 | ### Example 3A: Amazon QuickSight forecast 183 | [Amazon QuickSight](https://docs.aws.amazon.com/quicksight/latest/user/welcome.html) has ML features to give you hidden insights and trends in your data. One of these ML features is **ML-powered forecast**. The built-in ML forecast uses [Random Cut Forest (RCF) algorithm](https://docs.aws.amazon.com/quicksight/latest/user/concept-of-ml-algorithms.html) to detect seasonality, trends, exclude outliers, and impute missing values. For more details on how QuickSight uses RCF to generate forecasts, see the [developer guide](https://docs.aws.amazon.com/quicksight/latest/user/how-does-rcf-generate-forecasts.html). 184 | 185 | ![](img/quicksight_filter_item_101_store_001.png) 186 | 187 | You can customize multiple settings on the **Forecast properties** pane, such as number of forecast periods, prediction interval, seasonality, and forecast boundaries. 188 | 189 | For more details refer to the Developer Guide [Forecasting and creating what-if scenarios with Amazon QuickSight](https://docs.aws.amazon.com/quicksight/latest/user/forecasts-and-whatifs.html). 190 | 191 | Besides a graphical forecasting, you can also add a forecast as a narrative in an insight widget. To learn more, see [Creating autonarratives with Amazon QuickSight](https://docs.aws.amazon.com/quicksight/latest/user/narratives-creating.html). 192 | 193 | Additional resources for Amazon QuickSight forecasting: 194 | - [ML-powered forecasting](https://docs.aws.amazon.com/quicksight/latest/user/forecast-function.html) 195 | 196 | ## Results and comparison 197 | 198 | Open the [lab 6 notebook](./notebooks/lab_final_results.ipnyb) and follow the instructions. 199 | 200 | Additional resources about time series forecast accuracy evaluation 201 | - [Evaluating Predictor Accuracy](https://docs.aws.amazon.com/forecast/latest/dg/metrics.html) 202 | - [Evaluating Chronos models](https://github.com/amazon-science/chronos-forecasting/tree/main/scripts#evaluating-chronos-models) 203 | - [Forecasting time series - evaluation metrics](https://auto.gluon.ai/stable/tutorials/timeseries/forecasting-metrics.html) 204 | 205 | ## Clean up 206 | To avoid unnecessary costs, you must remove all project-provisioned and generated resources from your AWS account. 207 | 208 | ### Shut down SageMaker resources 209 | You must complete this section before deleting the SageMaker domain or the CloudFormation stack. 210 | 211 | Complete the following activities to shut down your Amazon SageMaker resources: 212 | - [Log out of Canvas](https://docs.aws.amazon.com/sagemaker/latest/dg/canvas-log-out.html) 213 | - Make sure to delete all endpoints created by this workshop including Canvas asynchronous endpoints 214 | - [Stop running applications and spaces in Studio](https://docs.aws.amazon.com/sagemaker/latest/dg/studio-updated-running.html#studio-updated-running-stop) > follow the instructions in the section **Use the Studio UI to delete your domain applications** 215 | 216 | ### Remove the SageMaker domain 217 | You don't need to complete this section if you run an AWS-instructor led workshop in an AWS-provisioned account. 218 | 219 | If you used the AWS Console to provision a Studio domain for this workshop, and don't need the domain, you can delete the domain by following the instructions in the [Developer Guide](https://docs.aws.amazon.com/sagemaker/latest/dg/gs-studio-delete-domain.html). 220 | 221 | If you provisioned a Studio domain with the provided CloudFormation template, you can delete the CloudFormation stack in the AWS console. 222 | 223 | If you provisioned a new VPC for the domain, go to the [VPC console](https://console.aws.amazon.com/vpc/home?#vpcs) and delete the provisioned VPC. 224 | 225 | 226 | ## Resources 227 | 228 | ### Algorithms 229 | - [References for machine learning and RCF](https://docs.aws.amazon.com/quicksight/latest/user/learn-more-about-machine-learning-and-rcf.html) 230 | - [Chronos forecasting GitHub repository](https://github.com/amazon-science/chronos-forecasting) 231 | - [Adapting language model architectures for time series forecasting](https://www.amazon.science/blog/adapting-language-model-architectures-for-time-series-forecasting) 232 | - [Chronos: Learning the Language of Time Series](https://arxiv.org/pdf/2403.07815.pdf) 233 | - [AutoGluon](https://github.com/autogluon/autogluon) 234 | - [AutoGluon Time series forecasting](https://auto.gluon.ai/stable/tutorials/timeseries/index.html) 235 | - [GluonTS - Probabilistic Time Series Modeling in Python](https://github.com/awslabs/gluonts) 236 | - [Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting](https://arxiv.org/abs/2012.07436) 237 | - [Sundial: A Family of Highly Capable Time Series Foundation Models](https://arxiv.org/abs/2502.00816v1) 238 | 239 | ### Books and whitepapers 240 | - [Time Series Analysis on AWS: Learn how to build forecasting models and detect anomalies in your time series data](https://www.amazon.com/Time-Analysis-AWS-forecasting-anomalies-ebook/dp/B09MMLLWDY) 241 | - [Time Series Forecasting Principles with Amazon Forecast](https://docs.aws.amazon.com/whitepapers/latest/time-series-forecasting-principles-with-amazon-forecast/time-series-forecasting-principles-with-amazon-forecast.html) 242 | - [Large Language Models Are Zero-Shot Time Series Forecasters](https://arxiv.org/pdf/2310.07820) 243 | - [An Evaluation of Standard Statistical Models and LLMs on Time Series Forecasting](https://arxiv.org/html/2408.04867v1) 244 | - [Forecasting: Principles and Practice](https://otexts.com/fpp3/) 245 | - [A simple combination of univariate models](https://www.sciencedirect.com/science/article/abs/pii/S0169207019300585) 246 | 247 | ### Blog posts 248 | - [Robust time series forecasting with MLOps on Amazon SageMaker](https://aws.amazon.com/blogs/machine-learning/robust-time-series-forecasting-with-mlops-on-amazon-sagemaker/) 249 | - [Deep demand forecasting with Amazon SageMaker](https://aws.amazon.com/blogs/machine-learning/deep-demand-forecasting-with-amazon-sagemaker/) 250 | - [Capture public health insights more quickly with no-code machine learning using Amazon SageMaker Canvas](https://aws.amazon.com/blogs/machine-learning/capture-public-health-insights-more-quickly-with-no-code-machine-learning-using-amazon-sagemaker-canvas/) 251 | - [Speed up your time series forecasting by up to 50 percent with Amazon SageMaker Canvas UI and AutoML APIs](https://aws.amazon.com/blogs/machine-learning/speed-up-your-time-series-forecasting-by-up-to-50-percent-with-amazon-sagemaker-canvas-ui-and-automl-apis/) 252 | - [Sagemaker Automated Model Tuning](https://aws.amazon.com/blogs/aws/sagemaker-automatic-model-tuning/) 253 | - [Time series forecasting with Amazon SageMaker AutoML](https://aws.amazon.com/it/blogs/machine-learning/time-series-forecasting-with-amazon-sagemaker-automl/) 254 | 255 | ### Workshops and notebooks 256 | - [Time series forecasting with AWS services workshop](https://catalog.us-east-1.prod.workshops.aws/workshops/caef4710-3721-4957-a2ce-33799920ef72/en-US) 257 | - [Time series Forecasting with Amazon SageMaker Autopilot](https://github.com/aws/amazon-sagemaker-examples/blob/main/autopilot/autopilot_time_series.ipynb) 258 | - [Deep Demand Forecasting with Amazon SageMaker notebook](https://github.com/awslabs/sagemaker-deep-demand-forecast/blob/mainline/src/deep-demand-forecast.ipynb) 259 | - [Timeseries Forecasting on AWS](https://github.com/aws-samples/time-series-forecasting-on-aws) 260 | - [Inventory Forecasting using Amazon SageMaker](https://catalog.us-east-1.prod.workshops.aws/workshops/866925a4-cb5f-4a3d-9cd7-80edc0aa5f0c/en-US) 261 | - [Tutorial at IJCAI 2021 (with videos)](https://lovvge.github.io/Forecasting-Tutorial-IJCAI-2021/) 262 | 263 | 264 | ## QR codes and links 265 | 266 | ### This GitHub repository 267 | Link: https://github.com/aws-samples/modern-time-series-forecasting-on-aws 268 | Short link: https://bit.ly/47hnKH6 269 | 270 | ![](./img/git-repo-qr-code.png) 271 | 272 | ### AWS workshop 273 | Link: https://catalog.workshops.aws/modern-time-series-forecasting-on-aws/en-US 274 | Short link: https://bit.ly/4dBQ0G8 275 | 276 | ![](./img/workshop-qr-code.svg) 277 | --- 278 | 279 | Copyright Amazon.com, Inc. or its affiliates. All Rights Reserved. 280 | SPDX-License-Identifier: MIT-0 -------------------------------------------------------------------------------- /img/canvas-add-to-model-registry.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/canvas-add-to-model-registry.png -------------------------------------------------------------------------------- /img/canvas-analyze-quick-build.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/canvas-analyze-quick-build.png -------------------------------------------------------------------------------- /img/canvas-analyze-standard-build.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/canvas-analyze-standard-build.png -------------------------------------------------------------------------------- /img/canvas-build-comparison.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/canvas-build-comparison.png -------------------------------------------------------------------------------- /img/canvas-configure-model-config.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/canvas-configure-model-config.png -------------------------------------------------------------------------------- /img/canvas-create-data-flow.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/canvas-create-data-flow.png -------------------------------------------------------------------------------- /img/canvas-create-model.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/canvas-create-model.png -------------------------------------------------------------------------------- /img/canvas-datasets.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/canvas-datasets.png -------------------------------------------------------------------------------- /img/canvas-download-performance-report.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/canvas-download-performance-report.png -------------------------------------------------------------------------------- /img/canvas-endpoint-resource-error.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/canvas-endpoint-resource-error.png -------------------------------------------------------------------------------- /img/canvas-model-deploy.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/canvas-model-deploy.png -------------------------------------------------------------------------------- /img/canvas-model-leaderboard.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/canvas-model-leaderboard.png -------------------------------------------------------------------------------- /img/canvas-model-new-version.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/canvas-model-new-version.png -------------------------------------------------------------------------------- /img/canvas-model-registry-details.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/canvas-model-registry-details.png -------------------------------------------------------------------------------- /img/canvas-predictions.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/canvas-predictions.png -------------------------------------------------------------------------------- /img/canvas-start-quick-build.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/canvas-start-quick-build.png -------------------------------------------------------------------------------- /img/chronos-main-figure.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/chronos-main-figure.png -------------------------------------------------------------------------------- /img/chronos.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/chronos.png -------------------------------------------------------------------------------- /img/git-repo-qr-code.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/git-repo-qr-code.png -------------------------------------------------------------------------------- /img/quicksight_filter_item_101_store_001.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/quicksight_filter_item_101_store_001.png -------------------------------------------------------------------------------- /img/smd-select.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/aws-samples/modern-time-series-forecasting-on-aws/f8b8a32cd0a3b7af37796834fa55dc6bf8748d31/img/smd-select.png -------------------------------------------------------------------------------- /img/workshop-qr-code.svg: -------------------------------------------------------------------------------- 1 | 2 | -------------------------------------------------------------------------------- /notebooks/additional/gluonts_pipeline/inference.py: -------------------------------------------------------------------------------- 1 | import os 2 | import json 3 | from typing import Any, List, Dict, Union 4 | from pathlib import Path 5 | from gluonts.model.predictor import Predictor 6 | from gluonts.dataset.common import ListDataset 7 | from gluonts.dataset.field_names import FieldName 8 | from gluonts.model.forecast import QuantileForecast 9 | import numpy as np 10 | 11 | class QuantileForecastEncoder(json.JSONEncoder): 12 | def default(self, obj): 13 | if isinstance(obj, QuantileForecast): 14 | return { 15 | "__type__": "QuantileForecast", 16 | "forecast_arrays": obj.forecast_array.tolist(), 17 | "start_date": obj.start_date.to_timestamp().isoformat() if obj.start_date else None, 18 | "forecast_keys": obj.forecast_keys, 19 | "item_id": obj.item_id, 20 | "info": obj.info, 21 | "freq": obj.freq.freqstr, 22 | "prediction_length": obj.prediction_length, 23 | 24 | } 25 | if isinstance(obj, np.ndarray): 26 | return obj.tolist() 27 | return super().default(obj) 28 | 29 | 30 | def model_fn(model_dir: str) -> Predictor: 31 | print("loading model from {model_dir}") 32 | predictor = Predictor.deserialize(Path(model_dir)) 33 | print("model was loaded successfully from {model_dir}") 34 | return predictor 35 | 36 | 37 | def transform_fn( 38 | model: Predictor, 39 | request_body: Any, 40 | content_type: Any, 41 | accept_type: Any 42 | ): 43 | # print(f'get {request_body}') 44 | request_data = json.loads(request_body) 45 | 46 | parameters = request_data['parameters'] 47 | request_list_data = ListDataset( 48 | request_data['inputs'], 49 | freq=parameters['freq'], 50 | ) 51 | 52 | forecasts = list(model.predict(request_list_data, num_samples=parameters['num_samples'])) 53 | return json.dumps(forecasts, cls=QuantileForecastEncoder), content_type -------------------------------------------------------------------------------- /notebooks/additional/gluonts_pipeline/preprocess.py: -------------------------------------------------------------------------------- 1 | import os 2 | from typing import Dict 3 | from gluonts.dataset.jsonl import JsonLinesWriter 4 | from pathlib import Path 5 | from gluonts.dataset.split import DateSplitter 6 | from gluonts.dataset.pandas import PandasDataset 7 | from gluonts.dataset.field_names import FieldName 8 | import pandas as pd 9 | import numpy as np 10 | import boto3 11 | import zipfile 12 | import json 13 | import random 14 | import string 15 | 16 | def _generate_unique_id(length=12): 17 | characters = string.ascii_lowercase + string.digits 18 | return ''.join(random.choices(characters, k=length)) 19 | 20 | def preprocess( 21 | input_data_s3_path, 22 | output_s3_prefix, 23 | freq, 24 | prediction_length, 25 | data_start, 26 | data_end, 27 | backtest_windows=4, 28 | sample_size=0, 29 | pipeline_run_id=None, 30 | ) -> Dict[str,str]: 31 | """ 32 | Prepares time series data for training. 33 | """ 34 | # if called without pipeline_run_id, generate a unique run_id 35 | if not pipeline_run_id: 36 | pipeline_run_id = _generate_unique_id() 37 | 38 | # load raw dataset 39 | print(f'Downloading from {input_data_s3_path}') 40 | 41 | os.makedirs("./data", exist_ok=True) 42 | s3 = boto3.client('s3') 43 | 44 | dataset_zip_filename = input_data_s3_path.split('/')[-1] 45 | s3.download_file( 46 | input_data_s3_path.split('/')[2], 47 | '/'.join(input_data_s3_path.split('/')[3:]), 48 | f'./data/{dataset_zip_filename}' 49 | ) 50 | 51 | print(f'Unzipping {dataset_zip_filename}') 52 | zip_ref = zipfile.ZipFile(f'./data/{dataset_zip_filename}', 'r') 53 | zip_ref.extractall('./data') 54 | zip_ref.close() 55 | dataset_path = '.'.join(zip_ref.filename.split('.')[:-1]) 56 | 57 | # load into DataFrame and resample 58 | # supported frequences for this example are 1h or 1d only 59 | print(f'Load dataset from {dataset_path} and resample to {freq} frequency') 60 | data_kw = pd.read_csv( 61 | dataset_path, 62 | sep=';', 63 | index_col=0, 64 | decimal=',', 65 | parse_dates=True, 66 | ).resample(freq).sum() / {'1h':4, '1d':'96'}[freq] 67 | 68 | # get the full dataset or a random sample of sample_size 69 | if sample_size != 0: 70 | print(f'Get a sample of {sample_size} time series out of the full dataset') 71 | ts_sample = data_kw[np.random.choice(data_kw.columns.to_list(), size=sample_size, replace=False)] 72 | else: 73 | print(f'Get the full dataset') 74 | ts_sample = data_kw 75 | 76 | # calculate the end of the training part based on backtest_windows 77 | end_training_date = pd.Period(data_end, freq=freq) - backtest_windows*prediction_length 78 | 79 | # convert to GluonTS format 80 | ts_dataset = PandasDataset( 81 | dict(ts_sample[(ts_sample.index > data_start) & (ts_sample.index <= data_end)]) 82 | ) 83 | # split to get the train dataset 84 | train_ds, _ = DateSplitter(date=end_training_date).split(ts_dataset) 85 | 86 | test_entry = next(iter(ts_dataset)) 87 | train_entry = next(iter(train_ds)) 88 | len_test = len(test_entry['target']) 89 | len_train = len(train_entry['target']) 90 | 91 | print(f'--------------------------------------------------------') 92 | print(f"The test dataset contains {len(train_ds)} time series: {[e[FieldName.ITEM_ID] for e in train_ds]}") 93 | print(f"The test dataset starts {test_entry['start'].to_timestamp()} and ends {test_entry['start'] + len_test} and contains {len_test} data points") 94 | print(f"The train dataset starts {train_entry['start']} and ends {train_entry['start'] + len_train} and contains {len_train} data points") 95 | print(f"The backtest contains {len_test-len_train} data points and has {(len_test-len_train)/prediction_length} windows of {prediction_length} length") 96 | print(f'--------------------------------------------------------') 97 | 98 | # save train and test datasets 99 | train_file_name = 'train.jsonl.gz' 100 | test_file_name = 'test.jsonl.gz' 101 | train_file_path = Path(f'./data/{train_file_name}') 102 | test_file_path = Path(f'./data/{test_file_name}') 103 | train_file_s3_path = f'{output_s3_prefix}/{pipeline_run_id}/train/{train_file_name}' 104 | test_file_s3_path = f'{output_s3_prefix}/{pipeline_run_id}/test/{test_file_name}' 105 | 106 | JsonLinesWriter().write_to_file(train_ds, train_file_path) 107 | JsonLinesWriter().write_to_file(ts_dataset, test_file_path) 108 | 109 | # upload files to S3 110 | print(f'Upload train and test datasets to {train_file_s3_path} and {test_file_s3_path}') 111 | s3.upload_file( 112 | train_file_path, 113 | train_file_s3_path.split('/')[2], 114 | '/'.join(train_file_s3_path.split('/')[3:]) 115 | ) 116 | s3.upload_file( 117 | test_file_path, 118 | test_file_s3_path.split('/')[2], 119 | '/'.join(test_file_s3_path.split('/')[3:]) 120 | ) 121 | 122 | print('### Data processing completed. Exiting.') 123 | 124 | return { 125 | 'train_data':train_file_s3_path, 126 | 'test_data':test_file_s3_path, 127 | 'pipeline_run_id':pipeline_run_id, 128 | } 129 | 130 | 131 | 132 | 133 | 134 | 135 | 136 | 137 | 138 | 139 | 140 | 141 | 142 | -------------------------------------------------------------------------------- /notebooks/additional/gluonts_pipeline/register.py: -------------------------------------------------------------------------------- 1 | from typing import Dict 2 | from sagemaker.estimator import Estimator 3 | 4 | def register( 5 | training_job_name, 6 | model_package_group_name, 7 | model_approval_status='PendingManualApproval', 8 | pipeline_run_id=None, 9 | )-> Dict[str,str]: 10 | """ 11 | Register model trained by the pipeline trained job in SageMaker model registry. 12 | """ 13 | 14 | print(f'Attaching estimator to the job {training_job_name}') 15 | estimator = Estimator.attach(training_job_name) 16 | 17 | print(f'Registering the model in {model_package_group_name}') 18 | supported_instances = ['ml.m5.xlarge', 'ml.m5.2xlarge', "ml.g5.xlarge", 'ml.g5.2xlarge'] 19 | 20 | model_package = estimator.register( 21 | content_types=["application/json"], 22 | response_types=["application/json"], 23 | inference_instances=supported_instances, 24 | transform_instances=supported_instances, 25 | model_package_group_name=model_package_group_name, 26 | approval_status=model_approval_status, 27 | model_name="gluonts-tft-model", 28 | domain="MACHINE_LEARNING", 29 | task="OTHER", 30 | framework='PYTORCH', 31 | framework_version='2.3', 32 | description='GluonTS TFT model group', 33 | customer_metadata_properties={ 34 | 'pipeline_run_id':pipeline_run_id, 35 | } 36 | ) 37 | 38 | print('### Model registration completed. Exiting.') 39 | 40 | return { 41 | "model_package_arn":model_package.model_package_arn, 42 | "model_package_group_name":model_package_group_name, 43 | } -------------------------------------------------------------------------------- /notebooks/additional/gluonts_pipeline/train.py: -------------------------------------------------------------------------------- 1 | import os 2 | import argparse 3 | import json 4 | import pandas as pd 5 | import numpy as np 6 | from typing import List, Dict, Tuple, Union 7 | from gluonts.dataset.jsonl import JsonLinesFile 8 | from pathlib import Path 9 | from gluonts.dataset.common import ListDataset 10 | from gluonts.model.predictor import Predictor 11 | from gluonts.dataset.field_names import FieldName 12 | from gluonts.dataset.split import OffsetSplitter 13 | from gluonts.dataset.util import to_pandas 14 | from gluonts.evaluation import Evaluator 15 | from gluonts.torch import TemporalFusionTransformerEstimator 16 | 17 | 18 | def _load_dataset(path, freq): 19 | return ListDataset(JsonLinesFile(path=path), freq=freq) 20 | 21 | 22 | def _evaluate( 23 | predictor: Predictor, 24 | test_data: ListDataset, 25 | prediction_length: int, 26 | quantiles: List[float] = None, 27 | num_windows: int = 1, 28 | num_samples: int = 20, 29 | ) -> Tuple[Dict[str, float], pd.DataFrame]: 30 | 31 | # prepare test pairs 32 | # the testing windows are taken from the end of the dataset 33 | _, test_template = OffsetSplitter(offset=-num_windows*prediction_length).split(test_data) 34 | 35 | test_pairs = test_template.generate_instances( 36 | prediction_length=prediction_length, 37 | windows=num_windows, 38 | ) 39 | 40 | # predict 41 | forecasts = predictor.predict(test_pairs.input, num_samples) 42 | 43 | # evaluate 44 | evaluator = Evaluator(quantiles=quantiles if quantiles else (np.arange(10) / 10.0)[1:]) 45 | 46 | return evaluator([to_pandas(l) for l in test_pairs.label], forecasts) 47 | 48 | 49 | def _train_predictor( 50 | dataset: ListDataset, 51 | trainer_hp, 52 | model_hp, 53 | ) -> Predictor: 54 | return TemporalFusionTransformerEstimator( 55 | **model_hp, 56 | trainer_kwargs={"max_epochs": trainer_hp['epochs']}, 57 | ).train(dataset) 58 | 59 | def _save_predictor(predictor: Predictor, model_dir: Path): 60 | predictor.serialize(model_dir) 61 | 62 | 63 | if __name__ == "__main__": 64 | 65 | parser = argparse.ArgumentParser() 66 | aa = parser.add_argument 67 | 68 | # data, model, and output directories. Defaults are set in the environment variables. 69 | aa('--output_data_dir', type=str, default=os.environ.get('SM_OUTPUT_DATA_DIR')) 70 | aa('--model_dir', type=str, default=os.environ.get('SM_MODEL_DIR')) 71 | aa('--train_dir', type=str, default=os.environ.get('SM_CHANNEL_TRAIN')) 72 | aa('--test_dir', type=str, default=os.environ.get('SM_CHANNEL_TEST')) 73 | aa('--sm_training_env', type=str, default=os.environ.get('SM_TRAINING_ENV')) 74 | 75 | args, _ = parser.parse_known_args() 76 | print(f'Passed arguments: {args}') 77 | 78 | # get SageMaker enviroment setup 79 | sm_training_env = json.loads(args.sm_training_env) 80 | 81 | # hyperparameters 82 | hyperparameters = sm_training_env['hyperparameters'] 83 | 84 | # load datasets into GluonTS format 85 | print(f'Load datasets into GluonTS format') 86 | train_ds = _load_dataset(Path(f'{args.train_dir}/train.jsonl.gz'), hyperparameters['freq']) 87 | test_ds = _load_dataset(Path(f'{args.test_dir}/test.jsonl.gz'), hyperparameters['freq']) 88 | 89 | train_entry = next(iter(train_ds)) 90 | test_entry = next(iter(test_ds)) 91 | len_train = train_entry[FieldName.TARGET].shape[0] 92 | len_test = test_entry[FieldName.TARGET].shape[0] 93 | 94 | print(f'--------------------------------------------------------') 95 | print(f"The test dataset contains {len(train_ds)} time series: {[e[FieldName.ITEM_ID] for e in train_ds]}") 96 | print(f"The test dataset starts {test_entry[FieldName.START].to_timestamp()} and ends {test_entry[FieldName.START] + len_test} and contains {len_test} data points") 97 | print(f"The train dataset starts {train_entry[FieldName.START]} and ends {train_entry[FieldName.START] + len_train} and contains {len_train} data points") 98 | print(f"The backtest contains {len_test-len_train} data points and has {(len_test-len_train)/hyperparameters['prediction_length']} windows of {hyperparameters['prediction_length']} length") 99 | print(f'--------------------------------------------------------') 100 | 101 | # training 102 | print(f"Training the predictor for {hyperparameters['epochs']} epochs") 103 | predictor = _train_predictor( 104 | train_ds, 105 | { 106 | 'epochs':hyperparameters['epochs'], 107 | }, 108 | { 109 | 'freq':hyperparameters['freq'], 110 | 'prediction_length':hyperparameters['prediction_length'], 111 | 'context_length':hyperparameters['context_length'], 112 | }, 113 | ) 114 | 115 | # evaluation 116 | print(f"Evaluating the model on {hyperparameters['backtest_windows']} rolling windows") 117 | agg_metrics, item_metrics = _evaluate( 118 | predictor, 119 | test_ds, 120 | hyperparameters['prediction_length'], 121 | [float(x) for x in hyperparameters['quantiles'].split(',')], 122 | hyperparameters['backtest_windows'], 123 | hyperparameters['num_samples'], 124 | ) 125 | 126 | # emit test metrics - SageMaker collects them from the log stream 127 | print(f"test_MSE={agg_metrics['MSE']}") 128 | print(f"test_MAPE={agg_metrics['MAPE']}") 129 | print(f"test_sMAPE={agg_metrics['sMAPE']}") 130 | print(f"test_RMSE={agg_metrics['RMSE']}") 131 | print(f"test_mean_wQuantileLoss={agg_metrics['mean_wQuantileLoss']}") 132 | print(f"test_mean_absolute_QuantileLoss={agg_metrics['mean_absolute_QuantileLoss']}") 133 | 134 | # save predictor and results 135 | # os.makedirs('./output/model', exist_ok=True) 136 | 137 | with open(os.path.join(args.output_data_dir, 'agg_metrics.json'), 'w', encoding="utf-8") as fout: 138 | json.dump(agg_metrics, fout) 139 | 140 | item_metrics.to_csv( 141 | os.path.join(args.output_data_dir, 'item_metrics.csv.gz'), 142 | index=False, 143 | encoding="utf-8", 144 | compression="gzip", 145 | ) 146 | 147 | _save_predictor(predictor, Path(args.model_dir)) 148 | 149 | print('### Training completed. Exiting.') -------------------------------------------------------------------------------- /notebooks/additional/gluonts_pipeline/train_step.py: -------------------------------------------------------------------------------- 1 | import os 2 | import json 3 | import pandas as pd 4 | import numpy as np 5 | from typing import List, Dict, Tuple, Union 6 | from gluonts.dataset.jsonl import JsonLinesFile 7 | from pathlib import Path 8 | from gluonts.dataset.common import ListDataset 9 | from gluonts.model.predictor import Predictor 10 | from gluonts.dataset.split import OffsetSplitter 11 | from gluonts.dataset.util import to_pandas 12 | from gluonts.evaluation import Evaluator 13 | from gluonts.torch import TemporalFusionTransformerEstimator 14 | import boto3 15 | 16 | def _upload_directory_to_s3(local_dir, bucket_name, s3_prefix): 17 | s3_client = boto3.client('s3') 18 | 19 | # Ensure the local directory path ends with a separator 20 | local_dir = os.path.join(local_dir, '') 21 | 22 | # Walk through all files in the directory 23 | for root, dirs, files in os.walk(local_dir): 24 | for filename in files: 25 | # Get the full local path 26 | local_path = os.path.join(root, filename) 27 | 28 | # Calculate relative path from the local directory 29 | relative_path = os.path.relpath(local_path, local_dir) 30 | 31 | # Create S3 key with prefix 32 | s3_key = os.path.join(s3_prefix, relative_path).replace("\\", "/") 33 | 34 | try: 35 | print(f"Uploading {local_path} to {bucket_name}/{s3_key}") 36 | s3_client.upload_file(local_path, bucket_name, s3_key) 37 | except Exception as e: 38 | print(f"Error uploading {local_path}: {e}") 39 | 40 | 41 | def _load_dataset(path, freq): 42 | return ListDataset(JsonLinesFile(path=path), freq=freq) 43 | 44 | 45 | def _evaluate( 46 | predictor: Predictor, 47 | test_data: ListDataset, 48 | prediction_length: int, 49 | quantiles: List[float] = None, 50 | num_windows: int = 1, 51 | num_samples: int = 20, 52 | ) -> Tuple[Dict[str, float], pd.DataFrame]: 53 | 54 | # prepare test pairs 55 | # the testing windows are taken from the end of the dataset 56 | _, test_template = OffsetSplitter(offset=-num_windows*prediction_length).split(test_data) 57 | 58 | test_pairs = test_template.generate_instances( 59 | prediction_length=prediction_length, 60 | windows=num_windows, 61 | ) 62 | 63 | # predict 64 | forecasts = predictor.predict(test_pairs.input, num_samples) 65 | 66 | # evaluate 67 | evaluator = Evaluator(quantiles=quantiles if quantiles else (np.arange(10) / 10.0)[1:]) 68 | 69 | return evaluator([to_pandas(l) for l in test_pairs.label], forecasts) 70 | 71 | 72 | def _train_predictor( 73 | dataset: ListDataset, 74 | trainer_hp, 75 | model_hp, 76 | ) -> Predictor: 77 | return TemporalFusionTransformerEstimator( 78 | **model_hp, 79 | trainer_kwargs={"max_epochs": trainer_hp['epochs']}, 80 | ).train(dataset) 81 | 82 | def _save_predictor(predictor: Predictor, model_dir: Path): 83 | predictor.serialize(model_dir) 84 | 85 | 86 | def train( 87 | train_data_s3_path, 88 | test_data_s3_path, 89 | output_s3_prefix, 90 | hyperparameters, 91 | 92 | )-> Dict[str, Union[str,float]]: 93 | """ 94 | Trains the TFT predictor 95 | """ 96 | 97 | freq = hyperparameters['freq'] 98 | prediction_length = hyperparameters['prediction_length'] 99 | 100 | # download datasets from S3 101 | print(f'Download datasets from {train_data_s3_path} and {test_data_s3_path}') 102 | os.makedirs("./data", exist_ok=True) 103 | s3 = boto3.client('s3') 104 | 105 | train_filename = train_data_s3_path.split('/')[-1] 106 | test_filename = test_data_s3_path.split('/')[-1] 107 | s3.download_file( 108 | train_data_s3_path.split('/')[2], 109 | '/'.join(train_data_s3_path.split('/')[3:]), 110 | f'./data/{train_filename}' 111 | ) 112 | s3.download_file( 113 | test_data_s3_path.split('/')[2], 114 | '/'.join(test_data_s3_path.split('/')[3:]), 115 | f'./data/{test_filename}' 116 | ) 117 | 118 | # load datasets into GluonTS format 119 | print(f'Load datasets into GluonTS format') 120 | train_ds = _load_dataset(Path(f'./data/{train_filename}'), freq) 121 | test_ds = _load_dataset(Path(f'./data/{test_filename}'), freq) 122 | 123 | train_entry = next(iter(train_ds)) 124 | test_entry = next(iter(test_ds)) 125 | len_train = train_entry['target'].shape[0] 126 | len_test = test_entry['target'].shape[0] 127 | 128 | print(f'--------------------------------------------------------') 129 | print(f"The test dataset contains {len(train_ds)} time series: {[e['item_id'] for e in train_ds]}") 130 | print(f"The test dataset starts {test_entry['start'].to_timestamp()} and ends {test_entry['start'] + len_test} and contains {len_test} data points") 131 | print(f"The train dataset starts {train_entry['start']} and ends {train_entry['start'] + len_train} and contains {len_train} data points") 132 | print(f"The backtest contains {len_test-len_train} data points and has {(len_test-len_train)/prediction_length} windows of {prediction_length} length") 133 | print(f'--------------------------------------------------------') 134 | 135 | # training 136 | print(f"Training the predictor for {hyperparameters['epochs']} epochs") 137 | predictor = _train_predictor( 138 | train_ds, 139 | { 140 | 'epochs':hyperparameters['epochs'], 141 | }, 142 | { 143 | 'freq':hyperparameters['freq'], 144 | 'prediction_length':hyperparameters['prediction_length'], 145 | 'context_length':hyperparameters['context_length'], 146 | }, 147 | ) 148 | 149 | # evaluation 150 | print(f"Evaluating the model on {hyperparameters['backtest_windows']} rolling windows") 151 | agg_metrics, item_metrics = _evaluate( 152 | predictor, 153 | test_ds, 154 | prediction_length, 155 | [float(x) for x in hyperparameters['quantiles'].split(',')], 156 | hyperparameters['backtest_windows'], 157 | hyperparameters['num_samples'], 158 | ) 159 | 160 | # save predictor and results 161 | os.makedirs('./output/model', exist_ok=True) 162 | 163 | with open(os.path.join('./output', 'agg_metrics.json'), 'w', encoding="utf-8") as fout: 164 | json.dump(agg_metrics, fout) 165 | 166 | item_metrics.to_csv( 167 | os.path.join('./output', 'item_metrics.csv.gz'), 168 | index=False, 169 | encoding="utf-8", 170 | compression="gzip", 171 | ) 172 | 173 | _save_predictor(predictor, Path('./output/model')) 174 | 175 | # upload artifacts to S3 176 | metrics_s3_path = f'{output_s3_prefix}' 177 | model_s3_path = f'{output_s3_prefix}/model' 178 | 179 | print(f'Upload metrics to {metrics_s3_path} and {model_s3_path}') 180 | _upload_directory_to_s3('./output', metrics_s3_path.split('/')[2], '/'.join(metrics_s3_path.split('/')[3:])) 181 | _upload_directory_to_s3('./output/model', model_s3_path.split('/')[2], '/'.join(model_s3_path.split('/')[3:])) 182 | 183 | print('### Training completed. Exiting.') 184 | 185 | return { 186 | 'metrics_s3_path':metrics_s3_path, 187 | 'model_s3_path':model_s3_path, 188 | 'agg_metrics':agg_metrics, 189 | } 190 | 191 | 192 | -------------------------------------------------------------------------------- /notebooks/additional/lab2a_tide.ipynb: -------------------------------------------------------------------------------- 1 | { 2 | "cells": [ 3 | { 4 | "cell_type": "markdown", 5 | "id": "79856ce0-f2bf-4c44-b7be-f7fbffb26648", 6 | "metadata": {}, 7 | "source": [ 8 | "\"GluonTS\n", 9 | "\n", 10 | "\n", 11 | "\n", 12 | "# Lab 2A: Implement a custom algorithm on Amazon SageMaker AI\n", 13 | "\n", 14 | "This notebook demonstrates how to use SageMaker infrastructure, features, and [Python SDK](https://sagemaker.readthedocs.io/en/stable/index.html) to build, evaluate, and train any time series forecasting algorithm. You learn how to train a model in a notebook, in a remote function, and with a built-in SageMaker training container.\n", 15 | "\n", 16 | "As an example of a time series forecast algorithm, this notebook uses a popular state-of-the-art algorithm TiDE.\n", 17 | "\n", 18 | "TiDE (Time-series Dense Encoder) is a powerful MLP-based encoder-decoder model developed by Google Research that excels at long-term time series forecasting. Unlike complex Transformer-based approaches, TiDE achieves state-of-the-art performance using a simple architecture built on dense neural networks. The model supports both univariate and multivariate forecasting, handles past and future covariates, and provides probabilistic forecasting capabilities while maintaining linear computational scaling. In this notebook, you explore how to implement and train TiDE models using the [Darts library](https://unit8co.github.io/darts/) ([Darts paper](https://www.jmlr.org/papers/v23/21-1177.html)), which offers a comprehensive implementation of the architecture. \n", 19 | "\n", 20 | "For more details on TiDE refer to the [orginal paper](https://arxiv.org/pdf/2304.08424.pdf)." 21 | ] 22 | }, 23 | { 24 | "cell_type": "code", 25 | "execution_count": null, 26 | "id": "f8d92142-dbe4-47b1-a426-4f87f762f0e7", 27 | "metadata": {}, 28 | "outputs": [], 29 | "source": [] 30 | }, 31 | { 32 | "cell_type": "code", 33 | "execution_count": null, 34 | "id": "b0ea5193-5810-48a5-b444-f0f4bd33a88b", 35 | "metadata": {}, 36 | "outputs": [], 37 | "source": [] 38 | }, 39 | { 40 | "cell_type": "code", 41 | "execution_count": null, 42 | "id": "857576f5-bf49-4c7e-93e9-ec24c6d19307", 43 | "metadata": {}, 44 | "outputs": [], 45 | "source": [] 46 | }, 47 | { 48 | "cell_type": "code", 49 | "execution_count": null, 50 | "id": "af163342-0529-4417-8221-dde428fa7c6e", 51 | "metadata": {}, 52 | "outputs": [], 53 | "source": [] 54 | } 55 | ], 56 | "metadata": { 57 | "kernelspec": { 58 | "display_name": "Python 3 (ipykernel)", 59 | "language": "python", 60 | "name": "python3" 61 | }, 62 | "language_info": { 63 | "codemirror_mode": { 64 | "name": "ipython", 65 | "version": 3 66 | }, 67 | "file_extension": ".py", 68 | "mimetype": "text/x-python", 69 | "name": "python", 70 | "nbconvert_exporter": "python", 71 | "pygments_lexer": "ipython3", 72 | "version": "3.11.10" 73 | } 74 | }, 75 | "nbformat": 4, 76 | "nbformat_minor": 5 77 | } 78 | -------------------------------------------------------------------------------- /scripts/lcc-script.sh: -------------------------------------------------------------------------------- 1 | #!/bin/bash 2 | 3 | set -e # Exit immediately if any command exits with a non-zero status 4 | 5 | if [ ! -z "${SM_JOB_DEF_VERSION}" ] 6 | then 7 | echo "Running in job mode, skip lcc" 8 | else 9 | echo "Cloning repository..." 10 | git clone https://github.com/aws-samples/modern-time-series-forecasting-on-aws.git || { echo "Error: Failed to clone repository"; exit 0; } 11 | echo "Files cloned from GitHub repo" 12 | fi 13 | -------------------------------------------------------------------------------- /test/model-performance/autogluon-2H-2-17533-20240919-145927.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20240919-145927,WQL,0.1156577823719694,autogluon-2H-2-17533 3 | 20240919-145927,MAPE,0.2885169559140486,autogluon-2H-2-17533 4 | 20240919-145927,WAPE,0.15921398464384248,autogluon-2H-2-17533 5 | 20240919-145927,RMSE,360.86992704956197,autogluon-2H-2-17533 6 | 20240919-145927,MASE,2.2880993443987947,autogluon-2H-2-17533 7 | -------------------------------------------------------------------------------- /test/model-performance/autogluon-2h-370-17533-20240927-145601.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20240927-145601,WQL,0.11037998847054764,autogluon-2h-370-17533 3 | 20240927-145601,MAPE,1.6516148818064893,autogluon-2h-370-17533 4 | 20240927-145601,WAPE,0.1406391102446078,autogluon-2h-370-17533 5 | 20240927-145601,RMSE,469.43349226376284,autogluon-2h-370-17533 6 | 20240927-145601,MASE,2.2952414901217364,autogluon-2h-370-17533 7 | -------------------------------------------------------------------------------- /test/model-performance/autogluon-Chronos[base]-2h-370-17533-20240930-114821.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20240930-114821,WQL,0.12102911796784806,autogluon-Chronos[base]-2h-370-17533 3 | 20240930-114821,MAPE,1.47092451846345,autogluon-Chronos[base]-2h-370-17533 4 | 20240930-114821,WAPE,0.15875072294548903,autogluon-Chronos[base]-2h-370-17533 5 | 20240930-114821,RMSE,486.90959145556263,autogluon-Chronos[base]-2h-370-17533 6 | 20240930-114821,MASE,2.58511763010597,autogluon-Chronos[base]-2h-370-17533 7 | -------------------------------------------------------------------------------- /test/model-performance/autogluon-Chronos[base]-2h-370-17533-bt4-20240930-122110.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20240930-122110,WQL,0.07136258019209654,autogluon-Chronos[base]-2h-370-17533-bt4 3 | 20240930-122110,MAPE,0.7416368759663339,autogluon-Chronos[base]-2h-370-17533-bt4 4 | 20240930-122110,WAPE,0.09690419086435706,autogluon-Chronos[base]-2h-370-17533-bt4 5 | 20240930-122110,RMSE,332.57348314704785,autogluon-Chronos[base]-2h-370-17533-bt4 6 | 20240930-122110,MASE,1.8060419826796434,autogluon-Chronos[base]-2h-370-17533-bt4 7 | -------------------------------------------------------------------------------- /test/model-performance/autogluon-Chronos[mini]-2H-2-17533-20240919-181432.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20240919-181432,WQL,0.18883697269731292,autogluon-Chronos[mini]-2H-2-17533 3 | 20240919-181432,MAPE,0.31105072680353685,autogluon-Chronos[mini]-2H-2-17533 4 | 20240919-181432,WAPE,0.24859525461343024,autogluon-Chronos[mini]-2H-2-17533 5 | 20240919-181432,RMSE,480.77695008147555,autogluon-Chronos[mini]-2H-2-17533 6 | 20240919-181432,MASE,3.216504374910825,autogluon-Chronos[mini]-2H-2-17533 7 | -------------------------------------------------------------------------------- /test/model-performance/autogluon-Chronos[mini]-2H-2-17533-20240919-182135.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20240919-182135,WQL,0.18883697269731292,autogluon-Chronos[mini]-2H-2-17533 3 | 20240919-182135,MAPE,0.31105072680353685,autogluon-Chronos[mini]-2H-2-17533 4 | 20240919-182135,WAPE,0.24859525461343024,autogluon-Chronos[mini]-2H-2-17533 5 | 20240919-182135,RMSE,480.77695008147555,autogluon-Chronos[mini]-2H-2-17533 6 | 20240919-182135,MASE,3.216504374910825,autogluon-Chronos[mini]-2H-2-17533 7 | -------------------------------------------------------------------------------- /test/model-performance/autogluon-Chronos[mini]-2H-2-17533-bt4-20240919-182239.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20240919-182239,WQL,0.0940352201345739,autogluon-Chronos[mini]-2H-2-17533-bt4 3 | 20240919-182239,MAPE,0.14535047086521502,autogluon-Chronos[mini]-2H-2-17533-bt4 4 | 20240919-182239,WAPE,0.12784565920417557,autogluon-Chronos[mini]-2H-2-17533-bt4 5 | 20240919-182239,RMSE,261.14083988245284,autogluon-Chronos[mini]-2H-2-17533-bt4 6 | 20240919-182239,MASE,1.7555795864080155,autogluon-Chronos[mini]-2H-2-17533-bt4 7 | -------------------------------------------------------------------------------- /test/model-performance/autopilot-2H-370-17533-20240919-151404.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20240919-151404,test:mean_wQuantileLoss,0.14979657456312492,autopilot-2H-370-17533 3 | 20240919-151404,test:MAPE,2.305517748336631,autopilot-2H-370-17533 4 | 20240919-151404,test:WAPE,0.22661944788511876,autopilot-2H-370-17533 5 | 20240919-151404,test:MASE,3.4459806753541664,autopilot-2H-370-17533 6 | 20240919-151404,test:RMSE,857.5111828358137,autopilot-2H-370-17533 7 | -------------------------------------------------------------------------------- /test/model-performance/autopilot-full-2h-370-17533-20241001-091815.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20241001-091815,RMSE,510.97817799106963,autopilot-full-2h-370-17533 3 | 20241001-091815,MAPE,1.3929039028229513,autopilot-full-2h-370-17533 4 | 20241001-091815,AverageWeightedQuantileLoss,0.12616037450201764,autopilot-full-2h-370-17533 5 | 20241001-091815,MASE,3.3095442369554786,autopilot-full-2h-370-17533 6 | 20241001-091815,WAPE,0.16565958678793932,autopilot-full-2h-370-17533 7 | -------------------------------------------------------------------------------- /test/model-performance/canvas-1D-2-1462-20240927-150101.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20240927-150101,unweighted_p10_quantile_loss,79.35375816346044,canvas-1D-2-1462 3 | 20240927-150101,unweighted_p50_quantile_loss,110.36289838400728,canvas-1D-2-1462 4 | 20240927-150101,unweighted_p90_quantile_loss,115.68823908288182,canvas-1D-2-1462 5 | 20240927-150101,w_p10_quantile_loss,0.14812526890431274,canvas-1D-2-1462 6 | 20240927-150101,w_p50_quantile_loss,0.20600831489941798,canvas-1D-2-1462 7 | 20240927-150101,w_p90_quantile_loss,0.21594883367614676,canvas-1D-2-1462 8 | 20240927-150101,mse,117.38312582165692,canvas-1D-2-1462 9 | 20240927-150101,MAPE,5.471807199970325,canvas-1D-2-1462 10 | 20240927-150101,MASE,0.5659456483257921,canvas-1D-2-1462 11 | 20240927-150101,RMSE,10.834349349252909,canvas-1D-2-1462 12 | 20240927-150101,WAPE,0.20708747993771542,canvas-1D-2-1462 13 | 20240927-150101,total_demand,535.7206015586853,canvas-1D-2-1462 14 | 20240927-150101,mean_wQuantileLoss,0.19002747249329252,canvas-1D-2-1462 15 | -------------------------------------------------------------------------------- /test/model-performance/canvas-1D-2-35065-20240920-091227.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20240920-091227,unweighted_p10_quantile_loss,522.1545467516479,canvas-1D-2-35065 3 | 20240920-091227,unweighted_p50_quantile_loss,418.2918188645812,canvas-1D-2-35065 4 | 20240920-091227,unweighted_p90_quantile_loss,88.93523869396464,canvas-1D-2-35065 5 | 20240920-091227,w_p10_quantile_loss,0.2705572393809178,canvas-1D-2-35065 6 | 20240920-091227,w_p50_quantile_loss,0.21674019784309548,canvas-1D-2-35065 7 | 20240920-091227,w_p90_quantile_loss,0.04608228122193617,canvas-1D-2-35065 8 | 20240920-091227,mse,5214.695808655263,canvas-1D-2-35065 9 | 20240920-091227,MAPE,13.40919101592671,canvas-1D-2-35065 10 | 20240920-091227,MASE,0.8540038436366449,canvas-1D-2-35065 11 | 20240920-091227,RMSE,72.21285071685831,canvas-1D-2-35065 12 | 20240920-091227,WAPE,0.2129545380743781,canvas-1D-2-35065 13 | 20240920-091227,total_demand,1929.9226586818695,canvas-1D-2-35065 14 | 20240920-091227,mean_wQuantileLoss,0.17779323948198314,canvas-1D-2-35065 15 | -------------------------------------------------------------------------------- /test/model-performance/canvas-1D-370-35065-20240920-091316.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20240920-091316,unweighted_p10_quantile_loss,292683.4389936282,canvas-1D-370-35065 3 | 20240920-091316,unweighted_p50_quantile_loss,294649.06495108077,canvas-1D-370-35065 4 | 20240920-091316,unweighted_p90_quantile_loss,103816.21627109816,canvas-1D-370-35065 5 | 20240920-091316,w_p10_quantile_loss,0.24906812284923815,canvas-1D-370-35065 6 | 20240920-091316,w_p50_quantile_loss,0.25074083371094535,canvas-1D-370-35065 7 | 20240920-091316,w_p90_quantile_loss,0.08834565493989524,canvas-1D-370-35065 8 | 20240920-091316,mse,564504.6974620013,canvas-1D-370-35065 9 | 20240920-091316,MAPE,30.353922822828423,canvas-1D-370-35065 10 | 20240920-091316,MASE,5.516681237219801,canvas-1D-370-35065 11 | 20240920-091316,RMSE,751.3352763327443,canvas-1D-370-35065 12 | 20240920-091316,WAPE,0.25157162443989295,canvas-1D-370-35065 13 | 20240920-091316,total_demand,1175114.0035322406,canvas-1D-370-35065 14 | 20240920-091316,mean_wQuantileLoss,0.1960515371666929,canvas-1D-370-35065 15 | -------------------------------------------------------------------------------- /test/model-performance/canvas-1D-370-35065-20240927-105539.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20240927-105539,unweighted_p10_quantile_loss,297199.3998739211,canvas-1D-2-35065 3 | 20240927-105539,unweighted_p50_quantile_loss,295534.347844236,canvas-1D-2-35065 4 | 20240927-105539,unweighted_p90_quantile_loss,114231.87216739124,canvas-1D-2-35065 5 | 20240927-105539,w_p10_quantile_loss,0.25291112094705553,canvas-1D-2-35065 6 | 20240927-105539,w_p50_quantile_loss,0.2514941928663074,canvas-1D-2-35065 7 | 20240927-105539,w_p90_quantile_loss,0.09720918295929162,canvas-1D-2-35065 8 | 20240927-105539,mse,618814.8448009331,canvas-1D-2-35065 9 | 20240927-105539,MAPE,29.94114844437655,canvas-1D-2-35065 10 | 20240927-105539,MASE,5.388823451737506,canvas-1D-2-35065 11 | 20240927-105539,RMSE,786.6478531089582,canvas-1D-2-35065 12 | 20240927-105539,WAPE,0.25089821719852534,canvas-1D-2-35065 13 | 20240927-105539,total_demand,1175114.0035322406,canvas-1D-2-35065 14 | 20240927-105539,mean_wQuantileLoss,0.20053816559088486,canvas-1D-2-35065 15 | -------------------------------------------------------------------------------- /test/model-performance/canvas-1H-2-1462-20240927-191325.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20240927-191325,unweighted_p10_quantile_loss,1931.487541330404,canvas-1H-2-1462 3 | 20240927-191325,unweighted_p50_quantile_loss,2976.2502571636023,canvas-1H-2-1462 4 | 20240927-191325,unweighted_p90_quantile_loss,1489.1675719392233,canvas-1H-2-1462 5 | 20240927-191325,w_p10_quantile_loss,0.04400921315553926,canvas-1H-2-1462 6 | 20240927-191325,w_p50_quantile_loss,0.06781427742553345,canvas-1H-2-1462 7 | 20240927-191325,w_p90_quantile_loss,0.03393089093013167,canvas-1H-2-1462 8 | 20240927-191325,mse,295.2931500637976,canvas-1H-2-1462 9 | 20240927-191325,MAPE,0.09448020517932307,canvas-1H-2-1462 10 | 20240927-191325,MASE,0.6833017471674712,canvas-1H-2-1462 11 | 20240927-191325,RMSE,17.184095846561075,canvas-1H-2-1462 12 | 20240927-191325,WAPE,0.06851069553250001,canvas-1H-2-1462 13 | 20240927-191325,total_demand,43888.25436401367,canvas-1H-2-1462 14 | 20240927-191325,mean_wQuantileLoss,0.04858479383706813,canvas-1H-2-1462 15 | -------------------------------------------------------------------------------- /test/model-performance/canvas-1H-370-35065-20240920-091430.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20240920-091430,unweighted_p10_quantile_loss,4422313.214299129,canvas-1H-370-35065 3 | 20240920-091430,unweighted_p50_quantile_loss,4959004.337389071,canvas-1H-370-35065 4 | 20240920-091430,unweighted_p90_quantile_loss,1788631.0110390235,canvas-1H-370-35065 5 | 20240920-091430,w_p10_quantile_loss,0.14370379287627855,canvas-1H-370-35065 6 | 20240920-091430,w_p50_quantile_loss,0.16114365890423862,canvas-1H-370-35065 7 | 20240920-091430,w_p90_quantile_loss,0.05812185792524794,canvas-1H-370-35065 8 | 20240920-091430,mse,276069.7514223092,canvas-1H-370-35065 9 | 20240920-091430,MAPE,0.5323419548980269,canvas-1H-370-35065 10 | 20240920-091430,MASE,2.5127543740670513,canvas-1H-370-35065 11 | 20240920-091430,RMSE,525.4234020504884,canvas-1H-370-35065 12 | 20240920-091430,WAPE,0.16186690776404453,canvas-1H-370-35065 13 | 20240920-091430,total_demand,30773809.97247936,canvas-1H-370-35065 14 | 20240920-091430,mean_wQuantileLoss,0.12098976990192171,canvas-1H-370-35065 15 | -------------------------------------------------------------------------------- /test/model-performance/canvas-1H-370-35065-20240927-134914.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20240927-134914,unweighted_p10_quantile_loss,4461949.856010195,canvas-1H-370-35065 3 | 20240927-134914,unweighted_p50_quantile_loss,4936261.369882716,canvas-1H-370-35065 4 | 20240927-134914,unweighted_p90_quantile_loss,1790727.625640128,canvas-1H-370-35065 5 | 20240927-134914,w_p10_quantile_loss,0.14499179204656368,canvas-1H-370-35065 6 | 20240927-134914,w_p50_quantile_loss,0.16040462244672188,canvas-1H-370-35065 7 | 20240927-134914,w_p90_quantile_loss,0.05818998776042206,canvas-1H-370-35065 8 | 20240927-134914,mse,276198.33052579156,canvas-1H-370-35065 9 | 20240927-134914,MAPE,0.5682595636394644,canvas-1H-370-35065 10 | 20240927-134914,MASE,2.7190232957157017,canvas-1H-370-35065 11 | 20240927-134914,RMSE,525.5457454168871,canvas-1H-370-35065 12 | 20240927-134914,WAPE,0.16123915241712355,canvas-1H-370-35065 13 | 20240927-134914,total_demand,30773809.97247936,canvas-1H-370-35065 14 | 20240927-134914,mean_wQuantileLoss,0.12119546741790255,canvas-1H-370-35065 15 | -------------------------------------------------------------------------------- /test/model-performance/chronos-2H-370-17533-bt4-20240919-130527.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20240919-130527,WQL,0.08387127944380027,chronos-chronos-t5-small-2H-370-17533-bt4 3 | 20240919-130527,MAPE,0.952375908485444,chronos-chronos-t5-small-2H-370-17533-bt4 4 | 20240919-130527,WAPE,0.11370558726111012,chronos-chronos-t5-small-2H-370-17533-bt4 5 | 20240919-130527,RMSE,308.83172065685693,chronos-chronos-t5-small-2H-370-17533-bt4 6 | 20240919-130527,MASE,1.9435703009669876,chronos-chronos-t5-small-2H-370-17533-bt4 7 | 20240919-130527,WQL,0.07530871644875207,chronos-chronos-t5-base-2H-370-17533-bt4 8 | 20240919-130527,MAPE,0.6694141157625808,chronos-chronos-t5-base-2H-370-17533-bt4 9 | 20240919-130527,WAPE,0.10237149753335748,chronos-chronos-t5-base-2H-370-17533-bt4 10 | 20240919-130527,RMSE,264.7049115043852,chronos-chronos-t5-base-2H-370-17533-bt4 11 | 20240919-130527,MASE,1.808932873018792,chronos-chronos-t5-base-2H-370-17533-bt4 12 | 20240919-130527,WQL,0.07304669144187251,chronos-chronos-t5-large-2H-370-17533-bt4 13 | 20240919-130527,MAPE,0.7852700020916967,chronos-chronos-t5-large-2H-370-17533-bt4 14 | 20240919-130527,WAPE,0.0994895806641236,chronos-chronos-t5-large-2H-370-17533-bt4 15 | 20240919-130527,RMSE,268.68509695234684,chronos-chronos-t5-large-2H-370-17533-bt4 16 | 20240919-130527,MASE,1.780262578979926,chronos-chronos-t5-large-2H-370-17533-bt4 17 | -------------------------------------------------------------------------------- /test/model-performance/chronos-2H-370-17533-off10-20240919-130527.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20240919-130527,WQL,0.062098830633961564,chronos-chronos-t5-small-2H-370-17533-off10 3 | 20240919-130527,MAPE,0.5717769900528037,chronos-chronos-t5-small-2H-370-17533-off10 4 | 20240919-130527,WAPE,0.08857892065787676,chronos-chronos-t5-small-2H-370-17533-off10 5 | 20240919-130527,RMSE,345.87860162878553,chronos-chronos-t5-small-2H-370-17533-off10 6 | 20240919-130527,MASE,1.64961587503781,chronos-chronos-t5-small-2H-370-17533-off10 7 | 20240919-130527,WQL,0.05673146361178913,chronos-chronos-t5-base-2H-370-17533-off10 8 | 20240919-130527,MAPE,0.484702837086737,chronos-chronos-t5-base-2H-370-17533-off10 9 | 20240919-130527,WAPE,0.08153096465865009,chronos-chronos-t5-base-2H-370-17533-off10 10 | 20240919-130527,RMSE,285.10273880993134,chronos-chronos-t5-base-2H-370-17533-off10 11 | 20240919-130527,MASE,1.57590851804799,chronos-chronos-t5-base-2H-370-17533-off10 12 | 20240919-130527,WQL,0.05686362673922743,chronos-chronos-t5-large-2H-370-17533-off10 13 | 20240919-130527,MAPE,0.41598719208847507,chronos-chronos-t5-large-2H-370-17533-off10 14 | 20240919-130527,WAPE,0.08108158218258046,chronos-chronos-t5-large-2H-370-17533-off10 15 | 20240919-130527,RMSE,282.1125940188181,chronos-chronos-t5-large-2H-370-17533-off10 16 | 20240919-130527,MASE,1.5247410696140768,chronos-chronos-t5-large-2H-370-17533-off10 17 | -------------------------------------------------------------------------------- /test/model-performance/deepar-2H-370-17533-20240919-202341.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20240919-202341,test:mean_wQuantileLoss,0.057275453987167074,deepar-2H-370-17533 3 | 20240919-202341,train:final_loss,3.725597706708041,deepar-2H-370-17533 4 | 20240919-202341,test:RMSE,621.6844993824717,deepar-2H-370-17533 5 | -------------------------------------------------------------------------------- /test/model-performance/deepar-2h-2-17533-20240927-105431.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20240927-105431,test:mean_wQuantileLoss,0.07233166282643218,deepar-2h-2-17533 3 | 20240927-105431,train:final_loss,3.733606360175393,deepar-2h-2-17533 4 | 20240927-105431,test:RMSE,877.3897729764718,deepar-2h-2-17533 5 | -------------------------------------------------------------------------------- /test/model-performance/gluonts-1h-10-8736-bt4-20241106-080455.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20241106-080455,MSE,20554.96211210322,gluonts-SimpleFeedForward-1h-10-8736-bt4 3 | 20241106-080455,abs_error,310388.6882653013,gluonts-SimpleFeedForward-1h-10-8736-bt4 4 | 20241106-080455,abs_target_sum,1933850.430311788,gluonts-SimpleFeedForward-1h-10-8736-bt4 5 | 20241106-080455,abs_target_mean,287.77536165353985,gluonts-SimpleFeedForward-1h-10-8736-bt4 6 | 20241106-080455,MAPE,0.49496113229443317,gluonts-SimpleFeedForward-1h-10-8736-bt4 7 | 20241106-080455,sMAPE,0.3394244618303942,gluonts-SimpleFeedForward-1h-10-8736-bt4 8 | 20241106-080455,num_masked_target_values,0.0,gluonts-SimpleFeedForward-1h-10-8736-bt4 9 | 20241106-080455,QuantileLoss[0.1],240710.38248697147,gluonts-SimpleFeedForward-1h-10-8736-bt4 10 | 20241106-080455,Coverage[0.1],0.16800595238095237,gluonts-SimpleFeedForward-1h-10-8736-bt4 11 | 20241106-080455,QuantileLoss[0.2],289160.4046860036,gluonts-SimpleFeedForward-1h-10-8736-bt4 12 | 20241106-080455,Coverage[0.2],0.2541666666666667,gluonts-SimpleFeedForward-1h-10-8736-bt4 13 | 20241106-080455,QuantileLoss[0.3],312643.3398831085,gluonts-SimpleFeedForward-1h-10-8736-bt4 14 | 20241106-080455,Coverage[0.3],0.33556547619047616,gluonts-SimpleFeedForward-1h-10-8736-bt4 15 | 20241106-080455,QuantileLoss[0.4],316419.8080766086,gluonts-SimpleFeedForward-1h-10-8736-bt4 16 | 20241106-080455,Coverage[0.4],0.4153273809523809,gluonts-SimpleFeedForward-1h-10-8736-bt4 17 | 20241106-080455,QuantileLoss[0.5],310388.6882653013,gluonts-SimpleFeedForward-1h-10-8736-bt4 18 | 20241106-080455,Coverage[0.5],0.5055059523809524,gluonts-SimpleFeedForward-1h-10-8736-bt4 19 | 20241106-080455,QuantileLoss[0.6],294462.0134910503,gluonts-SimpleFeedForward-1h-10-8736-bt4 20 | 20241106-080455,Coverage[0.6],0.5465773809523811,gluonts-SimpleFeedForward-1h-10-8736-bt4 21 | 20241106-080455,QuantileLoss[0.7],262103.71350753936,gluonts-SimpleFeedForward-1h-10-8736-bt4 22 | 20241106-080455,Coverage[0.7],0.6364583333333333,gluonts-SimpleFeedForward-1h-10-8736-bt4 23 | 20241106-080455,QuantileLoss[0.8],217507.30710448523,gluonts-SimpleFeedForward-1h-10-8736-bt4 24 | 20241106-080455,Coverage[0.8],0.7311011904761904,gluonts-SimpleFeedForward-1h-10-8736-bt4 25 | 20241106-080455,QuantileLoss[0.9],158937.5408318817,gluonts-SimpleFeedForward-1h-10-8736-bt4 26 | 20241106-080455,Coverage[0.9],0.8313988095238095,gluonts-SimpleFeedForward-1h-10-8736-bt4 27 | 20241106-080455,RMSE,143.37001817710433,gluonts-SimpleFeedForward-1h-10-8736-bt4 28 | 20241106-080455,NRMSE,0.4982011571571272,gluonts-SimpleFeedForward-1h-10-8736-bt4 29 | 20241106-080455,ND,0.16050294448845168,gluonts-SimpleFeedForward-1h-10-8736-bt4 30 | 20241106-080455,wQuantileLoss[0.1],0.12447207845756848,gluonts-SimpleFeedForward-1h-10-8736-bt4 31 | 20241106-080455,wQuantileLoss[0.2],0.14952573381767856,gluonts-SimpleFeedForward-1h-10-8736-bt4 32 | 20241106-080455,wQuantileLoss[0.3],0.16166883176828836,gluonts-SimpleFeedForward-1h-10-8736-bt4 33 | 20241106-080455,wQuantileLoss[0.4],0.16362165507577195,gluonts-SimpleFeedForward-1h-10-8736-bt4 34 | 20241106-080455,wQuantileLoss[0.5],0.16050294448845168,gluonts-SimpleFeedForward-1h-10-8736-bt4 35 | 20241106-080455,wQuantileLoss[0.6],0.15226721202196344,gluonts-SimpleFeedForward-1h-10-8736-bt4 36 | 20241106-080455,wQuantileLoss[0.7],0.135534635667393,gluonts-SimpleFeedForward-1h-10-8736-bt4 37 | 20241106-080455,wQuantileLoss[0.8],0.112473696877073,gluonts-SimpleFeedForward-1h-10-8736-bt4 38 | 20241106-080455,wQuantileLoss[0.9],0.0821870907597837,gluonts-SimpleFeedForward-1h-10-8736-bt4 39 | 20241106-080455,mean_absolute_QuantileLoss,266925.91092588333,gluonts-SimpleFeedForward-1h-10-8736-bt4 40 | 20241106-080455,mean_wQuantileLoss,0.13802820877044134,gluonts-SimpleFeedForward-1h-10-8736-bt4 41 | 20241106-080455,MAE_Coverage,0.3466435185185185,gluonts-SimpleFeedForward-1h-10-8736-bt4 42 | 20241106-080455,MSE,20076.85618215075,gluonts-NBEATS-1h-10-8736-bt4 43 | 20241106-080455,abs_error,355277.7003341072,gluonts-NBEATS-1h-10-8736-bt4 44 | 20241106-080455,abs_target_sum,1933850.430311788,gluonts-NBEATS-1h-10-8736-bt4 45 | 20241106-080455,abs_target_mean,287.77536165353985,gluonts-NBEATS-1h-10-8736-bt4 46 | 20241106-080455,MAPE,0.31730921659271843,gluonts-NBEATS-1h-10-8736-bt4 47 | 20241106-080455,sMAPE,0.3608155859211519,gluonts-NBEATS-1h-10-8736-bt4 48 | 20241106-080455,num_masked_target_values,0.0,gluonts-NBEATS-1h-10-8736-bt4 49 | 20241106-080455,QuantileLoss[0.1],346993.6474386893,gluonts-NBEATS-1h-10-8736-bt4 50 | 20241106-080455,Coverage[0.1],0.4866071428571429,gluonts-NBEATS-1h-10-8736-bt4 51 | 20241106-080455,QuantileLoss[0.2],349064.66066254384,gluonts-NBEATS-1h-10-8736-bt4 52 | 20241106-080455,Coverage[0.2],0.4866071428571429,gluonts-NBEATS-1h-10-8736-bt4 53 | 20241106-080455,QuantileLoss[0.3],351135.67388639826,gluonts-NBEATS-1h-10-8736-bt4 54 | 20241106-080455,Coverage[0.3],0.4866071428571429,gluonts-NBEATS-1h-10-8736-bt4 55 | 20241106-080455,QuantileLoss[0.4],353206.6871102528,gluonts-NBEATS-1h-10-8736-bt4 56 | 20241106-080455,Coverage[0.4],0.4866071428571429,gluonts-NBEATS-1h-10-8736-bt4 57 | 20241106-080455,QuantileLoss[0.5],355277.7003341072,gluonts-NBEATS-1h-10-8736-bt4 58 | 20241106-080455,Coverage[0.5],0.4866071428571429,gluonts-NBEATS-1h-10-8736-bt4 59 | 20241106-080455,QuantileLoss[0.6],357348.71355796175,gluonts-NBEATS-1h-10-8736-bt4 60 | 20241106-080455,Coverage[0.6],0.4866071428571429,gluonts-NBEATS-1h-10-8736-bt4 61 | 20241106-080455,QuantileLoss[0.7],359419.7267818162,gluonts-NBEATS-1h-10-8736-bt4 62 | 20241106-080455,Coverage[0.7],0.4866071428571429,gluonts-NBEATS-1h-10-8736-bt4 63 | 20241106-080455,QuantileLoss[0.8],361490.7400056707,gluonts-NBEATS-1h-10-8736-bt4 64 | 20241106-080455,Coverage[0.8],0.4866071428571429,gluonts-NBEATS-1h-10-8736-bt4 65 | 20241106-080455,QuantileLoss[0.9],363561.75322952523,gluonts-NBEATS-1h-10-8736-bt4 66 | 20241106-080455,Coverage[0.9],0.4866071428571429,gluonts-NBEATS-1h-10-8736-bt4 67 | 20241106-080455,RMSE,141.6928233262036,gluonts-NBEATS-1h-10-8736-bt4 68 | 20241106-080455,NRMSE,0.49237301800976013,gluonts-NBEATS-1h-10-8736-bt4 69 | 20241106-080455,ND,0.18371519056767333,gluonts-NBEATS-1h-10-8736-bt4 70 | 20241106-080455,wQuantileLoss[0.1],0.17943148135957168,gluonts-NBEATS-1h-10-8736-bt4 71 | 20241106-080455,wQuantileLoss[0.2],0.18050240866159714,gluonts-NBEATS-1h-10-8736-bt4 72 | 20241106-080455,wQuantileLoss[0.3],0.18157333596362252,gluonts-NBEATS-1h-10-8736-bt4 73 | 20241106-080455,wQuantileLoss[0.4],0.18264426326564795,gluonts-NBEATS-1h-10-8736-bt4 74 | 20241106-080455,wQuantileLoss[0.5],0.18371519056767333,gluonts-NBEATS-1h-10-8736-bt4 75 | 20241106-080455,wQuantileLoss[0.6],0.18478611786969876,gluonts-NBEATS-1h-10-8736-bt4 76 | 20241106-080455,wQuantileLoss[0.7],0.18585704517172416,gluonts-NBEATS-1h-10-8736-bt4 77 | 20241106-080455,wQuantileLoss[0.8],0.18692797247374957,gluonts-NBEATS-1h-10-8736-bt4 78 | 20241106-080455,wQuantileLoss[0.9],0.18799889977577502,gluonts-NBEATS-1h-10-8736-bt4 79 | 20241106-080455,mean_absolute_QuantileLoss,355277.70033410727,gluonts-NBEATS-1h-10-8736-bt4 80 | 20241106-080455,mean_wQuantileLoss,0.18371519056767335,gluonts-NBEATS-1h-10-8736-bt4 81 | 20241106-080455,MAE_Coverage,0.22371031746031747,gluonts-NBEATS-1h-10-8736-bt4 82 | 20241106-080455,MSE,24120.076428087465,gluonts-GaussianProcess-1h-10-8736-bt4 83 | 20241106-080455,abs_error,378382.5389505496,gluonts-GaussianProcess-1h-10-8736-bt4 84 | 20241106-080455,abs_target_sum,1933850.430311788,gluonts-GaussianProcess-1h-10-8736-bt4 85 | 20241106-080455,abs_target_mean,287.77536165353985,gluonts-GaussianProcess-1h-10-8736-bt4 86 | 20241106-080455,MAPE,0.5678405054544273,gluonts-GaussianProcess-1h-10-8736-bt4 87 | 20241106-080455,sMAPE,0.37280373740447525,gluonts-GaussianProcess-1h-10-8736-bt4 88 | 20241106-080455,num_masked_target_values,0.0,gluonts-GaussianProcess-1h-10-8736-bt4 89 | 20241106-080455,QuantileLoss[0.1],257311.11205047648,gluonts-GaussianProcess-1h-10-8736-bt4 90 | 20241106-080455,Coverage[0.1],0.11696428571428572,gluonts-GaussianProcess-1h-10-8736-bt4 91 | 20241106-080455,QuantileLoss[0.2],332085.40687584557,gluonts-GaussianProcess-1h-10-8736-bt4 92 | 20241106-080455,Coverage[0.2],0.16889880952380953,gluonts-GaussianProcess-1h-10-8736-bt4 93 | 20241106-080455,QuantileLoss[0.3],371906.35510617006,gluonts-GaussianProcess-1h-10-8736-bt4 94 | 20241106-080455,Coverage[0.3],0.2331845238095238,gluonts-GaussianProcess-1h-10-8736-bt4 95 | 20241106-080455,QuantileLoss[0.4],385296.89450299414,gluonts-GaussianProcess-1h-10-8736-bt4 96 | 20241106-080455,Coverage[0.4],0.3171130952380953,gluonts-GaussianProcess-1h-10-8736-bt4 97 | 20241106-080455,QuantileLoss[0.5],378382.5389505496,gluonts-GaussianProcess-1h-10-8736-bt4 98 | 20241106-080455,Coverage[0.5],0.4105654761904762,gluonts-GaussianProcess-1h-10-8736-bt4 99 | 20241106-080455,QuantileLoss[0.6],367635.3604437279,gluonts-GaussianProcess-1h-10-8736-bt4 100 | 20241106-080455,Coverage[0.6],0.4641369047619047,gluonts-GaussianProcess-1h-10-8736-bt4 101 | 20241106-080455,QuantileLoss[0.7],329628.2808167407,gluonts-GaussianProcess-1h-10-8736-bt4 102 | 20241106-080455,Coverage[0.7],0.5686011904761905,gluonts-GaussianProcess-1h-10-8736-bt4 103 | 20241106-080455,QuantileLoss[0.8],270698.6428038415,gluonts-GaussianProcess-1h-10-8736-bt4 104 | 20241106-080455,Coverage[0.8],0.6680059523809524,gluonts-GaussianProcess-1h-10-8736-bt4 105 | 20241106-080455,QuantileLoss[0.9],187117.86916145514,gluonts-GaussianProcess-1h-10-8736-bt4 106 | 20241106-080455,Coverage[0.9],0.7733630952380952,gluonts-GaussianProcess-1h-10-8736-bt4 107 | 20241106-080455,RMSE,155.30639532256058,gluonts-GaussianProcess-1h-10-8736-bt4 108 | 20241106-080455,NRMSE,0.5396792638194576,gluonts-GaussianProcess-1h-10-8736-bt4 109 | 20241106-080455,ND,0.19566277361457798,gluonts-GaussianProcess-1h-10-8736-bt4 110 | 20241106-080455,wQuantileLoss[0.1],0.13305636672686785,gluonts-GaussianProcess-1h-10-8736-bt4 111 | 20241106-080455,wQuantileLoss[0.2],0.1717223843533259,gluonts-GaussianProcess-1h-10-8736-bt4 112 | 20241106-080455,wQuantileLoss[0.3],0.1923139190481287,gluonts-GaussianProcess-1h-10-8736-bt4 113 | 20241106-080455,wQuantileLoss[0.4],0.19923820811771573,gluonts-GaussianProcess-1h-10-8736-bt4 114 | 20241106-080455,wQuantileLoss[0.5],0.19566277361457798,gluonts-GaussianProcess-1h-10-8736-bt4 115 | 20241106-080455,wQuantileLoss[0.6],0.1901053745839358,gluonts-GaussianProcess-1h-10-8736-bt4 116 | 20241106-080455,wQuantileLoss[0.7],0.17045179691771503,gluonts-GaussianProcess-1h-10-8736-bt4 117 | 20241106-080455,wQuantileLoss[0.8],0.139979100017677,gluonts-GaussianProcess-1h-10-8736-bt4 118 | 20241106-080455,wQuantileLoss[0.9],0.09675922513370734,gluonts-GaussianProcess-1h-10-8736-bt4 119 | 20241106-080455,mean_absolute_QuantileLoss,320006.94007908896,gluonts-GaussianProcess-1h-10-8736-bt4 120 | 20241106-080455,mean_wQuantileLoss,0.1654765720570724,gluonts-GaussianProcess-1h-10-8736-bt4 121 | 20241106-080455,MAE_Coverage,0.3074239417989418,gluonts-GaussianProcess-1h-10-8736-bt4 122 | 20241106-080455,MSE,21334.89029071459,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 123 | 20241106-080455,abs_error,396163.6205423127,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 124 | 20241106-080455,abs_target_sum,1933850.430311788,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 125 | 20241106-080455,abs_target_mean,287.77536165353985,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 126 | 20241106-080455,MAPE,0.8344180096876155,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 127 | 20241106-080455,sMAPE,0.4051719308738019,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 128 | 20241106-080455,num_masked_target_values,0.0,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 129 | 20241106-080455,QuantileLoss[0.1],233901.37818849954,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 130 | 20241106-080455,Coverage[0.1],0.18675595238095238,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 131 | 20241106-080455,QuantileLoss[0.2],322680.2048905965,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 132 | 20241106-080455,Coverage[0.2],0.2788690476190476,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 133 | 20241106-080455,QuantileLoss[0.3],378082.01421908627,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 134 | 20241106-080455,Coverage[0.3],0.3461309523809524,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 135 | 20241106-080455,QuantileLoss[0.4],400299.18625961663,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 136 | 20241106-080455,Coverage[0.4],0.4157738095238095,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 137 | 20241106-080455,QuantileLoss[0.5],396163.6205423127,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 138 | 20241106-080455,Coverage[0.5],0.5026785714285714,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 139 | 20241106-080455,QuantileLoss[0.6],371417.69543256215,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 140 | 20241106-080455,Coverage[0.6],0.5976190476190476,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 141 | 20241106-080455,QuantileLoss[0.7],325011.75360091677,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 142 | 20241106-080455,Coverage[0.7],0.6869047619047619,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 143 | 20241106-080455,QuantileLoss[0.8],257238.8791061459,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 144 | 20241106-080455,Coverage[0.8],0.7680059523809524,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 145 | 20241106-080455,QuantileLoss[0.9],165589.1830871597,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 146 | 20241106-080455,Coverage[0.9],0.8327380952380953,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 147 | 20241106-080455,RMSE,146.06467845004346,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 148 | 20241106-080455,NRMSE,0.5075649201195149,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 149 | 20241106-080455,ND,0.20485742554476696,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 150 | 20241106-080455,wQuantileLoss[0.1],0.12095112141159149,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 151 | 20241106-080455,wQuantileLoss[0.2],0.1668589255057186,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 152 | 20241106-080455,wQuantileLoss[0.3],0.19550737135246254,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 153 | 20241106-080455,wQuantileLoss[0.4],0.20699593928527232,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 154 | 20241106-080455,wQuantileLoss[0.5],0.20485742554476696,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 155 | 20241106-080455,wQuantileLoss[0.6],0.19206123162931465,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 156 | 20241106-080455,wQuantileLoss[0.7],0.16806457650838913,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 157 | 20241106-080455,wQuantileLoss[0.8],0.13301901484939152,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 158 | 20241106-080455,wQuantileLoss[0.9],0.08562667540967081,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 159 | 20241106-080455,mean_absolute_QuantileLoss,316709.3239252107,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 160 | 20241106-080455,mean_wQuantileLoss,0.1637713646107309,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 161 | 20241106-080455,MAE_Coverage,0.3476851851851852,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 162 | 20241106-080455,MSE,20453.903226090588,gluonts-MQCNN-1h-10-8736-bt4 163 | 20241106-080455,abs_error,324250.87280253635,gluonts-MQCNN-1h-10-8736-bt4 164 | 20241106-080455,abs_target_sum,1933850.430311788,gluonts-MQCNN-1h-10-8736-bt4 165 | 20241106-080455,abs_target_mean,287.77536165353985,gluonts-MQCNN-1h-10-8736-bt4 166 | 20241106-080455,MAPE,0.533053060136319,gluonts-MQCNN-1h-10-8736-bt4 167 | 20241106-080455,sMAPE,0.36282727055510205,gluonts-MQCNN-1h-10-8736-bt4 168 | 20241106-080455,num_masked_target_values,0.0,gluonts-MQCNN-1h-10-8736-bt4 169 | 20241106-080455,QuantileLoss[0.1],242852.0840527105,gluonts-MQCNN-1h-10-8736-bt4 170 | 20241106-080455,Coverage[0.1],0.15907738095238094,gluonts-MQCNN-1h-10-8736-bt4 171 | 20241106-080455,QuantileLoss[0.2],311015.5716887718,gluonts-MQCNN-1h-10-8736-bt4 172 | 20241106-080455,Coverage[0.2],0.28288690476190476,gluonts-MQCNN-1h-10-8736-bt4 173 | 20241106-080455,QuantileLoss[0.3],322810.83915022,gluonts-MQCNN-1h-10-8736-bt4 174 | 20241106-080455,Coverage[0.3],0.30297619047619045,gluonts-MQCNN-1h-10-8736-bt4 175 | 20241106-080455,QuantileLoss[0.4],331178.3973383634,gluonts-MQCNN-1h-10-8736-bt4 176 | 20241106-080455,Coverage[0.4],0.4001488095238096,gluonts-MQCNN-1h-10-8736-bt4 177 | 20241106-080455,QuantileLoss[0.5],324250.87280253635,gluonts-MQCNN-1h-10-8736-bt4 178 | 20241106-080455,Coverage[0.5],0.5504464285714287,gluonts-MQCNN-1h-10-8736-bt4 179 | 20241106-080455,QuantileLoss[0.6],297429.26302844606,gluonts-MQCNN-1h-10-8736-bt4 180 | 20241106-080455,Coverage[0.6],0.5703869047619048,gluonts-MQCNN-1h-10-8736-bt4 181 | 20241106-080455,QuantileLoss[0.7],269201.9485019067,gluonts-MQCNN-1h-10-8736-bt4 182 | 20241106-080455,Coverage[0.7],0.7178571428571429,gluonts-MQCNN-1h-10-8736-bt4 183 | 20241106-080455,QuantileLoss[0.8],216633.78681745386,gluonts-MQCNN-1h-10-8736-bt4 184 | 20241106-080455,Coverage[0.8],0.8019345238095237,gluonts-MQCNN-1h-10-8736-bt4 185 | 20241106-080455,QuantileLoss[0.9],142342.50592784095,gluonts-MQCNN-1h-10-8736-bt4 186 | 20241106-080455,Coverage[0.9],0.8706845238095238,gluonts-MQCNN-1h-10-8736-bt4 187 | 20241106-080455,RMSE,143.01714311959455,gluonts-MQCNN-1h-10-8736-bt4 188 | 20241106-080455,NRMSE,0.4969749400985084,gluonts-MQCNN-1h-10-8736-bt4 189 | 20241106-080455,ND,0.16767112270945303,gluonts-MQCNN-1h-10-8736-bt4 190 | 20241106-080455,wQuantileLoss[0.1],0.12557955891839903,gluonts-MQCNN-1h-10-8736-bt4 191 | 20241106-080455,wQuantileLoss[0.2],0.1608271078330643,gluonts-MQCNN-1h-10-8736-bt4 192 | 20241106-080455,wQuantileLoss[0.3],0.166926476882793,gluonts-MQCNN-1h-10-8736-bt4 193 | 20241106-080455,wQuantileLoss[0.4],0.17125336693436455,gluonts-MQCNN-1h-10-8736-bt4 194 | 20241106-080455,wQuantileLoss[0.5],0.16767112270945303,gluonts-MQCNN-1h-10-8736-bt4 195 | 20241106-080455,wQuantileLoss[0.6],0.1538015858757456,gluonts-MQCNN-1h-10-8736-bt4 196 | 20241106-080455,wQuantileLoss[0.7],0.13920515479498805,gluonts-MQCNN-1h-10-8736-bt4 197 | 20241106-080455,wQuantileLoss[0.8],0.1120219968524281,gluonts-MQCNN-1h-10-8736-bt4 198 | 20241106-080455,wQuantileLoss[0.9],0.07360574721639229,gluonts-MQCNN-1h-10-8736-bt4 199 | 20241106-080455,mean_absolute_QuantileLoss,273079.4743675833,gluonts-MQCNN-1h-10-8736-bt4 200 | 20241106-080455,mean_wQuantileLoss,0.14121023533529198,gluonts-MQCNN-1h-10-8736-bt4 201 | 20241106-080455,MAE_Coverage,0.37719907407407405,gluonts-MQCNN-1h-10-8736-bt4 202 | 20241106-080455,MSE,120626.77846042384,gluonts-MQRNN-1h-10-8736-bt4 203 | 20241106-080455,abs_error,1708696.9935996654,gluonts-MQRNN-1h-10-8736-bt4 204 | 20241106-080455,abs_target_sum,1933850.430311788,gluonts-MQRNN-1h-10-8736-bt4 205 | 20241106-080455,abs_target_mean,287.77536165353985,gluonts-MQRNN-1h-10-8736-bt4 206 | 20241106-080455,MAPE,4.642684025336889,gluonts-MQRNN-1h-10-8736-bt4 207 | 20241106-080455,sMAPE,1.0497421392402313,gluonts-MQRNN-1h-10-8736-bt4 208 | 20241106-080455,num_masked_target_values,0.0,gluonts-MQRNN-1h-10-8736-bt4 209 | 20241106-080455,QuantileLoss[0.1],469208.3396337308,gluonts-MQRNN-1h-10-8736-bt4 210 | 20241106-080455,Coverage[0.1],0.41636904761904764,gluonts-MQRNN-1h-10-8736-bt4 211 | 20241106-080455,QuantileLoss[0.2],923095.8049626815,gluonts-MQRNN-1h-10-8736-bt4 212 | 20241106-080455,Coverage[0.2],0.601190476190476,gluonts-MQRNN-1h-10-8736-bt4 213 | 20241106-080455,QuantileLoss[0.3],1297062.3036221645,gluonts-MQRNN-1h-10-8736-bt4 214 | 20241106-080455,Coverage[0.3],0.7607142857142858,gluonts-MQRNN-1h-10-8736-bt4 215 | 20241106-080455,QuantileLoss[0.4],1590916.9085901747,gluonts-MQRNN-1h-10-8736-bt4 216 | 20241106-080455,Coverage[0.4],0.8248511904761905,gluonts-MQRNN-1h-10-8736-bt4 217 | 20241106-080455,QuantileLoss[0.5],1708696.9935996654,gluonts-MQRNN-1h-10-8736-bt4 218 | 20241106-080455,Coverage[0.5],0.8410714285714285,gluonts-MQRNN-1h-10-8736-bt4 219 | 20241106-080455,QuantileLoss[0.6],1661523.812853003,gluonts-MQRNN-1h-10-8736-bt4 220 | 20241106-080455,Coverage[0.6],0.8644345238095237,gluonts-MQRNN-1h-10-8736-bt4 221 | 20241106-080455,QuantileLoss[0.7],1507215.6589001382,gluonts-MQRNN-1h-10-8736-bt4 222 | 20241106-080455,Coverage[0.7],0.9092261904761905,gluonts-MQRNN-1h-10-8736-bt4 223 | 20241106-080455,QuantileLoss[0.8],1242963.4756011772,gluonts-MQRNN-1h-10-8736-bt4 224 | 20241106-080455,Coverage[0.8],0.9325892857142858,gluonts-MQRNN-1h-10-8736-bt4 225 | 20241106-080455,QuantileLoss[0.9],806353.9466135174,gluonts-MQRNN-1h-10-8736-bt4 226 | 20241106-080455,Coverage[0.9],0.9461309523809524,gluonts-MQRNN-1h-10-8736-bt4 227 | 20241106-080455,RMSE,347.31366005445835,gluonts-MQRNN-1h-10-8736-bt4 228 | 20241106-080455,NRMSE,1.2068915770231858,gluonts-MQRNN-1h-10-8736-bt4 229 | 20241106-080455,ND,0.8835724660071969,gluonts-MQRNN-1h-10-8736-bt4 230 | 20241106-080455,wQuantileLoss[0.1],0.24262907424442434,gluonts-MQRNN-1h-10-8736-bt4 231 | 20241106-080455,wQuantileLoss[0.2],0.4773356773066746,gluonts-MQRNN-1h-10-8736-bt4 232 | 20241106-080455,wQuantileLoss[0.3],0.6707149029167905,gluonts-MQRNN-1h-10-8736-bt4 233 | 20241106-080455,wQuantileLoss[0.4],0.8226680221249979,gluonts-MQRNN-1h-10-8736-bt4 234 | 20241106-080455,wQuantileLoss[0.5],0.8835724660071969,gluonts-MQRNN-1h-10-8736-bt4 235 | 20241106-080455,wQuantileLoss[0.6],0.8591790692856847,gluonts-MQRNN-1h-10-8736-bt4 236 | 20241106-080455,wQuantileLoss[0.7],0.7793858487065801,gluonts-MQRNN-1h-10-8736-bt4 237 | 20241106-080455,wQuantileLoss[0.8],0.6427402327080582,gluonts-MQRNN-1h-10-8736-bt4 238 | 20241106-080455,wQuantileLoss[0.9],0.41696810362087394,gluonts-MQRNN-1h-10-8736-bt4 239 | 20241106-080455,mean_absolute_QuantileLoss,1245226.3604862504,gluonts-MQRNN-1h-10-8736-bt4 240 | 20241106-080455,mean_wQuantileLoss,0.6439103774356979,gluonts-MQRNN-1h-10-8736-bt4 241 | 20241106-080455,MAE_Coverage,0.4461309523809523,gluonts-MQRNN-1h-10-8736-bt4 242 | 20241106-080455,MSE,23963.740487498282,gluonts-SeasonalNaive-1h-10-8736-bt4 243 | 20241106-080455,abs_error,353611.9989626674,gluonts-SeasonalNaive-1h-10-8736-bt4 244 | 20241106-080455,abs_target_sum,1933850.430311788,gluonts-SeasonalNaive-1h-10-8736-bt4 245 | 20241106-080455,abs_target_mean,287.77536165353985,gluonts-SeasonalNaive-1h-10-8736-bt4 246 | 20241106-080455,MAPE,0.5920573391566246,gluonts-SeasonalNaive-1h-10-8736-bt4 247 | 20241106-080455,sMAPE,0.3103506820879077,gluonts-SeasonalNaive-1h-10-8736-bt4 248 | 20241106-080455,num_masked_target_values,0.0,gluonts-SeasonalNaive-1h-10-8736-bt4 249 | 20241106-080455,QuantileLoss[0.1],482413.70590227813,gluonts-SeasonalNaive-1h-10-8736-bt4 250 | 20241106-080455,Coverage[0.1],0.5431547619047619,gluonts-SeasonalNaive-1h-10-8736-bt4 251 | 20241106-080455,QuantileLoss[0.2],450213.27916737547,gluonts-SeasonalNaive-1h-10-8736-bt4 252 | 20241106-080455,Coverage[0.2],0.5431547619047619,gluonts-SeasonalNaive-1h-10-8736-bt4 253 | 20241106-080455,QuantileLoss[0.3],418012.85243247275,gluonts-SeasonalNaive-1h-10-8736-bt4 254 | 20241106-080455,Coverage[0.3],0.5431547619047619,gluonts-SeasonalNaive-1h-10-8736-bt4 255 | 20241106-080455,QuantileLoss[0.4],385812.4256975701,gluonts-SeasonalNaive-1h-10-8736-bt4 256 | 20241106-080455,Coverage[0.4],0.5431547619047619,gluonts-SeasonalNaive-1h-10-8736-bt4 257 | 20241106-080455,QuantileLoss[0.5],353611.9989626674,gluonts-SeasonalNaive-1h-10-8736-bt4 258 | 20241106-080455,Coverage[0.5],0.5431547619047619,gluonts-SeasonalNaive-1h-10-8736-bt4 259 | 20241106-080455,QuantileLoss[0.6],321411.57222776476,gluonts-SeasonalNaive-1h-10-8736-bt4 260 | 20241106-080455,Coverage[0.6],0.5431547619047619,gluonts-SeasonalNaive-1h-10-8736-bt4 261 | 20241106-080455,QuantileLoss[0.7],289211.1454928621,gluonts-SeasonalNaive-1h-10-8736-bt4 262 | 20241106-080455,Coverage[0.7],0.5431547619047619,gluonts-SeasonalNaive-1h-10-8736-bt4 263 | 20241106-080455,QuantileLoss[0.8],257010.71875795937,gluonts-SeasonalNaive-1h-10-8736-bt4 264 | 20241106-080455,Coverage[0.8],0.5431547619047619,gluonts-SeasonalNaive-1h-10-8736-bt4 265 | 20241106-080455,QuantileLoss[0.9],224810.2920230567,gluonts-SeasonalNaive-1h-10-8736-bt4 266 | 20241106-080455,Coverage[0.9],0.5431547619047619,gluonts-SeasonalNaive-1h-10-8736-bt4 267 | 20241106-080455,RMSE,154.80226253998447,gluonts-SeasonalNaive-1h-10-8736-bt4 268 | 20241106-080455,NRMSE,0.5379274363534807,gluonts-SeasonalNaive-1h-10-8736-bt4 269 | 20241106-080455,ND,0.18285385127000528,gluonts-SeasonalNaive-1h-10-8736-bt4 270 | 20241106-080455,wQuantileLoss[0.1],0.2494576097203651,gluonts-SeasonalNaive-1h-10-8736-bt4 271 | 20241106-080455,wQuantileLoss[0.2],0.23280667010777514,gluonts-SeasonalNaive-1h-10-8736-bt4 272 | 20241106-080455,wQuantileLoss[0.3],0.21615573049518516,gluonts-SeasonalNaive-1h-10-8736-bt4 273 | 20241106-080455,wQuantileLoss[0.4],0.19950479088259523,gluonts-SeasonalNaive-1h-10-8736-bt4 274 | 20241106-080455,wQuantileLoss[0.5],0.18285385127000528,gluonts-SeasonalNaive-1h-10-8736-bt4 275 | 20241106-080455,wQuantileLoss[0.6],0.16620291165741535,gluonts-SeasonalNaive-1h-10-8736-bt4 276 | 20241106-080455,wQuantileLoss[0.7],0.1495519720448254,gluonts-SeasonalNaive-1h-10-8736-bt4 277 | 20241106-080455,wQuantileLoss[0.8],0.13290103243223542,gluonts-SeasonalNaive-1h-10-8736-bt4 278 | 20241106-080455,wQuantileLoss[0.9],0.11625009281964548,gluonts-SeasonalNaive-1h-10-8736-bt4 279 | 20241106-080455,mean_absolute_QuantileLoss,353611.9989626674,gluonts-SeasonalNaive-1h-10-8736-bt4 280 | 20241106-080455,mean_wQuantileLoss,0.18285385127000528,gluonts-SeasonalNaive-1h-10-8736-bt4 281 | 20241106-080455,MAE_Coverage,0.22701719576719576,gluonts-SeasonalNaive-1h-10-8736-bt4 282 | 20241106-080455,MSE,21392.03482248147,gluonts-Prophet-1h-10-8736-bt4 283 | 20241106-080455,abs_error,376606.29346479836,gluonts-Prophet-1h-10-8736-bt4 284 | 20241106-080455,abs_target_sum,1933850.430311788,gluonts-Prophet-1h-10-8736-bt4 285 | 20241106-080455,abs_target_mean,287.77536165353985,gluonts-Prophet-1h-10-8736-bt4 286 | 20241106-080455,MAPE,0.6738884865652881,gluonts-Prophet-1h-10-8736-bt4 287 | 20241106-080455,sMAPE,0.37342936957494466,gluonts-Prophet-1h-10-8736-bt4 288 | 20241106-080455,num_masked_target_values,0.0,gluonts-Prophet-1h-10-8736-bt4 289 | 20241106-080455,QuantileLoss[0.1],244542.52854257333,gluonts-Prophet-1h-10-8736-bt4 290 | 20241106-080455,Coverage[0.1],0.16979166666666667,gluonts-Prophet-1h-10-8736-bt4 291 | 20241106-080455,QuantileLoss[0.2],310879.564058149,gluonts-Prophet-1h-10-8736-bt4 292 | 20241106-080455,Coverage[0.2],0.25982142857142854,gluonts-Prophet-1h-10-8736-bt4 293 | 20241106-080455,QuantileLoss[0.3],353786.1223710363,gluonts-Prophet-1h-10-8736-bt4 294 | 20241106-080455,Coverage[0.3],0.35089285714285723,gluonts-Prophet-1h-10-8736-bt4 295 | 20241106-080455,QuantileLoss[0.4],374364.71891049703,gluonts-Prophet-1h-10-8736-bt4 296 | 20241106-080455,Coverage[0.4],0.4392857142857142,gluonts-Prophet-1h-10-8736-bt4 297 | 20241106-080455,QuantileLoss[0.5],376606.29346479836,gluonts-Prophet-1h-10-8736-bt4 298 | 20241106-080455,Coverage[0.5],0.5199404761904762,gluonts-Prophet-1h-10-8736-bt4 299 | 20241106-080455,QuantileLoss[0.6],362616.94570055755,gluonts-Prophet-1h-10-8736-bt4 300 | 20241106-080455,Coverage[0.6],0.5547619047619048,gluonts-Prophet-1h-10-8736-bt4 301 | 20241106-080455,QuantileLoss[0.7],329327.2961050865,gluonts-Prophet-1h-10-8736-bt4 302 | 20241106-080455,Coverage[0.7],0.6232142857142857,gluonts-Prophet-1h-10-8736-bt4 303 | 20241106-080455,QuantileLoss[0.8],273821.4479207531,gluonts-Prophet-1h-10-8736-bt4 304 | 20241106-080455,Coverage[0.8],0.6933035714285715,gluonts-Prophet-1h-10-8736-bt4 305 | 20241106-080455,QuantileLoss[0.9],193635.45118206405,gluonts-Prophet-1h-10-8736-bt4 306 | 20241106-080455,Coverage[0.9],0.7703869047619047,gluonts-Prophet-1h-10-8736-bt4 307 | 20241106-080455,RMSE,146.2601614332538,gluonts-Prophet-1h-10-8736-bt4 308 | 20241106-080455,NRMSE,0.5082442103203407,gluonts-Prophet-1h-10-8736-bt4 309 | 20241106-080455,ND,0.19474427161571095,gluonts-Prophet-1h-10-8736-bt4 310 | 20241106-080455,wQuantileLoss[0.1],0.1264536929586362,gluonts-Prophet-1h-10-8736-bt4 311 | 20241106-080455,wQuantileLoss[0.2],0.16075677786933448,gluonts-Prophet-1h-10-8736-bt4 312 | 20241106-080455,wQuantileLoss[0.3],0.1829438910195328,gluonts-Prophet-1h-10-8736-bt4 313 | 20241106-080455,wQuantileLoss[0.4],0.19358514652559738,gluonts-Prophet-1h-10-8736-bt4 314 | 20241106-080455,wQuantileLoss[0.5],0.19474427161571095,gluonts-Prophet-1h-10-8736-bt4 315 | 20241106-080455,wQuantileLoss[0.6],0.18751033689926788,gluonts-Prophet-1h-10-8736-bt4 316 | 20241106-080455,wQuantileLoss[0.7],0.17029615679842944,gluonts-Prophet-1h-10-8736-bt4 317 | 20241106-080455,wQuantileLoss[0.8],0.14159391213963005,gluonts-Prophet-1h-10-8736-bt4 318 | 20241106-080455,wQuantileLoss[0.9],0.10012948682429637,gluonts-Prophet-1h-10-8736-bt4 319 | 20241106-080455,mean_absolute_QuantileLoss,313286.7075839461,gluonts-Prophet-1h-10-8736-bt4 320 | 20241106-080455,mean_wQuantileLoss,0.16200151918338174,gluonts-Prophet-1h-10-8736-bt4 321 | 20241106-080455,MAE_Coverage,0.30577050264550265,gluonts-Prophet-1h-10-8736-bt4 322 | 20241106-080455,MSE,19185.081160918926,gluonts-NPTS-1h-10-8736-bt4 323 | 20241106-080455,abs_error,349404.60605022905,gluonts-NPTS-1h-10-8736-bt4 324 | 20241106-080455,abs_target_sum,1933850.430311788,gluonts-NPTS-1h-10-8736-bt4 325 | 20241106-080455,abs_target_mean,287.77536165353985,gluonts-NPTS-1h-10-8736-bt4 326 | 20241106-080455,MAPE,0.8968081786639033,gluonts-NPTS-1h-10-8736-bt4 327 | 20241106-080455,sMAPE,0.3171503899298057,gluonts-NPTS-1h-10-8736-bt4 328 | 20241106-080455,num_masked_target_values,0.0,gluonts-NPTS-1h-10-8736-bt4 329 | 20241106-080455,QuantileLoss[0.1],198787.71156291745,gluonts-NPTS-1h-10-8736-bt4 330 | 20241106-080455,Coverage[0.1],0.21294642857142856,gluonts-NPTS-1h-10-8736-bt4 331 | 20241106-080455,QuantileLoss[0.2],276528.5470673586,gluonts-NPTS-1h-10-8736-bt4 332 | 20241106-080455,Coverage[0.2],0.28377976190476195,gluonts-NPTS-1h-10-8736-bt4 333 | 20241106-080455,QuantileLoss[0.3],328469.0184515643,gluonts-NPTS-1h-10-8736-bt4 334 | 20241106-080455,Coverage[0.3],0.3453869047619048,gluonts-NPTS-1h-10-8736-bt4 335 | 20241106-080455,QuantileLoss[0.4],351305.5860314814,gluonts-NPTS-1h-10-8736-bt4 336 | 20241106-080455,Coverage[0.4],0.4058035714285714,gluonts-NPTS-1h-10-8736-bt4 337 | 20241106-080455,QuantileLoss[0.5],349404.60605022905,gluonts-NPTS-1h-10-8736-bt4 338 | 20241106-080455,Coverage[0.5],0.4711309523809524,gluonts-NPTS-1h-10-8736-bt4 339 | 20241106-080455,QuantileLoss[0.6],329019.6566468009,gluonts-NPTS-1h-10-8736-bt4 340 | 20241106-080455,Coverage[0.6],0.5127976190476191,gluonts-NPTS-1h-10-8736-bt4 341 | 20241106-080455,QuantileLoss[0.7],293654.63564441935,gluonts-NPTS-1h-10-8736-bt4 342 | 20241106-080455,Coverage[0.7],0.5950892857142858,gluonts-NPTS-1h-10-8736-bt4 343 | 20241106-080455,QuantileLoss[0.8],238033.35188405242,gluonts-NPTS-1h-10-8736-bt4 344 | 20241106-080455,Coverage[0.8],0.6949404761904763,gluonts-NPTS-1h-10-8736-bt4 345 | 20241106-080455,QuantileLoss[0.9],149487.60377643866,gluonts-NPTS-1h-10-8736-bt4 346 | 20241106-080455,Coverage[0.9],0.7967261904761905,gluonts-NPTS-1h-10-8736-bt4 347 | 20241106-080455,RMSE,138.5102204204402,gluonts-NPTS-1h-10-8736-bt4 348 | 20241106-080455,NRMSE,0.4813136872613723,gluonts-NPTS-1h-10-8736-bt4 349 | 20241106-080455,ND,0.18067819546617975,gluonts-NPTS-1h-10-8736-bt4 350 | 20241106-080455,wQuantileLoss[0.1],0.10279373649950145,gluonts-NPTS-1h-10-8736-bt4 351 | 20241106-080455,wQuantileLoss[0.2],0.14299376142692424,gluonts-NPTS-1h-10-8736-bt4 352 | 20241106-080455,wQuantileLoss[0.3],0.16985233878641087,gluonts-NPTS-1h-10-8736-bt4 353 | 20241106-080455,wQuantileLoss[0.4],0.18166119805596423,gluonts-NPTS-1h-10-8736-bt4 354 | 20241106-080455,wQuantileLoss[0.5],0.18067819546617975,gluonts-NPTS-1h-10-8736-bt4 355 | 20241106-080455,wQuantileLoss[0.6],0.17013707548921156,gluonts-NPTS-1h-10-8736-bt4 356 | 20241106-080455,wQuantileLoss[0.7],0.15184971445649725,gluonts-NPTS-1h-10-8736-bt4 357 | 20241106-080455,wQuantileLoss[0.8],0.12308777770661153,gluonts-NPTS-1h-10-8736-bt4 358 | 20241106-080455,wQuantileLoss[0.9],0.0773004992699137,gluonts-NPTS-1h-10-8736-bt4 359 | 20241106-080455,mean_absolute_QuantileLoss,279410.07967947354,gluonts-NPTS-1h-10-8736-bt4 360 | 20241106-080455,mean_wQuantileLoss,0.14448381079524608,gluonts-NPTS-1h-10-8736-bt4 361 | 20241106-080455,MAE_Coverage,0.3204034391534392,gluonts-NPTS-1h-10-8736-bt4 362 | -------------------------------------------------------------------------------- /test/model-performance/gluonts-1h-10-8736-bt4-20241106-181402.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20241106-181402,MSE,2775.3851223107026,gluonts-SimpleFeedForward-1h-10-8736-bt4 3 | 20241106-181402,abs_error,160010.71253004688,gluonts-SimpleFeedForward-1h-10-8736-bt4 4 | 20241106-181402,abs_target_sum,1060363.985675315,gluonts-SimpleFeedForward-1h-10-8736-bt4 5 | 20241106-181402,abs_target_mean,157.79225977311233,gluonts-SimpleFeedForward-1h-10-8736-bt4 6 | 20241106-181402,MAPE,0.25801824711235377,gluonts-SimpleFeedForward-1h-10-8736-bt4 7 | 20241106-181402,sMAPE,0.20219454907019324,gluonts-SimpleFeedForward-1h-10-8736-bt4 8 | 20241106-181402,num_masked_target_values,0.0,gluonts-SimpleFeedForward-1h-10-8736-bt4 9 | 20241106-181402,QuantileLoss[0.1],103931.47809452907,gluonts-SimpleFeedForward-1h-10-8736-bt4 10 | 20241106-181402,Coverage[0.1],0.14122023809523807,gluonts-SimpleFeedForward-1h-10-8736-bt4 11 | 20241106-181402,QuantileLoss[0.2],133939.34989503742,gluonts-SimpleFeedForward-1h-10-8736-bt4 12 | 20241106-181402,Coverage[0.2],0.23258928571428578,gluonts-SimpleFeedForward-1h-10-8736-bt4 13 | 20241106-181402,QuantileLoss[0.3],149539.81611929752,gluonts-SimpleFeedForward-1h-10-8736-bt4 14 | 20241106-181402,Coverage[0.3],0.3339285714285714,gluonts-SimpleFeedForward-1h-10-8736-bt4 15 | 20241106-181402,QuantileLoss[0.4],157514.49414235397,gluonts-SimpleFeedForward-1h-10-8736-bt4 16 | 20241106-181402,Coverage[0.4],0.4386904761904762,gluonts-SimpleFeedForward-1h-10-8736-bt4 17 | 20241106-181402,QuantileLoss[0.5],160010.71253004688,gluonts-SimpleFeedForward-1h-10-8736-bt4 18 | 20241106-181402,Coverage[0.5],0.5433035714285714,gluonts-SimpleFeedForward-1h-10-8736-bt4 19 | 20241106-181402,QuantileLoss[0.6],153263.1297431849,gluonts-SimpleFeedForward-1h-10-8736-bt4 20 | 20241106-181402,Coverage[0.6],0.5919642857142857,gluonts-SimpleFeedForward-1h-10-8736-bt4 21 | 20241106-181402,QuantileLoss[0.7],140805.00303752205,gluonts-SimpleFeedForward-1h-10-8736-bt4 22 | 20241106-181402,Coverage[0.7],0.6953869047619048,gluonts-SimpleFeedForward-1h-10-8736-bt4 23 | 20241106-181402,QuantileLoss[0.8],119143.75565828788,gluonts-SimpleFeedForward-1h-10-8736-bt4 24 | 20241106-181402,Coverage[0.8],0.7918154761904761,gluonts-SimpleFeedForward-1h-10-8736-bt4 25 | 20241106-181402,QuantileLoss[0.9],85146.83691103829,gluonts-SimpleFeedForward-1h-10-8736-bt4 26 | 20241106-181402,Coverage[0.9],0.8867559523809525,gluonts-SimpleFeedForward-1h-10-8736-bt4 27 | 20241106-181402,RMSE,52.681924056650615,gluonts-SimpleFeedForward-1h-10-8736-bt4 28 | 20241106-181402,NRMSE,0.3338688737483153,gluonts-SimpleFeedForward-1h-10-8736-bt4 29 | 20241106-181402,ND,0.1509016853567888,gluonts-SimpleFeedForward-1h-10-8736-bt4 30 | 20241106-181402,wQuantileLoss[0.1],0.09801490761527339,gluonts-SimpleFeedForward-1h-10-8736-bt4 31 | 20241106-181402,wQuantileLoss[0.2],0.1263145030427786,gluonts-SimpleFeedForward-1h-10-8736-bt4 32 | 20241106-181402,wQuantileLoss[0.3],0.14102687203588868,gluonts-SimpleFeedForward-1h-10-8736-bt4 33 | 20241106-181402,wQuantileLoss[0.4],0.14854757071180383,gluonts-SimpleFeedForward-1h-10-8736-bt4 34 | 20241106-181402,wQuantileLoss[0.5],0.1509016853567888,gluonts-SimpleFeedForward-1h-10-8736-bt4 35 | 20241106-181402,wQuantileLoss[0.6],0.14453822632006505,gluonts-SimpleFeedForward-1h-10-8736-bt4 36 | 20241106-181402,wQuantileLoss[0.7],0.13278931097216345,gluonts-SimpleFeedForward-1h-10-8736-bt4 37 | 20241106-181402,wQuantileLoss[0.8],0.11236118659990955,gluonts-SimpleFeedForward-1h-10-8736-bt4 38 | 20241106-181402,wQuantileLoss[0.9],0.08029963112790063,gluonts-SimpleFeedForward-1h-10-8736-bt4 39 | 20241106-181402,mean_absolute_QuantileLoss,133699.397347922,gluonts-SimpleFeedForward-1h-10-8736-bt4 40 | 20241106-181402,mean_wQuantileLoss,0.12608821042028578,gluonts-SimpleFeedForward-1h-10-8736-bt4 41 | 20241106-181402,MAE_Coverage,0.38969907407407417,gluonts-SimpleFeedForward-1h-10-8736-bt4 42 | 20241106-181402,MSE,1950.4088925865894,gluonts-NBEATS-1h-10-8736-bt4 43 | 20241106-181402,abs_error,158319.2642732688,gluonts-NBEATS-1h-10-8736-bt4 44 | 20241106-181402,abs_target_sum,1060363.985675315,gluonts-NBEATS-1h-10-8736-bt4 45 | 20241106-181402,abs_target_mean,157.79225977311233,gluonts-NBEATS-1h-10-8736-bt4 46 | 20241106-181402,MAPE,0.2402972222950998,gluonts-NBEATS-1h-10-8736-bt4 47 | 20241106-181402,sMAPE,0.21527708453741096,gluonts-NBEATS-1h-10-8736-bt4 48 | 20241106-181402,num_masked_target_values,0.0,gluonts-NBEATS-1h-10-8736-bt4 49 | 20241106-181402,QuantileLoss[0.1],167958.68235580812,gluonts-NBEATS-1h-10-8736-bt4 50 | 20241106-181402,Coverage[0.1],0.46964285714285714,gluonts-NBEATS-1h-10-8736-bt4 51 | 20241106-181402,QuantileLoss[0.2],165548.82783517332,gluonts-NBEATS-1h-10-8736-bt4 52 | 20241106-181402,Coverage[0.2],0.46964285714285714,gluonts-NBEATS-1h-10-8736-bt4 53 | 20241106-181402,QuantileLoss[0.3],163138.97331453845,gluonts-NBEATS-1h-10-8736-bt4 54 | 20241106-181402,Coverage[0.3],0.46964285714285714,gluonts-NBEATS-1h-10-8736-bt4 55 | 20241106-181402,QuantileLoss[0.4],160729.11879390362,gluonts-NBEATS-1h-10-8736-bt4 56 | 20241106-181402,Coverage[0.4],0.46964285714285714,gluonts-NBEATS-1h-10-8736-bt4 57 | 20241106-181402,QuantileLoss[0.5],158319.2642732688,gluonts-NBEATS-1h-10-8736-bt4 58 | 20241106-181402,Coverage[0.5],0.46964285714285714,gluonts-NBEATS-1h-10-8736-bt4 59 | 20241106-181402,QuantileLoss[0.6],155909.40975263398,gluonts-NBEATS-1h-10-8736-bt4 60 | 20241106-181402,Coverage[0.6],0.46964285714285714,gluonts-NBEATS-1h-10-8736-bt4 61 | 20241106-181402,QuantileLoss[0.7],153499.55523199914,gluonts-NBEATS-1h-10-8736-bt4 62 | 20241106-181402,Coverage[0.7],0.46964285714285714,gluonts-NBEATS-1h-10-8736-bt4 63 | 20241106-181402,QuantileLoss[0.8],151089.7007113643,gluonts-NBEATS-1h-10-8736-bt4 64 | 20241106-181402,Coverage[0.8],0.46964285714285714,gluonts-NBEATS-1h-10-8736-bt4 65 | 20241106-181402,QuantileLoss[0.9],148679.8461907295,gluonts-NBEATS-1h-10-8736-bt4 66 | 20241106-181402,Coverage[0.9],0.46964285714285714,gluonts-NBEATS-1h-10-8736-bt4 67 | 20241106-181402,RMSE,44.163433885813156,gluonts-NBEATS-1h-10-8736-bt4 68 | 20241106-181402,NRMSE,0.27988339826880765,gluonts-NBEATS-1h-10-8736-bt4 69 | 20241106-181402,ND,0.14930652720390147,gluonts-NBEATS-1h-10-8736-bt4 70 | 20241106-181402,wQuantileLoss[0.1],0.15839719626920387,gluonts-NBEATS-1h-10-8736-bt4 71 | 20241106-181402,wQuantileLoss[0.2],0.1561245290028783,gluonts-NBEATS-1h-10-8736-bt4 72 | 20241106-181402,wQuantileLoss[0.3],0.15385186173655266,gluonts-NBEATS-1h-10-8736-bt4 73 | 20241106-181402,wQuantileLoss[0.4],0.15157919447022705,gluonts-NBEATS-1h-10-8736-bt4 74 | 20241106-181402,wQuantileLoss[0.5],0.14930652720390147,gluonts-NBEATS-1h-10-8736-bt4 75 | 20241106-181402,wQuantileLoss[0.6],0.1470338599375759,gluonts-NBEATS-1h-10-8736-bt4 76 | 20241106-181402,wQuantileLoss[0.7],0.1447611926712503,gluonts-NBEATS-1h-10-8736-bt4 77 | 20241106-181402,wQuantileLoss[0.8],0.14248852540492468,gluonts-NBEATS-1h-10-8736-bt4 78 | 20241106-181402,wQuantileLoss[0.9],0.1402158581385991,gluonts-NBEATS-1h-10-8736-bt4 79 | 20241106-181402,mean_absolute_QuantileLoss,158319.26427326878,gluonts-NBEATS-1h-10-8736-bt4 80 | 20241106-181402,mean_wQuantileLoss,0.14930652720390147,gluonts-NBEATS-1h-10-8736-bt4 81 | 20241106-181402,MAE_Coverage,0.2255952380952381,gluonts-NBEATS-1h-10-8736-bt4 82 | 20241106-181402,MSE,2680.7500844596952,gluonts-GaussianProcess-1h-10-8736-bt4 83 | 20241106-181402,abs_error,195038.23047864335,gluonts-GaussianProcess-1h-10-8736-bt4 84 | 20241106-181402,abs_target_sum,1060363.985675315,gluonts-GaussianProcess-1h-10-8736-bt4 85 | 20241106-181402,abs_target_mean,157.79225977311233,gluonts-GaussianProcess-1h-10-8736-bt4 86 | 20241106-181402,MAPE,0.3020655674338906,gluonts-GaussianProcess-1h-10-8736-bt4 87 | 20241106-181402,sMAPE,0.24901002795281907,gluonts-GaussianProcess-1h-10-8736-bt4 88 | 20241106-181402,num_masked_target_values,0.0,gluonts-GaussianProcess-1h-10-8736-bt4 89 | 20241106-181402,QuantileLoss[0.1],108122.26556972912,gluonts-GaussianProcess-1h-10-8736-bt4 90 | 20241106-181402,Coverage[0.1],0.1431547619047619,gluonts-GaussianProcess-1h-10-8736-bt4 91 | 20241106-181402,QuantileLoss[0.2],150737.5502003047,gluonts-GaussianProcess-1h-10-8736-bt4 92 | 20241106-181402,Coverage[0.2],0.21696428571428572,gluonts-GaussianProcess-1h-10-8736-bt4 93 | 20241106-181402,QuantileLoss[0.3],176394.10445404742,gluonts-GaussianProcess-1h-10-8736-bt4 94 | 20241106-181402,Coverage[0.3],0.3007440476190476,gluonts-GaussianProcess-1h-10-8736-bt4 95 | 20241106-181402,QuantileLoss[0.4],190410.02594109948,gluonts-GaussianProcess-1h-10-8736-bt4 96 | 20241106-181402,Coverage[0.4],0.3949404761904762,gluonts-GaussianProcess-1h-10-8736-bt4 97 | 20241106-181402,QuantileLoss[0.5],195038.23047864335,gluonts-GaussianProcess-1h-10-8736-bt4 98 | 20241106-181402,Coverage[0.5],0.48377976190476185,gluonts-GaussianProcess-1h-10-8736-bt4 99 | 20241106-181402,QuantileLoss[0.6],189187.5788318215,gluonts-GaussianProcess-1h-10-8736-bt4 100 | 20241106-181402,Coverage[0.6],0.5255952380952381,gluonts-GaussianProcess-1h-10-8736-bt4 101 | 20241106-181402,QuantileLoss[0.7],173720.23567715057,gluonts-GaussianProcess-1h-10-8736-bt4 102 | 20241106-181402,Coverage[0.7],0.6169642857142857,gluonts-GaussianProcess-1h-10-8736-bt4 103 | 20241106-181402,QuantileLoss[0.8],144857.75769209766,gluonts-GaussianProcess-1h-10-8736-bt4 104 | 20241106-181402,Coverage[0.8],0.7041666666666666,gluonts-GaussianProcess-1h-10-8736-bt4 105 | 20241106-181402,QuantileLoss[0.9],99212.02530412623,gluonts-GaussianProcess-1h-10-8736-bt4 106 | 20241106-181402,Coverage[0.9],0.7925595238095238,gluonts-GaussianProcess-1h-10-8736-bt4 107 | 20241106-181402,RMSE,51.775960488045946,gluonts-GaussianProcess-1h-10-8736-bt4 108 | 20241106-181402,NRMSE,0.32812737812674714,gluonts-GaussianProcess-1h-10-8736-bt4 109 | 20241106-181402,ND,0.1839351704824539,gluonts-GaussianProcess-1h-10-8736-bt4 110 | 20241106-181402,wQuantileLoss[0.1],0.10196712358244532,gluonts-GaussianProcess-1h-10-8736-bt4 111 | 20241106-181402,wQuantileLoss[0.2],0.1421564219802357,gluonts-GaussianProcess-1h-10-8736-bt4 112 | 20241106-181402,wQuantileLoss[0.3],0.16635240996204445,gluonts-GaussianProcess-1h-10-8736-bt4 113 | 20241106-181402,wQuantileLoss[0.4],0.17957043856014487,gluonts-GaussianProcess-1h-10-8736-bt4 114 | 20241106-181402,wQuantileLoss[0.5],0.1839351704824539,gluonts-GaussianProcess-1h-10-8736-bt4 115 | 20241106-181402,wQuantileLoss[0.6],0.17841758244112135,gluonts-GaussianProcess-1h-10-8736-bt4 116 | 20241106-181402,wQuantileLoss[0.7],0.16383075813963374,gluonts-GaussianProcess-1h-10-8736-bt4 117 | 20241106-181402,wQuantileLoss[0.8],0.13661135199706162,gluonts-GaussianProcess-1h-10-8736-bt4 118 | 20241106-181402,wQuantileLoss[0.9],0.09356412198490595,gluonts-GaussianProcess-1h-10-8736-bt4 119 | 20241106-181402,mean_absolute_QuantileLoss,158631.0860165578,gluonts-GaussianProcess-1h-10-8736-bt4 120 | 20241106-181402,mean_wQuantileLoss,0.14960059768111633,gluonts-GaussianProcess-1h-10-8736-bt4 121 | 20241106-181402,MAE_Coverage,0.31808862433862434,gluonts-GaussianProcess-1h-10-8736-bt4 122 | 20241106-181402,MSE,2319.9428871070177,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 123 | 20241106-181402,abs_error,177290.60106763607,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 124 | 20241106-181402,abs_target_sum,1060363.985675315,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 125 | 20241106-181402,abs_target_mean,157.79225977311233,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 126 | 20241106-181402,MAPE,0.28652680420823484,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 127 | 20241106-181402,sMAPE,0.25022229327789214,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 128 | 20241106-181402,num_masked_target_values,0.0,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 129 | 20241106-181402,QuantileLoss[0.1],89518.85284943001,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 130 | 20241106-181402,Coverage[0.1],0.13467261904761904,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 131 | 20241106-181402,QuantileLoss[0.2],129303.35146963695,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 132 | 20241106-181402,Coverage[0.2],0.18839285714285717,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 133 | 20241106-181402,QuantileLoss[0.3],156751.6344294235,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 134 | 20241106-181402,Coverage[0.3],0.24583333333333335,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 135 | 20241106-181402,QuantileLoss[0.4],172359.21026093402,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 136 | 20241106-181402,Coverage[0.4],0.30625,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 137 | 20241106-181402,QuantileLoss[0.5],177290.60106763607,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 138 | 20241106-181402,Coverage[0.5],0.384375,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 139 | 20241106-181402,QuantileLoss[0.6],170113.25586288184,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 140 | 20241106-181402,Coverage[0.6],0.5063988095238094,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 141 | 20241106-181402,QuantileLoss[0.7],156581.46458156954,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 142 | 20241106-181402,Coverage[0.7],0.621875,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 143 | 20241106-181402,QuantileLoss[0.8],132627.30403193354,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 144 | 20241106-181402,Coverage[0.8],0.7117559523809524,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 145 | 20241106-181402,QuantileLoss[0.9],95599.67192574969,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 146 | 20241106-181402,Coverage[0.9],0.7915178571428572,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 147 | 20241106-181402,RMSE,48.16578544056993,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 148 | 20241106-181402,NRMSE,0.3052480870090013,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 149 | 20241106-181402,ND,0.1671978711675358,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 150 | 20241106-181402,wQuantileLoss[0.1],0.08442275865529143,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 151 | 20241106-181402,wQuantileLoss[0.2],0.12194242092000834,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 152 | 20241106-181402,wQuantileLoss[0.3],0.14782813877783008,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 153 | 20241106-181402,wQuantileLoss[0.4],0.16254721264525357,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 154 | 20241106-181402,wQuantileLoss[0.5],0.1671978711675358,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 155 | 20241106-181402,wQuantileLoss[0.6],0.1604291150595252,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 156 | 20241106-181402,wQuantileLoss[0.7],0.1476676562924262,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 157 | 20241106-181402,wQuantileLoss[0.8],0.1250771488126948,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 158 | 20241106-181402,wQuantileLoss[0.9],0.09015741124484254,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 159 | 20241106-181402,mean_absolute_QuantileLoss,142238.37183102168,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 160 | 20241106-181402,mean_wQuantileLoss,0.13414108150837867,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 161 | 20241106-181402,MAE_Coverage,0.3175099206349207,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 162 | 20241106-181402,MSE,2132.347569452132,gluonts-MQCNN-1h-10-8736-bt4 163 | 20241106-181402,abs_error,157090.22064877092,gluonts-MQCNN-1h-10-8736-bt4 164 | 20241106-181402,abs_target_sum,1060363.985675315,gluonts-MQCNN-1h-10-8736-bt4 165 | 20241106-181402,abs_target_mean,157.79225977311233,gluonts-MQCNN-1h-10-8736-bt4 166 | 20241106-181402,MAPE,0.2903656085227503,gluonts-MQCNN-1h-10-8736-bt4 167 | 20241106-181402,sMAPE,0.2112283989107681,gluonts-MQCNN-1h-10-8736-bt4 168 | 20241106-181402,num_masked_target_values,0.0,gluonts-MQCNN-1h-10-8736-bt4 169 | 20241106-181402,QuantileLoss[0.1],100859.35444308957,gluonts-MQCNN-1h-10-8736-bt4 170 | 20241106-181402,Coverage[0.1],0.24553571428571427,gluonts-MQCNN-1h-10-8736-bt4 171 | 20241106-181402,QuantileLoss[0.2],134774.9373253173,gluonts-MQCNN-1h-10-8736-bt4 172 | 20241106-181402,Coverage[0.2],0.32901785714285714,gluonts-MQCNN-1h-10-8736-bt4 173 | 20241106-181402,QuantileLoss[0.3],152094.46182656707,gluonts-MQCNN-1h-10-8736-bt4 174 | 20241106-181402,Coverage[0.3],0.41830357142857144,gluonts-MQCNN-1h-10-8736-bt4 175 | 20241106-181402,QuantileLoss[0.4],155147.88285885996,gluonts-MQCNN-1h-10-8736-bt4 176 | 20241106-181402,Coverage[0.4],0.4697916666666667,gluonts-MQCNN-1h-10-8736-bt4 177 | 20241106-181402,QuantileLoss[0.5],157090.22064877092,gluonts-MQCNN-1h-10-8736-bt4 178 | 20241106-181402,Coverage[0.5],0.5912202380952382,gluonts-MQCNN-1h-10-8736-bt4 179 | 20241106-181402,QuantileLoss[0.6],144346.27805889462,gluonts-MQCNN-1h-10-8736-bt4 180 | 20241106-181402,Coverage[0.6],0.5994047619047619,gluonts-MQCNN-1h-10-8736-bt4 181 | 20241106-181402,QuantileLoss[0.7],133733.16929524357,gluonts-MQCNN-1h-10-8736-bt4 182 | 20241106-181402,Coverage[0.7],0.7633928571428571,gluonts-MQCNN-1h-10-8736-bt4 183 | 20241106-181402,QuantileLoss[0.8],108628.85800980823,gluonts-MQCNN-1h-10-8736-bt4 184 | 20241106-181402,Coverage[0.8],0.7744047619047618,gluonts-MQCNN-1h-10-8736-bt4 185 | 20241106-181402,QuantileLoss[0.9],77060.79549965278,gluonts-MQCNN-1h-10-8736-bt4 186 | 20241106-181402,Coverage[0.9],0.8636904761904762,gluonts-MQCNN-1h-10-8736-bt4 187 | 20241106-181402,RMSE,46.177349095115154,gluonts-MQCNN-1h-10-8736-bt4 188 | 20241106-181402,NRMSE,0.2926464781068034,gluonts-MQCNN-1h-10-8736-bt4 189 | 20241106-181402,ND,0.1481474500934929,gluonts-MQCNN-1h-10-8736-bt4 190 | 20241106-181402,wQuantileLoss[0.1],0.09511767261583784,gluonts-MQCNN-1h-10-8736-bt4 191 | 20241106-181402,wQuantileLoss[0.2],0.12710252247909293,gluonts-MQCNN-1h-10-8736-bt4 192 | 20241106-181402,wQuantileLoss[0.3],0.14343608787288503,gluonts-MQCNN-1h-10-8736-bt4 193 | 20241106-181402,wQuantileLoss[0.4],0.14631568494855168,gluonts-MQCNN-1h-10-8736-bt4 194 | 20241106-181402,wQuantileLoss[0.5],0.1481474500934929,gluonts-MQCNN-1h-10-8736-bt4 195 | 20241106-181402,wQuantileLoss[0.6],0.1361289896760919,gluonts-MQCNN-1h-10-8736-bt4 196 | 20241106-181402,wQuantileLoss[0.7],0.12612005981141733,gluonts-MQCNN-1h-10-8736-bt4 197 | 20241106-181402,wQuantileLoss[0.8],0.10244487692650715,gluonts-MQCNN-1h-10-8736-bt4 198 | 20241106-181402,wQuantileLoss[0.9],0.07267390871501073,gluonts-MQCNN-1h-10-8736-bt4 199 | 20241106-181402,mean_absolute_QuantileLoss,129303.99532957823,gluonts-MQCNN-1h-10-8736-bt4 200 | 20241106-181402,mean_wQuantileLoss,0.12194302812654306,gluonts-MQCNN-1h-10-8736-bt4 201 | 20241106-181402,MAE_Coverage,0.3717592592592593,gluonts-MQCNN-1h-10-8736-bt4 202 | 20241106-181402,MSE,19358.998269501197,gluonts-MQRNN-1h-10-8736-bt4 203 | 20241106-181402,abs_error,843923.815799206,gluonts-MQRNN-1h-10-8736-bt4 204 | 20241106-181402,abs_target_sum,1060363.985675315,gluonts-MQRNN-1h-10-8736-bt4 205 | 20241106-181402,abs_target_mean,157.79225977311233,gluonts-MQRNN-1h-10-8736-bt4 206 | 20241106-181402,MAPE,10.950899213778792,gluonts-MQRNN-1h-10-8736-bt4 207 | 20241106-181402,sMAPE,0.9882469344378965,gluonts-MQRNN-1h-10-8736-bt4 208 | 20241106-181402,num_masked_target_values,0.0,gluonts-MQRNN-1h-10-8736-bt4 209 | 20241106-181402,QuantileLoss[0.1],527963.2541242407,gluonts-MQRNN-1h-10-8736-bt4 210 | 20241106-181402,Coverage[0.1],0.6425595238095237,gluonts-MQRNN-1h-10-8736-bt4 211 | 20241106-181402,QuantileLoss[0.2],809787.3054428424,gluonts-MQRNN-1h-10-8736-bt4 212 | 20241106-181402,Coverage[0.2],0.7729166666666666,gluonts-MQRNN-1h-10-8736-bt4 213 | 20241106-181402,QuantileLoss[0.3],982833.5523863547,gluonts-MQRNN-1h-10-8736-bt4 214 | 20241106-181402,Coverage[0.3],0.8785714285714287,gluonts-MQRNN-1h-10-8736-bt4 215 | 20241106-181402,QuantileLoss[0.4],866151.7817861272,gluonts-MQRNN-1h-10-8736-bt4 216 | 20241106-181402,Coverage[0.4],0.8785714285714287,gluonts-MQRNN-1h-10-8736-bt4 217 | 20241106-181402,QuantileLoss[0.5],843923.815799206,gluonts-MQRNN-1h-10-8736-bt4 218 | 20241106-181402,Coverage[0.5],0.9223214285714286,gluonts-MQRNN-1h-10-8736-bt4 219 | 20241106-181402,QuantileLoss[0.6],701255.2492217594,gluonts-MQRNN-1h-10-8736-bt4 220 | 20241106-181402,Coverage[0.6],0.940922619047619,gluonts-MQRNN-1h-10-8736-bt4 221 | 20241106-181402,QuantileLoss[0.7],575379.3920975274,gluonts-MQRNN-1h-10-8736-bt4 222 | 20241106-181402,Coverage[0.7],0.9633928571428572,gluonts-MQRNN-1h-10-8736-bt4 223 | 20241106-181402,QuantileLoss[0.8],421511.03247122094,gluonts-MQRNN-1h-10-8736-bt4 224 | 20241106-181402,Coverage[0.8],0.9790178571428572,gluonts-MQRNN-1h-10-8736-bt4 225 | 20241106-181402,QuantileLoss[0.9],248910.9942025168,gluonts-MQRNN-1h-10-8736-bt4 226 | 20241106-181402,Coverage[0.9],0.990625,gluonts-MQRNN-1h-10-8736-bt4 227 | 20241106-181402,RMSE,139.13661728495916,gluonts-MQRNN-1h-10-8736-bt4 228 | 20241106-181402,NRMSE,0.8817708643315083,gluonts-MQRNN-1h-10-8736-bt4 229 | 20241106-181402,ND,0.795881251343835,gluonts-MQRNN-1h-10-8736-bt4 230 | 20241106-181402,wQuantileLoss[0.1],0.4979075687750714,gluonts-MQRNN-1h-10-8736-bt4 231 | 20241106-181402,wQuantileLoss[0.2],0.7636880508791635,gluonts-MQRNN-1h-10-8736-bt4 232 | 20241106-181402,wQuantileLoss[0.3],0.9268831888518135,gluonts-MQRNN-1h-10-8736-bt4 233 | 20241106-181402,wQuantileLoss[0.4],0.8168438323888381,gluonts-MQRNN-1h-10-8736-bt4 234 | 20241106-181402,wQuantileLoss[0.5],0.795881251343835,gluonts-MQRNN-1h-10-8736-bt4 235 | 20241106-181402,wQuantileLoss[0.6],0.6613344650470663,gluonts-MQRNN-1h-10-8736-bt4 236 | 20241106-181402,wQuantileLoss[0.7],0.542624419416777,gluonts-MQRNN-1h-10-8736-bt4 237 | 20241106-181402,wQuantileLoss[0.8],0.3975154175033329,gluonts-MQRNN-1h-10-8736-bt4 238 | 20241106-181402,wQuantileLoss[0.9],0.2347410866128131,gluonts-MQRNN-1h-10-8736-bt4 239 | 20241106-181402,mean_absolute_QuantileLoss,664190.708614644,gluonts-MQRNN-1h-10-8736-bt4 240 | 20241106-181402,mean_wQuantileLoss,0.6263799200909679,gluonts-MQRNN-1h-10-8736-bt4 241 | 20241106-181402,MAE_Coverage,0.4906249999999999,gluonts-MQRNN-1h-10-8736-bt4 242 | 20241106-181402,MSE,2066.67041301526,gluonts-SeasonalNaive-1h-10-8736-bt4 243 | 20241106-181402,abs_error,147631.08123499842,gluonts-SeasonalNaive-1h-10-8736-bt4 244 | 20241106-181402,abs_target_sum,1060363.985675315,gluonts-SeasonalNaive-1h-10-8736-bt4 245 | 20241106-181402,abs_target_mean,157.79225977311233,gluonts-SeasonalNaive-1h-10-8736-bt4 246 | 20241106-181402,MAPE,0.29139723096273695,gluonts-SeasonalNaive-1h-10-8736-bt4 247 | 20241106-181402,sMAPE,0.1964473079101054,gluonts-SeasonalNaive-1h-10-8736-bt4 248 | 20241106-181402,num_masked_target_values,0.0,gluonts-SeasonalNaive-1h-10-8736-bt4 249 | 20241106-181402,QuantileLoss[0.1],179223.60020769734,gluonts-SeasonalNaive-1h-10-8736-bt4 250 | 20241106-181402,Coverage[0.1],0.5349702380952381,gluonts-SeasonalNaive-1h-10-8736-bt4 251 | 20241106-181402,QuantileLoss[0.2],171325.47046452263,gluonts-SeasonalNaive-1h-10-8736-bt4 252 | 20241106-181402,Coverage[0.2],0.5349702380952381,gluonts-SeasonalNaive-1h-10-8736-bt4 253 | 20241106-181402,QuantileLoss[0.3],163427.34072134786,gluonts-SeasonalNaive-1h-10-8736-bt4 254 | 20241106-181402,Coverage[0.3],0.5349702380952381,gluonts-SeasonalNaive-1h-10-8736-bt4 255 | 20241106-181402,QuantileLoss[0.4],155529.21097817313,gluonts-SeasonalNaive-1h-10-8736-bt4 256 | 20241106-181402,Coverage[0.4],0.5349702380952381,gluonts-SeasonalNaive-1h-10-8736-bt4 257 | 20241106-181402,QuantileLoss[0.5],147631.08123499842,gluonts-SeasonalNaive-1h-10-8736-bt4 258 | 20241106-181402,Coverage[0.5],0.5349702380952381,gluonts-SeasonalNaive-1h-10-8736-bt4 259 | 20241106-181402,QuantileLoss[0.6],139732.9514918237,gluonts-SeasonalNaive-1h-10-8736-bt4 260 | 20241106-181402,Coverage[0.6],0.5349702380952381,gluonts-SeasonalNaive-1h-10-8736-bt4 261 | 20241106-181402,QuantileLoss[0.7],131834.82174864894,gluonts-SeasonalNaive-1h-10-8736-bt4 262 | 20241106-181402,Coverage[0.7],0.5349702380952381,gluonts-SeasonalNaive-1h-10-8736-bt4 263 | 20241106-181402,QuantileLoss[0.8],123936.69200547425,gluonts-SeasonalNaive-1h-10-8736-bt4 264 | 20241106-181402,Coverage[0.8],0.5349702380952381,gluonts-SeasonalNaive-1h-10-8736-bt4 265 | 20241106-181402,QuantileLoss[0.9],116038.56226229951,gluonts-SeasonalNaive-1h-10-8736-bt4 266 | 20241106-181402,Coverage[0.9],0.5349702380952381,gluonts-SeasonalNaive-1h-10-8736-bt4 267 | 20241106-181402,RMSE,45.46064686094182,gluonts-SeasonalNaive-1h-10-8736-bt4 268 | 20241106-181402,NRMSE,0.2881044160614035,gluonts-SeasonalNaive-1h-10-8736-bt4 269 | 20241106-181402,ND,0.13922679686350956,gluonts-SeasonalNaive-1h-10-8736-bt4 270 | 20241106-181402,wQuantileLoss[0.1],0.16902082928962836,gluonts-SeasonalNaive-1h-10-8736-bt4 271 | 20241106-181402,wQuantileLoss[0.2],0.1615723211830987,gluonts-SeasonalNaive-1h-10-8736-bt4 272 | 20241106-181402,wQuantileLoss[0.3],0.15412381307656894,gluonts-SeasonalNaive-1h-10-8736-bt4 273 | 20241106-181402,wQuantileLoss[0.4],0.14667530497003922,gluonts-SeasonalNaive-1h-10-8736-bt4 274 | 20241106-181402,wQuantileLoss[0.5],0.13922679686350956,gluonts-SeasonalNaive-1h-10-8736-bt4 275 | 20241106-181402,wQuantileLoss[0.6],0.13177828875697986,gluonts-SeasonalNaive-1h-10-8736-bt4 276 | 20241106-181402,wQuantileLoss[0.7],0.12432978065045013,gluonts-SeasonalNaive-1h-10-8736-bt4 277 | 20241106-181402,wQuantileLoss[0.8],0.11688127254392046,gluonts-SeasonalNaive-1h-10-8736-bt4 278 | 20241106-181402,wQuantileLoss[0.9],0.10943276443739075,gluonts-SeasonalNaive-1h-10-8736-bt4 279 | 20241106-181402,mean_absolute_QuantileLoss,147631.08123499842,gluonts-SeasonalNaive-1h-10-8736-bt4 280 | 20241106-181402,mean_wQuantileLoss,0.13922679686350956,gluonts-SeasonalNaive-1h-10-8736-bt4 281 | 20241106-181402,MAE_Coverage,0.22610780423280424,gluonts-SeasonalNaive-1h-10-8736-bt4 282 | 20241106-181402,MSE,2041.7211777950765,gluonts-Prophet-1h-10-8736-bt4 283 | 20241106-181402,abs_error,173335.60835965935,gluonts-Prophet-1h-10-8736-bt4 284 | 20241106-181402,abs_target_sum,1060363.985675315,gluonts-Prophet-1h-10-8736-bt4 285 | 20241106-181402,abs_target_mean,157.79225977311233,gluonts-Prophet-1h-10-8736-bt4 286 | 20241106-181402,MAPE,0.34439058992896676,gluonts-Prophet-1h-10-8736-bt4 287 | 20241106-181402,sMAPE,0.2615444067105221,gluonts-Prophet-1h-10-8736-bt4 288 | 20241106-181402,num_masked_target_values,0.0,gluonts-Prophet-1h-10-8736-bt4 289 | 20241106-181402,QuantileLoss[0.1],99152.38494383259,gluonts-Prophet-1h-10-8736-bt4 290 | 20241106-181402,Coverage[0.1],0.13199404761904762,gluonts-Prophet-1h-10-8736-bt4 291 | 20241106-181402,QuantileLoss[0.2],133257.26048313058,gluonts-Prophet-1h-10-8736-bt4 292 | 20241106-181402,Coverage[0.2],0.20089285714285712,gluonts-Prophet-1h-10-8736-bt4 293 | 20241106-181402,QuantileLoss[0.3],155211.24386134886,gluonts-Prophet-1h-10-8736-bt4 294 | 20241106-181402,Coverage[0.3],0.290327380952381,gluonts-Prophet-1h-10-8736-bt4 295 | 20241106-181402,QuantileLoss[0.4],168381.82553495688,gluonts-Prophet-1h-10-8736-bt4 296 | 20241106-181402,Coverage[0.4],0.39613095238095236,gluonts-Prophet-1h-10-8736-bt4 297 | 20241106-181402,QuantileLoss[0.5],173335.60835965935,gluonts-Prophet-1h-10-8736-bt4 298 | 20241106-181402,Coverage[0.5],0.5114583333333333,gluonts-Prophet-1h-10-8736-bt4 299 | 20241106-181402,QuantileLoss[0.6],168946.8305044715,gluonts-Prophet-1h-10-8736-bt4 300 | 20241106-181402,Coverage[0.6],0.5602678571428571,gluonts-Prophet-1h-10-8736-bt4 301 | 20241106-181402,QuantileLoss[0.7],156724.31224252278,gluonts-Prophet-1h-10-8736-bt4 302 | 20241106-181402,Coverage[0.7],0.65625,gluonts-Prophet-1h-10-8736-bt4 303 | 20241106-181402,QuantileLoss[0.8],133471.9806772892,gluonts-Prophet-1h-10-8736-bt4 304 | 20241106-181402,Coverage[0.8],0.7346726190476189,gluonts-Prophet-1h-10-8736-bt4 305 | 20241106-181402,QuantileLoss[0.9],96277.16146839075,gluonts-Prophet-1h-10-8736-bt4 306 | 20241106-181402,Coverage[0.9],0.8072916666666667,gluonts-Prophet-1h-10-8736-bt4 307 | 20241106-181402,RMSE,45.18540890370559,gluonts-Prophet-1h-10-8736-bt4 308 | 20241106-181402,NRMSE,0.2863601102403703,gluonts-Prophet-1h-10-8736-bt4 309 | 20241106-181402,ND,0.16346802673542987,gluonts-Prophet-1h-10-8736-bt4 310 | 20241106-181402,wQuantileLoss[0.1],0.09350787680768441,gluonts-Prophet-1h-10-8736-bt4 311 | 20241106-181402,wQuantileLoss[0.2],0.1256712433497663,gluonts-Prophet-1h-10-8736-bt4 312 | 20241106-181402,wQuantileLoss[0.3],0.14637543896070682,gluonts-Prophet-1h-10-8736-bt4 313 | 20241106-181402,wQuantileLoss[0.4],0.1587962509191779,gluonts-Prophet-1h-10-8736-bt4 314 | 20241106-181402,wQuantileLoss[0.5],0.16346802673542987,gluonts-Prophet-1h-10-8736-bt4 315 | 20241106-181402,wQuantileLoss[0.6],0.15932909150707733,gluonts-Prophet-1h-10-8736-bt4 316 | 20241106-181402,wQuantileLoss[0.7],0.14780237197768428,gluonts-Prophet-1h-10-8736-bt4 317 | 20241106-181402,wQuantileLoss[0.8],0.12587374003680893,gluonts-Prophet-1h-10-8736-bt4 318 | 20241106-181402,wQuantileLoss[0.9],0.09079633292814508,gluonts-Prophet-1h-10-8736-bt4 319 | 20241106-181402,mean_absolute_QuantileLoss,142750.95645284472,gluonts-Prophet-1h-10-8736-bt4 320 | 20241106-181402,mean_wQuantileLoss,0.13462448591360898,gluonts-Prophet-1h-10-8736-bt4 321 | 20241106-181402,MAE_Coverage,0.32789351851851856,gluonts-Prophet-1h-10-8736-bt4 322 | 20241106-181402,MSE,2048.5676183551177,gluonts-NPTS-1h-10-8736-bt4 323 | 20241106-181402,abs_error,161316.0429316016,gluonts-NPTS-1h-10-8736-bt4 324 | 20241106-181402,abs_target_sum,1060363.985675315,gluonts-NPTS-1h-10-8736-bt4 325 | 20241106-181402,abs_target_mean,157.79225977311233,gluonts-NPTS-1h-10-8736-bt4 326 | 20241106-181402,MAPE,0.2924837016851417,gluonts-NPTS-1h-10-8736-bt4 327 | 20241106-181402,sMAPE,0.21299489574639283,gluonts-NPTS-1h-10-8736-bt4 328 | 20241106-181402,num_masked_target_values,0.0,gluonts-NPTS-1h-10-8736-bt4 329 | 20241106-181402,QuantileLoss[0.1],87846.8666409161,gluonts-NPTS-1h-10-8736-bt4 330 | 20241106-181402,Coverage[0.1],0.20967261904761902,gluonts-NPTS-1h-10-8736-bt4 331 | 20241106-181402,QuantileLoss[0.2],123593.96445873434,gluonts-NPTS-1h-10-8736-bt4 332 | 20241106-181402,Coverage[0.2],0.2901785714285714,gluonts-NPTS-1h-10-8736-bt4 333 | 20241106-181402,QuantileLoss[0.3],146881.18105940163,gluonts-NPTS-1h-10-8736-bt4 334 | 20241106-181402,Coverage[0.3],0.35892857142857143,gluonts-NPTS-1h-10-8736-bt4 335 | 20241106-181402,QuantileLoss[0.4],157408.59694362,gluonts-NPTS-1h-10-8736-bt4 336 | 20241106-181402,Coverage[0.4],0.4197916666666667,gluonts-NPTS-1h-10-8736-bt4 337 | 20241106-181402,QuantileLoss[0.5],161316.0429316016,gluonts-NPTS-1h-10-8736-bt4 338 | 20241106-181402,Coverage[0.5],0.4787202380952381,gluonts-NPTS-1h-10-8736-bt4 339 | 20241106-181402,QuantileLoss[0.6],156298.37572599584,gluonts-NPTS-1h-10-8736-bt4 340 | 20241106-181402,Coverage[0.6],0.5165178571428573,gluonts-NPTS-1h-10-8736-bt4 341 | 20241106-181402,QuantileLoss[0.7],145057.4421188393,gluonts-NPTS-1h-10-8736-bt4 342 | 20241106-181402,Coverage[0.7],0.5876488095238096,gluonts-NPTS-1h-10-8736-bt4 343 | 20241106-181402,QuantileLoss[0.8],121466.12843783917,gluonts-NPTS-1h-10-8736-bt4 344 | 20241106-181402,Coverage[0.8],0.6677083333333333,gluonts-NPTS-1h-10-8736-bt4 345 | 20241106-181402,QuantileLoss[0.9],85589.66618242188,gluonts-NPTS-1h-10-8736-bt4 346 | 20241106-181402,Coverage[0.9],0.7633928571428571,gluonts-NPTS-1h-10-8736-bt4 347 | 20241106-181402,RMSE,45.26110491752403,gluonts-NPTS-1h-10-8736-bt4 348 | 20241106-181402,NRMSE,0.28683982967608457,gluonts-NPTS-1h-10-8736-bt4 349 | 20241106-181402,ND,0.15213270642048837,gluonts-NPTS-1h-10-8736-bt4 350 | 20241106-181402,wQuantileLoss[0.1],0.08284595462280717,gluonts-NPTS-1h-10-8736-bt4 351 | 20241106-181402,wQuantileLoss[0.2],0.11655805565672898,gluonts-NPTS-1h-10-8736-bt4 352 | 20241106-181402,wQuantileLoss[0.3],0.1385195867114039,gluonts-NPTS-1h-10-8736-bt4 353 | 20241106-181402,wQuantileLoss[0.4],0.1484477019873238,gluonts-NPTS-1h-10-8736-bt4 354 | 20241106-181402,wQuantileLoss[0.5],0.15213270642048837,gluonts-NPTS-1h-10-8736-bt4 355 | 20241106-181402,wQuantileLoss[0.6],0.14740068300834827,gluonts-NPTS-1h-10-8736-bt4 356 | 20241106-181402,wQuantileLoss[0.7],0.13679966886696596,gluonts-NPTS-1h-10-8736-bt4 357 | 20241106-181402,wQuantileLoss[0.8],0.114551352251445,gluonts-NPTS-1h-10-8736-bt4 358 | 20241106-181402,wQuantileLoss[0.9],0.080717251187961,gluonts-NPTS-1h-10-8736-bt4 359 | 20241106-181402,mean_absolute_QuantileLoss,131717.58494437442,gluonts-NPTS-1h-10-8736-bt4 360 | 20241106-181402,mean_wQuantileLoss,0.1242192178570525,gluonts-NPTS-1h-10-8736-bt4 361 | 20241106-181402,MAE_Coverage,0.3018849206349207,gluonts-NPTS-1h-10-8736-bt4 362 | -------------------------------------------------------------------------------- /test/model-performance/gluonts-1h-10-8736-bt4-20241107-214657.csv: -------------------------------------------------------------------------------- 1 | timestamp,metric_name,value,experiment 2 | 20241107-214657,MSE,621365.1752981541,gluonts-SimpleFeedForward-1h-10-8736-bt4 3 | 20241107-214657,abs_error,847399.32650858,gluonts-SimpleFeedForward-1h-10-8736-bt4 4 | 20241107-214657,abs_target_sum,7840969.521757782,gluonts-SimpleFeedForward-1h-10-8736-bt4 5 | 20241107-214657,abs_target_mean,1166.8109407377653,gluonts-SimpleFeedForward-1h-10-8736-bt4 6 | 20241107-214657,MAPE,0.19890838701033067,gluonts-SimpleFeedForward-1h-10-8736-bt4 7 | 20241107-214657,sMAPE,0.16907286257610193,gluonts-SimpleFeedForward-1h-10-8736-bt4 8 | 20241107-214657,num_masked_target_values,0.0,gluonts-SimpleFeedForward-1h-10-8736-bt4 9 | 20241107-214657,QuantileLoss[0.1],758006.3894938908,gluonts-SimpleFeedForward-1h-10-8736-bt4 10 | 20241107-214657,Coverage[0.1],0.171875,gluonts-SimpleFeedForward-1h-10-8736-bt4 11 | 20241107-214657,QuantileLoss[0.2],861823.9839150833,gluonts-SimpleFeedForward-1h-10-8736-bt4 12 | 20241107-214657,Coverage[0.2],0.278125,gluonts-SimpleFeedForward-1h-10-8736-bt4 13 | 20241107-214657,QuantileLoss[0.3],895174.3729425969,gluonts-SimpleFeedForward-1h-10-8736-bt4 14 | 20241107-214657,Coverage[0.3],0.39404761904761904,gluonts-SimpleFeedForward-1h-10-8736-bt4 15 | 20241107-214657,QuantileLoss[0.4],886984.2024955886,gluonts-SimpleFeedForward-1h-10-8736-bt4 16 | 20241107-214657,Coverage[0.4],0.5019345238095239,gluonts-SimpleFeedForward-1h-10-8736-bt4 17 | 20241107-214657,QuantileLoss[0.5],847399.32650858,gluonts-SimpleFeedForward-1h-10-8736-bt4 18 | 20241107-214657,Coverage[0.5],0.6023809523809524,gluonts-SimpleFeedForward-1h-10-8736-bt4 19 | 20241107-214657,QuantileLoss[0.6],768886.3691748355,gluonts-SimpleFeedForward-1h-10-8736-bt4 20 | 20241107-214657,Coverage[0.6],0.6489583333333334,gluonts-SimpleFeedForward-1h-10-8736-bt4 21 | 20241107-214657,QuantileLoss[0.7],683213.9093425189,gluonts-SimpleFeedForward-1h-10-8736-bt4 22 | 20241107-214657,Coverage[0.7],0.7354166666666666,gluonts-SimpleFeedForward-1h-10-8736-bt4 23 | 20241107-214657,QuantileLoss[0.8],562789.6253904708,gluonts-SimpleFeedForward-1h-10-8736-bt4 24 | 20241107-214657,Coverage[0.8],0.8132440476190477,gluonts-SimpleFeedForward-1h-10-8736-bt4 25 | 20241107-214657,QuantileLoss[0.9],394919.6603022926,gluonts-SimpleFeedForward-1h-10-8736-bt4 26 | 20241107-214657,Coverage[0.9],0.8880952380952379,gluonts-SimpleFeedForward-1h-10-8736-bt4 27 | 20241107-214657,RMSE,788.2671979082689,gluonts-SimpleFeedForward-1h-10-8736-bt4 28 | 20241107-214657,NRMSE,0.6755740543621007,gluonts-SimpleFeedForward-1h-10-8736-bt4 29 | 20241107-214657,ND,0.10807328406992847,gluonts-SimpleFeedForward-1h-10-8736-bt4 30 | 20241107-214657,wQuantileLoss[0.1],0.09667253359300924,gluonts-SimpleFeedForward-1h-10-8736-bt4 31 | 20241107-214657,wQuantileLoss[0.2],0.10991293634334652,gluonts-SimpleFeedForward-1h-10-8736-bt4 32 | 20241107-214657,wQuantileLoss[0.3],0.11416628651069127,gluonts-SimpleFeedForward-1h-10-8736-bt4 33 | 20241107-214657,wQuantileLoss[0.4],0.11312175108375438,gluonts-SimpleFeedForward-1h-10-8736-bt4 34 | 20241107-214657,wQuantileLoss[0.5],0.10807328406992847,gluonts-SimpleFeedForward-1h-10-8736-bt4 35 | 20241107-214657,wQuantileLoss[0.6],0.09806011451023561,gluonts-SimpleFeedForward-1h-10-8736-bt4 36 | 20241107-214657,wQuantileLoss[0.7],0.08713385601699884,gluonts-SimpleFeedForward-1h-10-8736-bt4 37 | 20241107-214657,wQuantileLoss[0.8],0.07177551498303811,gluonts-SimpleFeedForward-1h-10-8736-bt4 38 | 20241107-214657,wQuantileLoss[0.9],0.05036617719357744,gluonts-SimpleFeedForward-1h-10-8736-bt4 39 | 20241107-214657,mean_absolute_QuantileLoss,739910.871062873,gluonts-SimpleFeedForward-1h-10-8736-bt4 40 | 20241107-214657,mean_wQuantileLoss,0.09436471714495331,gluonts-SimpleFeedForward-1h-10-8736-bt4 41 | 20241107-214657,MAE_Coverage,0.3907407407407406,gluonts-SimpleFeedForward-1h-10-8736-bt4 42 | 20241107-214657,MSE,513058.7270739113,gluonts-NBEATS-1h-10-8736-bt4 43 | 20241107-214657,abs_error,830151.258899364,gluonts-NBEATS-1h-10-8736-bt4 44 | 20241107-214657,abs_target_sum,7840969.521757782,gluonts-NBEATS-1h-10-8736-bt4 45 | 20241107-214657,abs_target_mean,1166.8109407377653,gluonts-NBEATS-1h-10-8736-bt4 46 | 20241107-214657,MAPE,0.16987577294867512,gluonts-NBEATS-1h-10-8736-bt4 47 | 20241107-214657,sMAPE,0.16795732212397113,gluonts-NBEATS-1h-10-8736-bt4 48 | 20241107-214657,num_masked_target_values,0.0,gluonts-NBEATS-1h-10-8736-bt4 49 | 20241107-214657,QuantileLoss[0.1],927928.3856524556,gluonts-NBEATS-1h-10-8736-bt4 50 | 20241107-214657,Coverage[0.1],0.4413690476190476,gluonts-NBEATS-1h-10-8736-bt4 51 | 20241107-214657,QuantileLoss[0.2],903484.1039641828,gluonts-NBEATS-1h-10-8736-bt4 52 | 20241107-214657,Coverage[0.2],0.4413690476190476,gluonts-NBEATS-1h-10-8736-bt4 53 | 20241107-214657,QuantileLoss[0.3],879039.8222759096,gluonts-NBEATS-1h-10-8736-bt4 54 | 20241107-214657,Coverage[0.3],0.4413690476190476,gluonts-NBEATS-1h-10-8736-bt4 55 | 20241107-214657,QuantileLoss[0.4],854595.5405876366,gluonts-NBEATS-1h-10-8736-bt4 56 | 20241107-214657,Coverage[0.4],0.4413690476190476,gluonts-NBEATS-1h-10-8736-bt4 57 | 20241107-214657,QuantileLoss[0.5],830151.258899364,gluonts-NBEATS-1h-10-8736-bt4 58 | 20241107-214657,Coverage[0.5],0.4413690476190476,gluonts-NBEATS-1h-10-8736-bt4 59 | 20241107-214657,QuantileLoss[0.6],805706.9772110912,gluonts-NBEATS-1h-10-8736-bt4 60 | 20241107-214657,Coverage[0.6],0.4413690476190476,gluonts-NBEATS-1h-10-8736-bt4 61 | 20241107-214657,QuantileLoss[0.7],781262.6955228181,gluonts-NBEATS-1h-10-8736-bt4 62 | 20241107-214657,Coverage[0.7],0.4413690476190476,gluonts-NBEATS-1h-10-8736-bt4 63 | 20241107-214657,QuantileLoss[0.8],756818.4138345454,gluonts-NBEATS-1h-10-8736-bt4 64 | 20241107-214657,Coverage[0.8],0.4413690476190476,gluonts-NBEATS-1h-10-8736-bt4 65 | 20241107-214657,QuantileLoss[0.9],732374.1321462727,gluonts-NBEATS-1h-10-8736-bt4 66 | 20241107-214657,Coverage[0.9],0.4413690476190476,gluonts-NBEATS-1h-10-8736-bt4 67 | 20241107-214657,RMSE,716.2811787796126,gluonts-NBEATS-1h-10-8736-bt4 68 | 20241107-214657,NRMSE,0.6138793816303382,gluonts-NBEATS-1h-10-8736-bt4 69 | 20241107-214657,ND,0.10587354747340752,gluonts-NBEATS-1h-10-8736-bt4 70 | 20241107-214657,wQuantileLoss[0.1],0.1183435776758935,gluonts-NBEATS-1h-10-8736-bt4 71 | 20241107-214657,wQuantileLoss[0.2],0.11522607012527201,gluonts-NBEATS-1h-10-8736-bt4 72 | 20241107-214657,wQuantileLoss[0.3],0.11210856257465047,gluonts-NBEATS-1h-10-8736-bt4 73 | 20241107-214657,wQuantileLoss[0.4],0.10899105502402898,gluonts-NBEATS-1h-10-8736-bt4 74 | 20241107-214657,wQuantileLoss[0.5],0.10587354747340752,gluonts-NBEATS-1h-10-8736-bt4 75 | 20241107-214657,wQuantileLoss[0.6],0.10275603992278604,gluonts-NBEATS-1h-10-8736-bt4 76 | 20241107-214657,wQuantileLoss[0.7],0.09963853237216452,gluonts-NBEATS-1h-10-8736-bt4 77 | 20241107-214657,wQuantileLoss[0.8],0.09652102482154304,gluonts-NBEATS-1h-10-8736-bt4 78 | 20241107-214657,wQuantileLoss[0.9],0.09340351727092157,gluonts-NBEATS-1h-10-8736-bt4 79 | 20241107-214657,mean_absolute_QuantileLoss,830151.258899364,gluonts-NBEATS-1h-10-8736-bt4 80 | 20241107-214657,mean_wQuantileLoss,0.10587354747340752,gluonts-NBEATS-1h-10-8736-bt4 81 | 20241107-214657,MAE_Coverage,0.22873677248677252,gluonts-NBEATS-1h-10-8736-bt4 82 | 20241107-214657,MSE,517353.7761313581,gluonts-DeepAR-1h-10-8736-bt4 83 | 20241107-214657,abs_error,1078050.1977001084,gluonts-DeepAR-1h-10-8736-bt4 84 | 20241107-214657,abs_target_sum,7840969.521757782,gluonts-DeepAR-1h-10-8736-bt4 85 | 20241107-214657,abs_target_mean,1166.8109407377653,gluonts-DeepAR-1h-10-8736-bt4 86 | 20241107-214657,MAPE,0.21178039029426438,gluonts-DeepAR-1h-10-8736-bt4 87 | 20241107-214657,sMAPE,0.21684134632204927,gluonts-DeepAR-1h-10-8736-bt4 88 | 20241107-214657,num_masked_target_values,0.0,gluonts-DeepAR-1h-10-8736-bt4 89 | 20241107-214657,QuantileLoss[0.1],705143.2536384666,gluonts-DeepAR-1h-10-8736-bt4 90 | 20241107-214657,Coverage[0.1],0.059375,gluonts-DeepAR-1h-10-8736-bt4 91 | 20241107-214657,QuantileLoss[0.2],875018.9825982942,gluonts-DeepAR-1h-10-8736-bt4 92 | 20241107-214657,Coverage[0.2],0.07693452380952381,gluonts-DeepAR-1h-10-8736-bt4 93 | 20241107-214657,QuantileLoss[0.3],989605.915557466,gluonts-DeepAR-1h-10-8736-bt4 94 | 20241107-214657,Coverage[0.3],0.10758928571428572,gluonts-DeepAR-1h-10-8736-bt4 95 | 20241107-214657,QuantileLoss[0.4],1055248.679701458,gluonts-DeepAR-1h-10-8736-bt4 96 | 20241107-214657,Coverage[0.4],0.1483630952380952,gluonts-DeepAR-1h-10-8736-bt4 97 | 20241107-214657,QuantileLoss[0.5],1078050.1977001084,gluonts-DeepAR-1h-10-8736-bt4 98 | 20241107-214657,Coverage[0.5],0.19806547619047618,gluonts-DeepAR-1h-10-8736-bt4 99 | 20241107-214657,QuantileLoss[0.6],1111712.389153796,gluonts-DeepAR-1h-10-8736-bt4 100 | 20241107-214657,Coverage[0.6],0.23080357142857144,gluonts-DeepAR-1h-10-8736-bt4 101 | 20241107-214657,QuantileLoss[0.7],1034245.9985696841,gluonts-DeepAR-1h-10-8736-bt4 102 | 20241107-214657,Coverage[0.7],0.30044642857142856,gluonts-DeepAR-1h-10-8736-bt4 103 | 20241107-214657,QuantileLoss[0.8],884955.4130222801,gluonts-DeepAR-1h-10-8736-bt4 104 | 20241107-214657,Coverage[0.8],0.40476190476190477,gluonts-DeepAR-1h-10-8736-bt4 105 | 20241107-214657,QuantileLoss[0.9],649305.9710002927,gluonts-DeepAR-1h-10-8736-bt4 106 | 20241107-214657,Coverage[0.9],0.5428571428571429,gluonts-DeepAR-1h-10-8736-bt4 107 | 20241107-214657,RMSE,719.273088702308,gluonts-DeepAR-1h-10-8736-bt4 108 | 20241107-214657,NRMSE,0.6164435587547005,gluonts-DeepAR-1h-10-8736-bt4 109 | 20241107-214657,ND,0.13748940034885276,gluonts-DeepAR-1h-10-8736-bt4 110 | 20241107-214657,wQuantileLoss[0.1],0.08993062040118581,gluonts-DeepAR-1h-10-8736-bt4 111 | 20241107-214657,wQuantileLoss[0.2],0.11159576378536071,gluonts-DeepAR-1h-10-8736-bt4 112 | 20241107-214657,wQuantileLoss[0.3],0.12620963680721167,gluonts-DeepAR-1h-10-8736-bt4 113 | 20241107-214657,wQuantileLoss[0.4],0.13458140307436026,gluonts-DeepAR-1h-10-8736-bt4 114 | 20241107-214657,wQuantileLoss[0.5],0.13748940034885276,gluonts-DeepAR-1h-10-8736-bt4 115 | 20241107-214657,wQuantileLoss[0.6],0.14178251631624417,gluonts-DeepAR-1h-10-8736-bt4 116 | 20241107-214657,wQuantileLoss[0.7],0.13190282090751293,gluonts-DeepAR-1h-10-8736-bt4 117 | 20241107-214657,wQuantileLoss[0.8],0.11286300891320025,gluonts-DeepAR-1h-10-8736-bt4 118 | 20241107-214657,wQuantileLoss[0.9],0.08280939865899795,gluonts-DeepAR-1h-10-8736-bt4 119 | 20241107-214657,mean_absolute_QuantileLoss,931476.3112157607,gluonts-DeepAR-1h-10-8736-bt4 120 | 20241107-214657,mean_wQuantileLoss,0.11879606324588073,gluonts-DeepAR-1h-10-8736-bt4 121 | 20241107-214657,MAE_Coverage,0.226984126984127,gluonts-DeepAR-1h-10-8736-bt4 122 | 20241107-214657,MSE,591669.990028578,gluonts-GaussianProcess-1h-10-8736-bt4 123 | 20241107-214657,abs_error,1033484.9379525569,gluonts-GaussianProcess-1h-10-8736-bt4 124 | 20241107-214657,abs_target_sum,7840969.521757782,gluonts-GaussianProcess-1h-10-8736-bt4 125 | 20241107-214657,abs_target_mean,1166.8109407377653,gluonts-GaussianProcess-1h-10-8736-bt4 126 | 20241107-214657,MAPE,0.24041505643511182,gluonts-GaussianProcess-1h-10-8736-bt4 127 | 20241107-214657,sMAPE,0.22196994519295205,gluonts-GaussianProcess-1h-10-8736-bt4 128 | 20241107-214657,num_masked_target_values,0.0,gluonts-GaussianProcess-1h-10-8736-bt4 129 | 20241107-214657,QuantileLoss[0.1],782468.5503951254,gluonts-GaussianProcess-1h-10-8736-bt4 130 | 20241107-214657,Coverage[0.1],0.10922619047619046,gluonts-GaussianProcess-1h-10-8736-bt4 131 | 20241107-214657,QuantileLoss[0.2],968870.0443387064,gluonts-GaussianProcess-1h-10-8736-bt4 132 | 20241107-214657,Coverage[0.2],0.19166666666666668,gluonts-GaussianProcess-1h-10-8736-bt4 133 | 20241107-214657,QuantileLoss[0.3],1061806.0325222253,gluonts-GaussianProcess-1h-10-8736-bt4 134 | 20241107-214657,Coverage[0.3],0.2922619047619048,gluonts-GaussianProcess-1h-10-8736-bt4 135 | 20241107-214657,QuantileLoss[0.4],1078522.306882761,gluonts-GaussianProcess-1h-10-8736-bt4 136 | 20241107-214657,Coverage[0.4],0.40714285714285714,gluonts-GaussianProcess-1h-10-8736-bt4 137 | 20241107-214657,QuantileLoss[0.5],1033484.9379525569,gluonts-GaussianProcess-1h-10-8736-bt4 138 | 20241107-214657,Coverage[0.5],0.5357142857142857,gluonts-GaussianProcess-1h-10-8736-bt4 139 | 20241107-214657,QuantileLoss[0.6],974092.8959604588,gluonts-GaussianProcess-1h-10-8736-bt4 140 | 20241107-214657,Coverage[0.6],0.5928571428571429,gluonts-GaussianProcess-1h-10-8736-bt4 141 | 20241107-214657,QuantileLoss[0.7],849769.6314119821,gluonts-GaussianProcess-1h-10-8736-bt4 142 | 20241107-214657,Coverage[0.7],0.7016369047619048,gluonts-GaussianProcess-1h-10-8736-bt4 143 | 20241107-214657,QuantileLoss[0.8],679248.2890064333,gluonts-GaussianProcess-1h-10-8736-bt4 144 | 20241107-214657,Coverage[0.8],0.8063988095238095,gluonts-GaussianProcess-1h-10-8736-bt4 145 | 20241107-214657,QuantileLoss[0.9],425854.36995146907,gluonts-GaussianProcess-1h-10-8736-bt4 146 | 20241107-214657,Coverage[0.9],0.8879464285714287,gluonts-GaussianProcess-1h-10-8736-bt4 147 | 20241107-214657,RMSE,769.2008775531773,gluonts-GaussianProcess-1h-10-8736-bt4 148 | 20241107-214657,NRMSE,0.659233514785881,gluonts-GaussianProcess-1h-10-8736-bt4 149 | 20241107-214657,ND,0.13180575885223833,gluonts-GaussianProcess-1h-10-8736-bt4 150 | 20241107-214657,wQuantileLoss[0.1],0.09979232137350691,gluonts-GaussianProcess-1h-10-8736-bt4 151 | 20241107-214657,wQuantileLoss[0.2],0.12356508230904409,gluonts-GaussianProcess-1h-10-8736-bt4 152 | 20241107-214657,wQuantileLoss[0.3],0.13541769669883763,gluonts-GaussianProcess-1h-10-8736-bt4 153 | 20241107-214657,wQuantileLoss[0.4],0.13754961091099596,gluonts-GaussianProcess-1h-10-8736-bt4 154 | 20241107-214657,wQuantileLoss[0.5],0.13180575885223833,gluonts-GaussianProcess-1h-10-8736-bt4 155 | 20241107-214657,wQuantileLoss[0.6],0.12423117999087534,gluonts-GaussianProcess-1h-10-8736-bt4 156 | 20241107-214657,wQuantileLoss[0.7],0.10837558149588132,gluonts-GaussianProcess-1h-10-8736-bt4 157 | 20241107-214657,wQuantileLoss[0.8],0.08662809964017817,gluonts-GaussianProcess-1h-10-8736-bt4 158 | 20241107-214657,wQuantileLoss[0.9],0.05431144309001234,gluonts-GaussianProcess-1h-10-8736-bt4 159 | 20241107-214657,mean_absolute_QuantileLoss,872679.6731579688,gluonts-GaussianProcess-1h-10-8736-bt4 160 | 20241107-214657,mean_wQuantileLoss,0.11129741937350777,gluonts-GaussianProcess-1h-10-8736-bt4 161 | 20241107-214657,MAE_Coverage,0.3906250000000001,gluonts-GaussianProcess-1h-10-8736-bt4 162 | 20241107-214657,MSE,506420.21957510506,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 163 | 20241107-214657,abs_error,904810.0186330621,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 164 | 20241107-214657,abs_target_sum,7840969.521757782,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 165 | 20241107-214657,abs_target_mean,1166.8109407377653,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 166 | 20241107-214657,MAPE,0.303919967670033,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 167 | 20241107-214657,sMAPE,0.29610526186293484,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 168 | 20241107-214657,num_masked_target_values,0.0,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 169 | 20241107-214657,QuantileLoss[0.1],702962.5078474184,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 170 | 20241107-214657,Coverage[0.1],0.1418154761904762,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 171 | 20241107-214657,QuantileLoss[0.2],843698.9754430054,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 172 | 20241107-214657,Coverage[0.2],0.1842261904761905,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 173 | 20241107-214657,QuantileLoss[0.3],923172.4710499098,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 174 | 20241107-214657,Coverage[0.3],0.24017857142857144,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 175 | 20241107-214657,QuantileLoss[0.4],939963.6876435605,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 176 | 20241107-214657,Coverage[0.4],0.31101190476190477,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 177 | 20241107-214657,QuantileLoss[0.5],904810.0186330621,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 178 | 20241107-214657,Coverage[0.5],0.3949404761904762,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 179 | 20241107-214657,QuantileLoss[0.6],817274.2637810577,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 180 | 20241107-214657,Coverage[0.6],0.5181547619047618,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 181 | 20241107-214657,QuantileLoss[0.7],712211.058519595,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 182 | 20241107-214657,Coverage[0.7],0.637202380952381,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 183 | 20241107-214657,QuantileLoss[0.8],558132.5039513422,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 184 | 20241107-214657,Coverage[0.8],0.7220238095238095,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 185 | 20241107-214657,QuantileLoss[0.9],341950.85405915393,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 186 | 20241107-214657,Coverage[0.9],0.7910714285714285,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 187 | 20241107-214657,RMSE,711.6320816089625,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 188 | 20241107-214657,NRMSE,0.6098949339290589,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 189 | 20241107-214657,ND,0.11539517098265963,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 190 | 20241107-214657,wQuantileLoss[0.1],0.08965249844381858,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 191 | 20241107-214657,wQuantileLoss[0.2],0.10760135887556233,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 192 | 20241107-214657,wQuantileLoss[0.3],0.11773703092305271,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 193 | 20241107-214657,wQuantileLoss[0.4],0.11987850291156854,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 194 | 20241107-214657,wQuantileLoss[0.5],0.11539517098265963,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 195 | 20241107-214657,wQuantileLoss[0.6],0.10423127669521176,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 196 | 20241107-214657,wQuantileLoss[0.7],0.09083201465626052,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 197 | 20241107-214657,wQuantileLoss[0.8],0.0711815678408887,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 198 | 20241107-214657,wQuantileLoss[0.9],0.043610787302549754,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 199 | 20241107-214657,mean_absolute_QuantileLoss,749352.9267697895,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 200 | 20241107-214657,mean_wQuantileLoss,0.09556891207017472,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 201 | 20241107-214657,MAE_Coverage,0.31726190476190474,gluonts-TemporalFusionTransformer-1h-10-8736-bt4 202 | 20241107-214657,MSE,495405.71600424324,gluonts-MQCNN-1h-10-8736-bt4 203 | 20241107-214657,abs_error,809684.6485752523,gluonts-MQCNN-1h-10-8736-bt4 204 | 20241107-214657,abs_target_sum,7840969.521757782,gluonts-MQCNN-1h-10-8736-bt4 205 | 20241107-214657,abs_target_mean,1166.8109407377653,gluonts-MQCNN-1h-10-8736-bt4 206 | 20241107-214657,MAPE,0.2684776516556477,gluonts-MQCNN-1h-10-8736-bt4 207 | 20241107-214657,sMAPE,0.20535807771505876,gluonts-MQCNN-1h-10-8736-bt4 208 | 20241107-214657,num_masked_target_values,0.0,gluonts-MQCNN-1h-10-8736-bt4 209 | 20241107-214657,QuantileLoss[0.1],742749.07022983,gluonts-MQCNN-1h-10-8736-bt4 210 | 20241107-214657,Coverage[0.1],0.16577380952380955,gluonts-MQCNN-1h-10-8736-bt4 211 | 20241107-214657,QuantileLoss[0.2],844471.187695416,gluonts-MQCNN-1h-10-8736-bt4 212 | 20241107-214657,Coverage[0.2],0.2513392857142857,gluonts-MQCNN-1h-10-8736-bt4 213 | 20241107-214657,QuantileLoss[0.3],866008.1884494075,gluonts-MQCNN-1h-10-8736-bt4 214 | 20241107-214657,Coverage[0.3],0.2888392857142857,gluonts-MQCNN-1h-10-8736-bt4 215 | 20241107-214657,QuantileLoss[0.4],869947.2416351496,gluonts-MQCNN-1h-10-8736-bt4 216 | 20241107-214657,Coverage[0.4],0.4321428571428571,gluonts-MQCNN-1h-10-8736-bt4 217 | 20241107-214657,QuantileLoss[0.5],809684.6485752523,gluonts-MQCNN-1h-10-8736-bt4 218 | 20241107-214657,Coverage[0.5],0.5202380952380953,gluonts-MQCNN-1h-10-8736-bt4 219 | 20241107-214657,QuantileLoss[0.6],743172.5363041362,gluonts-MQCNN-1h-10-8736-bt4 220 | 20241107-214657,Coverage[0.6],0.5453869047619048,gluonts-MQCNN-1h-10-8736-bt4 221 | 20241107-214657,QuantileLoss[0.7],627556.8249516919,gluonts-MQCNN-1h-10-8736-bt4 222 | 20241107-214657,Coverage[0.7],0.7233630952380953,gluonts-MQCNN-1h-10-8736-bt4 223 | 20241107-214657,QuantileLoss[0.8],485640.02672345354,gluonts-MQCNN-1h-10-8736-bt4 224 | 20241107-214657,Coverage[0.8],0.8257440476190474,gluonts-MQCNN-1h-10-8736-bt4 225 | 20241107-214657,QuantileLoss[0.9],301772.67409519444,gluonts-MQCNN-1h-10-8736-bt4 226 | 20241107-214657,Coverage[0.9],0.8595238095238095,gluonts-MQCNN-1h-10-8736-bt4 227 | 20241107-214657,RMSE,703.8506347260357,gluonts-MQCNN-1h-10-8736-bt4 228 | 20241107-214657,NRMSE,0.6032259470253136,gluonts-MQCNN-1h-10-8736-bt4 229 | 20241107-214657,ND,0.10326333323047247,gluonts-MQCNN-1h-10-8736-bt4 230 | 20241107-214657,wQuantileLoss[0.1],0.0947266875822929,gluonts-MQCNN-1h-10-8736-bt4 231 | 20241107-214657,wQuantileLoss[0.2],0.10769984315741903,gluonts-MQCNN-1h-10-8736-bt4 232 | 20241107-214657,wQuantileLoss[0.3],0.11044656991030702,gluonts-MQCNN-1h-10-8736-bt4 233 | 20241107-214657,wQuantileLoss[0.4],0.11094893803899464,gluonts-MQCNN-1h-10-8736-bt4 234 | 20241107-214657,wQuantileLoss[0.5],0.10326333323047247,gluonts-MQCNN-1h-10-8736-bt4 235 | 20241107-214657,wQuantileLoss[0.6],0.0947806944335032,gluonts-MQCNN-1h-10-8736-bt4 236 | 20241107-214657,wQuantileLoss[0.7],0.08003561590314749,gluonts-MQCNN-1h-10-8736-bt4 237 | 20241107-214657,wQuantileLoss[0.8],0.061936221710320226,gluonts-MQCNN-1h-10-8736-bt4 238 | 20241107-214657,wQuantileLoss[0.9],0.03848665311831786,gluonts-MQCNN-1h-10-8736-bt4 239 | 20241107-214657,mean_absolute_QuantileLoss,699000.2665177258,gluonts-MQCNN-1h-10-8736-bt4 240 | 20241107-214657,mean_wQuantileLoss,0.08914717300941942,gluonts-MQCNN-1h-10-8736-bt4 241 | 20241107-214657,MAE_Coverage,0.36851851851851847,gluonts-MQCNN-1h-10-8736-bt4 242 | 20241107-214657,MSE,602056.1989273061,gluonts-SeasonalNaive-1h-10-8736-bt4 243 | 20241107-214657,abs_error,968718.28635148,gluonts-SeasonalNaive-1h-10-8736-bt4 244 | 20241107-214657,abs_target_sum,7840969.521757782,gluonts-SeasonalNaive-1h-10-8736-bt4 245 | 20241107-214657,abs_target_mean,1166.8109407377653,gluonts-SeasonalNaive-1h-10-8736-bt4 246 | 20241107-214657,MAPE,0.22112640814653392,gluonts-SeasonalNaive-1h-10-8736-bt4 247 | 20241107-214657,sMAPE,0.18031426801961994,gluonts-SeasonalNaive-1h-10-8736-bt4 248 | 20241107-214657,num_masked_target_values,0.0,gluonts-SeasonalNaive-1h-10-8736-bt4 249 | 20241107-214657,QuantileLoss[0.1],1214573.4515250581,gluonts-SeasonalNaive-1h-10-8736-bt4 250 | 20241107-214657,Coverage[0.1],0.5313988095238095,gluonts-SeasonalNaive-1h-10-8736-bt4 251 | 20241107-214657,QuantileLoss[0.2],1153109.6602316636,gluonts-SeasonalNaive-1h-10-8736-bt4 252 | 20241107-214657,Coverage[0.2],0.5313988095238095,gluonts-SeasonalNaive-1h-10-8736-bt4 253 | 20241107-214657,QuantileLoss[0.3],1091645.868938269,gluonts-SeasonalNaive-1h-10-8736-bt4 254 | 20241107-214657,Coverage[0.3],0.5313988095238095,gluonts-SeasonalNaive-1h-10-8736-bt4 255 | 20241107-214657,QuantileLoss[0.4],1030182.0776448745,gluonts-SeasonalNaive-1h-10-8736-bt4 256 | 20241107-214657,Coverage[0.4],0.5313988095238095,gluonts-SeasonalNaive-1h-10-8736-bt4 257 | 20241107-214657,QuantileLoss[0.5],968718.28635148,gluonts-SeasonalNaive-1h-10-8736-bt4 258 | 20241107-214657,Coverage[0.5],0.5313988095238095,gluonts-SeasonalNaive-1h-10-8736-bt4 259 | 20241107-214657,QuantileLoss[0.6],907254.4950580857,gluonts-SeasonalNaive-1h-10-8736-bt4 260 | 20241107-214657,Coverage[0.6],0.5313988095238095,gluonts-SeasonalNaive-1h-10-8736-bt4 261 | 20241107-214657,QuantileLoss[0.7],845790.7037646911,gluonts-SeasonalNaive-1h-10-8736-bt4 262 | 20241107-214657,Coverage[0.7],0.5313988095238095,gluonts-SeasonalNaive-1h-10-8736-bt4 263 | 20241107-214657,QuantileLoss[0.8],784326.9124712966,gluonts-SeasonalNaive-1h-10-8736-bt4 264 | 20241107-214657,Coverage[0.8],0.5313988095238095,gluonts-SeasonalNaive-1h-10-8736-bt4 265 | 20241107-214657,QuantileLoss[0.9],722863.1211779023,gluonts-SeasonalNaive-1h-10-8736-bt4 266 | 20241107-214657,Coverage[0.9],0.5313988095238095,gluonts-SeasonalNaive-1h-10-8736-bt4 267 | 20241107-214657,RMSE,775.9228047475509,gluonts-SeasonalNaive-1h-10-8736-bt4 268 | 20241107-214657,NRMSE,0.6649944542488958,gluonts-SeasonalNaive-1h-10-8736-bt4 269 | 20241107-214657,ND,0.12354572781636237,gluonts-SeasonalNaive-1h-10-8736-bt4 270 | 20241107-214657,wQuantileLoss[0.1],0.15490092751346085,gluonts-SeasonalNaive-1h-10-8736-bt4 271 | 20241107-214657,wQuantileLoss[0.2],0.14706212758918624,gluonts-SeasonalNaive-1h-10-8736-bt4 272 | 20241107-214657,wQuantileLoss[0.3],0.1392233276649116,gluonts-SeasonalNaive-1h-10-8736-bt4 273 | 20241107-214657,wQuantileLoss[0.4],0.13138452774063697,gluonts-SeasonalNaive-1h-10-8736-bt4 274 | 20241107-214657,wQuantileLoss[0.5],0.12354572781636237,gluonts-SeasonalNaive-1h-10-8736-bt4 275 | 20241107-214657,wQuantileLoss[0.6],0.11570692789208777,gluonts-SeasonalNaive-1h-10-8736-bt4 276 | 20241107-214657,wQuantileLoss[0.7],0.10786812796781314,gluonts-SeasonalNaive-1h-10-8736-bt4 277 | 20241107-214657,wQuantileLoss[0.8],0.10002932804353852,gluonts-SeasonalNaive-1h-10-8736-bt4 278 | 20241107-214657,wQuantileLoss[0.9],0.09219052811926394,gluonts-SeasonalNaive-1h-10-8736-bt4 279 | 20241107-214657,mean_absolute_QuantileLoss,968718.28635148,gluonts-SeasonalNaive-1h-10-8736-bt4 280 | 20241107-214657,mean_wQuantileLoss,0.12354572781636236,gluonts-SeasonalNaive-1h-10-8736-bt4 281 | 20241107-214657,MAE_Coverage,0.2257109788359788,gluonts-SeasonalNaive-1h-10-8736-bt4 282 | 20241107-214657,MSE,513524.1753865841,gluonts-Prophet-1h-10-8736-bt4 283 | 20241107-214657,abs_error,893692.8004313651,gluonts-Prophet-1h-10-8736-bt4 284 | 20241107-214657,abs_target_sum,7840969.521757782,gluonts-Prophet-1h-10-8736-bt4 285 | 20241107-214657,abs_target_mean,1166.8109407377653,gluonts-Prophet-1h-10-8736-bt4 286 | 20241107-214657,MAPE,0.3162780346209607,gluonts-Prophet-1h-10-8736-bt4 287 | 20241107-214657,sMAPE,0.3092261665660643,gluonts-Prophet-1h-10-8736-bt4 288 | 20241107-214657,num_masked_target_values,0.0,gluonts-Prophet-1h-10-8736-bt4 289 | 20241107-214657,QuantileLoss[0.1],694077.0088892874,gluonts-Prophet-1h-10-8736-bt4 290 | 20241107-214657,Coverage[0.1],0.11607142857142856,gluonts-Prophet-1h-10-8736-bt4 291 | 20241107-214657,QuantileLoss[0.2],826341.396470925,gluonts-Prophet-1h-10-8736-bt4 292 | 20241107-214657,Coverage[0.2],0.18705357142857143,gluonts-Prophet-1h-10-8736-bt4 293 | 20241107-214657,QuantileLoss[0.3],901139.5703235045,gluonts-Prophet-1h-10-8736-bt4 294 | 20241107-214657,Coverage[0.3],0.2663690476190476,gluonts-Prophet-1h-10-8736-bt4 295 | 20241107-214657,QuantileLoss[0.4],919363.0257825678,gluonts-Prophet-1h-10-8736-bt4 296 | 20241107-214657,Coverage[0.4],0.3546130952380952,gluonts-Prophet-1h-10-8736-bt4 297 | 20241107-214657,QuantileLoss[0.5],893692.8004313651,gluonts-Prophet-1h-10-8736-bt4 298 | 20241107-214657,Coverage[0.5],0.44538690476190473,gluonts-Prophet-1h-10-8736-bt4 299 | 20241107-214657,QuantileLoss[0.6],860659.0832785695,gluonts-Prophet-1h-10-8736-bt4 300 | 20241107-214657,Coverage[0.6],0.49598214285714287,gluonts-Prophet-1h-10-8736-bt4 301 | 20241107-214657,QuantileLoss[0.7],777101.5853908068,gluonts-Prophet-1h-10-8736-bt4 302 | 20241107-214657,Coverage[0.7],0.5910714285714286,gluonts-Prophet-1h-10-8736-bt4 303 | 20241107-214657,QuantileLoss[0.8],639399.2479384069,gluonts-Prophet-1h-10-8736-bt4 304 | 20241107-214657,Coverage[0.8],0.6912202380952381,gluonts-Prophet-1h-10-8736-bt4 305 | 20241107-214657,QuantileLoss[0.9],438213.497574374,gluonts-Prophet-1h-10-8736-bt4 306 | 20241107-214657,Coverage[0.9],0.7946428571428571,gluonts-Prophet-1h-10-8736-bt4 307 | 20241107-214657,RMSE,716.6060112688032,gluonts-Prophet-1h-10-8736-bt4 308 | 20241107-214657,NRMSE,0.614157775051114,gluonts-Prophet-1h-10-8736-bt4 309 | 20241107-214657,ND,0.11397733379162757,gluonts-Prophet-1h-10-8736-bt4 310 | 20241107-214657,wQuantileLoss[0.1],0.08851928412211067,gluonts-Prophet-1h-10-8736-bt4 311 | 20241107-214657,wQuantileLoss[0.2],0.10538765572011513,gluonts-Prophet-1h-10-8736-bt4 312 | 20241107-214657,wQuantileLoss[0.3],0.11492705944372651,gluonts-Prophet-1h-10-8736-bt4 313 | 20241107-214657,wQuantileLoss[0.4],0.11725119237250468,gluonts-Prophet-1h-10-8736-bt4 314 | 20241107-214657,wQuantileLoss[0.5],0.11397733379162757,gluonts-Prophet-1h-10-8736-bt4 315 | 20241107-214657,wQuantileLoss[0.6],0.10976437045066177,gluonts-Prophet-1h-10-8736-bt4 316 | 20241107-214657,wQuantileLoss[0.7],0.09910784415555243,gluonts-Prophet-1h-10-8736-bt4 317 | 20241107-214657,wQuantileLoss[0.8],0.08154594226697962,gluonts-Prophet-1h-10-8736-bt4 318 | 20241107-214657,wQuantileLoss[0.9],0.055887667508256765,gluonts-Prophet-1h-10-8736-bt4 319 | 20241107-214657,mean_absolute_QuantileLoss,772220.8017866452,gluonts-Prophet-1h-10-8736-bt4 320 | 20241107-214657,mean_wQuantileLoss,0.09848537220350391,gluonts-Prophet-1h-10-8736-bt4 321 | 20241107-214657,MAE_Coverage,0.3192460317460318,gluonts-Prophet-1h-10-8736-bt4 322 | 20241107-214657,MSE,497407.51645393326,gluonts-NPTS-1h-10-8736-bt4 323 | 20241107-214657,abs_error,753775.6228687668,gluonts-NPTS-1h-10-8736-bt4 324 | 20241107-214657,abs_target_sum,7840969.521757782,gluonts-NPTS-1h-10-8736-bt4 325 | 20241107-214657,abs_target_mean,1166.8109407377653,gluonts-NPTS-1h-10-8736-bt4 326 | 20241107-214657,MAPE,0.2222654429580105,gluonts-NPTS-1h-10-8736-bt4 327 | 20241107-214657,sMAPE,0.1743582554083264,gluonts-NPTS-1h-10-8736-bt4 328 | 20241107-214657,num_masked_target_values,0.0,gluonts-NPTS-1h-10-8736-bt4 329 | 20241107-214657,QuantileLoss[0.1],617423.9344890051,gluonts-NPTS-1h-10-8736-bt4 330 | 20241107-214657,Coverage[0.1],0.22752976190476196,gluonts-NPTS-1h-10-8736-bt4 331 | 20241107-214657,QuantileLoss[0.2],701661.2470645618,gluonts-NPTS-1h-10-8736-bt4 332 | 20241107-214657,Coverage[0.2],0.32559523809523805,gluonts-NPTS-1h-10-8736-bt4 333 | 20241107-214657,QuantileLoss[0.3],770585.2733677226,gluonts-NPTS-1h-10-8736-bt4 334 | 20241107-214657,Coverage[0.3],0.409970238095238,gluonts-NPTS-1h-10-8736-bt4 335 | 20241107-214657,QuantileLoss[0.4],766608.1028513353,gluonts-NPTS-1h-10-8736-bt4 336 | 20241107-214657,Coverage[0.4],0.4927083333333334,gluonts-NPTS-1h-10-8736-bt4 337 | 20241107-214657,QuantileLoss[0.5],753775.6228687668,gluonts-NPTS-1h-10-8736-bt4 338 | 20241107-214657,Coverage[0.5],0.5729166666666666,gluonts-NPTS-1h-10-8736-bt4 339 | 20241107-214657,QuantileLoss[0.6],698826.1362233879,gluonts-NPTS-1h-10-8736-bt4 340 | 20241107-214657,Coverage[0.6],0.6178571428571429,gluonts-NPTS-1h-10-8736-bt4 341 | 20241107-214657,QuantileLoss[0.7],595409.926329624,gluonts-NPTS-1h-10-8736-bt4 342 | 20241107-214657,Coverage[0.7],0.6994047619047619,gluonts-NPTS-1h-10-8736-bt4 343 | 20241107-214657,QuantileLoss[0.8],456182.5509127008,gluonts-NPTS-1h-10-8736-bt4 344 | 20241107-214657,Coverage[0.8],0.7811011904761905,gluonts-NPTS-1h-10-8736-bt4 345 | 20241107-214657,QuantileLoss[0.9],276380.42531627364,gluonts-NPTS-1h-10-8736-bt4 346 | 20241107-214657,Coverage[0.9],0.8532738095238095,gluonts-NPTS-1h-10-8736-bt4 347 | 20241107-214657,RMSE,705.2712360885939,gluonts-NPTS-1h-10-8736-bt4 348 | 20241107-214657,NRMSE,0.6044434547748211,gluonts-NPTS-1h-10-8736-bt4 349 | 20241107-214657,ND,0.09613296171820675,gluonts-NPTS-1h-10-8736-bt4 350 | 20241107-214657,wQuantileLoss[0.1],0.07874331519536266,gluonts-NPTS-1h-10-8736-bt4 351 | 20241107-214657,wQuantileLoss[0.2],0.08948654182592256,gluonts-NPTS-1h-10-8736-bt4 352 | 20241107-214657,wQuantileLoss[0.3],0.09827678467942488,gluonts-NPTS-1h-10-8736-bt4 353 | 20241107-214657,wQuantileLoss[0.4],0.09776955524748394,gluonts-NPTS-1h-10-8736-bt4 354 | 20241107-214657,wQuantileLoss[0.5],0.09613296171820675,gluonts-NPTS-1h-10-8736-bt4 355 | 20241107-214657,wQuantileLoss[0.6],0.08912496525898057,gluonts-NPTS-1h-10-8736-bt4 356 | 20241107-214657,wQuantileLoss[0.7],0.07593575318427528,gluonts-NPTS-1h-10-8736-bt4 357 | 20241107-214657,wQuantileLoss[0.8],0.05817935519923233,gluonts-NPTS-1h-10-8736-bt4 358 | 20241107-214657,wQuantileLoss[0.9],0.03524824634879015,gluonts-NPTS-1h-10-8736-bt4 359 | 20241107-214657,mean_absolute_QuantileLoss,626317.0243803752,gluonts-NPTS-1h-10-8736-bt4 360 | 20241107-214657,mean_wQuantileLoss,0.07987749762863101,gluonts-NPTS-1h-10-8736-bt4 361 | 20241107-214657,MAE_Coverage,0.3636574074074074,gluonts-NPTS-1h-10-8736-bt4 362 | --------------------------------------------------------------------------------