{"id":18508853,"url":"https://github.com/hfawaz/ijcnn19ensemble","last_synced_at":"2025-04-09T03:32:16.989Z","repository":{"id":45798345,"uuid":"174340948","full_name":"hfawaz/ijcnn19ensemble","owner":"hfawaz","description":"Deep Neural Network Ensembles for Time Series Classification","archived":false,"fork":false,"pushed_at":"2023-08-31T14:50:05.000Z","size":314,"stargazers_count":111,"open_issues_count":1,"forks_count":38,"subscribers_count":9,"default_branch":"master","last_synced_at":"2025-03-23T22:42:27.374Z","etag":null,"topics":["deep-learning","deep-neural-networks","ensemble","time-series-classification"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"gpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/hfawaz.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null}},"created_at":"2019-03-07T12:34:32.000Z","updated_at":"2024-11-13T13:18:38.000Z","dependencies_parsed_at":"2022-07-17T01:16:08.980Z","dependency_job_id":"694400d3-5d38-4bb9-b72a-3e8e163c21b7","html_url":"https://github.com/hfawaz/ijcnn19ensemble","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hfawaz%2Fijcnn19ensemble","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hfawaz%2Fijcnn19ensemble/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hfawaz%2Fijcnn19ensemble/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hfawaz%2Fijcnn19ensemble/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/hfawaz","download_url":"https://codeload.github.com/hfawaz/ijcnn19ensemble/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247973930,"owners_count":21026736,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["deep-learning","deep-neural-networks","ensemble","time-series-classification"],"created_at":"2024-11-06T15:15:37.974Z","updated_at":"2025-04-09T03:32:16.698Z","avatar_url":"https://github.com/hfawaz.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Deep Neural Network Ensembles for Time Series Classification\nThis is the companion repository for our paper also available on [ArXiv](https://arxiv.org/abs/1903.06602) titled \"Deep Neural Network Ensembles for Time Series Classification\". This paper has been accepted at the [IEEE International Joint Conference on Neural Networks (IJCNN) 2019](https://www.ijcnn.org/). \n\n## Approach\n![ensemble](https://github.com/hfawaz/ijcnn19ensemble/blob/master/png/ensemble.png)\n\n## Data \nThe data used in this project comes from the [UCR/UEA archive](http://timeseriesclassification.com/TSC.zip), which contains the 85 univariate time series datasets. \n\n## Code \nThe code is divided as follows: \n* The [main.py](https://github.com/hfawaz/ijcnn19ensemble/blob/master/src/main.py) python file contains the necessary code to run all experiements. \n* The [utils](https://github.com/hfawaz/ijcnn19ensemble/blob/master/src/utils/) folder contains the necessary functions to read the datasets and manipulate the data.\n* The [classifiers](https://github.com/hfawaz/ijcnn19ensemble/tree/master/src/classifiers) folder contains eight python files one for each deep individual/ensemble classifier presented in our paper. \n\nTo run a model on all datasets you should issue the following command: \n```\npython3 main.py\n```\nTo control which datasets and which individual/ensemble classifiers to run see the options in [constants.py](https://github.com/hfawaz/ijcnn19ensemble/blob/master/src/utils/constants.py). \n\nYou can control which algorithms to include in the ensemble by changing [this line of code](https://github.com/hfawaz/ijcnn19ensemble/blob/cb822a0783ea6bd10359348f727b8fd81ae2c131/src/classifiers/nne.py#L35). \n\n## Prerequisites\nAll python packages needed are listed in [pip-requirements.txt](https://github.com/hfawaz/ijcnn19ensemble/blob/master/src/utils/pip-requirements.txt) file and can be installed simply using the pip command.\n\n* [numpy](http://www.numpy.org/) \n* [pandas](https://pandas.pydata.org/) \n* [sklearn](http://scikit-learn.org/stable/) \n* [scipy](https://www.scipy.org/) \n* [matplotlib](https://matplotlib.org/) \n* [tensorflow-gpu](https://www.tensorflow.org/) \n* [keras](https://keras.io/) \n* [h5py](http://docs.h5py.org/en/latest/build.html)\n* [keras_contrib](https://www.github.com/keras-team/keras-contrib.git)\n\n## Results\nThe following table shows the results of four ensembles, the raw results can be found [here](https://github.com/hfawaz/ijcnn19ensemble/blob/master/results/results.csv). \n\n| | Fine-tuned FCNs | NNE | ALL | ResNets | \n|--------------------------------|-----------------|--------|--------|---------| \n| 50words | 66.81 | **80.00** | **80.00** | 77.14 | \n| Adiac | **85.17** | **85.17** | 83.38 | 83.63 | \n| ArrowHead | 84.00 | 86.29 | 86.29 | **86.86** | \n| Beef | 76.67 | 76.67 | **80.00** | 76.67 | \n| BeetleFly | **90.00** | 85.00 | 85.00 | 85.00 | \n| BirdChicken | 90.00 | **95.00** | 85.00 | 90.00 | \n| CBF | **99.78** | 99.44 | 98.56 | **99.78** | \n| Car | 91.67 | **95.00** | 86.67 | 93.33 | \n| ChlorineConcentration | 82.42 | 85.05 | 83.98 | **85.49** | \n| CinC_ECG_torso | 85.87 | 89.71 | **92.90** | 83.55 | \n| Coffee | **100.00** | **100.00** | **100.00** | **100.00** | \n| Computers | 83.20 | **83.60** | 71.60 | **83.60** | \n| Cricket_X | 78.97 | **82.05** | 77.95 | 81.54 | \n| Cricket_Y | 79.23 | **84.36** | 78.72 | 82.05 | \n| Cricket_Z | 82.05 | **83.85** | 79.49 | 82.05 | \n| DiatomSizeReduction | 30.07 | 30.07 | **88.56** | 30.07 | \n| DistalPhalanxOutlineAgeGroup | 71.94 | 72.66 | **76.26** | 73.38 | \n| DistalPhalanxOutlineCorrect | 77.54 | 77.90 | 77.90 | **78.99** | \n| DistalPhalanxTW | **71.22** | 65.47 | 67.63 | 66.19 | \n| ECG200 | 89.00 | 89.00 | **92.00** | 88.00 | \n| ECG5000 | 94.16 | 94.42 | **94.51** | 93.67 | \n| ECGFiveDays | 99.54 | **99.88** | 99.65 | 98.61 | \n| Earthquakes | 71.94 | **74.82** | **74.82** | 72.66 | \n| ElectricDevices | 71.74 | **74.39** | 73.03 | 74.22 | \n| FISH | 96.00 | 97.71 | 93.71 | **98.29** | \n| FaceAll | **92.84** | 86.39 | 83.91 | 84.02 | \n| FaceFour | 93.18 | **95.45** | 92.05 | **95.45** | \n| FacesUCR | 93.95 | 95.76 | 95.51 | **95.90** | \n| FordA | 90.67 | 93.70 | **94.22** | 92.56 | \n| FordB | 88.04 | **92.90** | 92.33 | 92.16 | \n| Gun_Point | **100.00** | **100.00** | 99.33 | 99.33 | \n| Ham | 74.29 | 75.24 | 74.29 | **78.10** | \n| HandOutlines | 92.70 | **95.14** | 93.78 | 93.78 | \n| Haptics | 50.65 | 52.60 | 50.97 | **53.25** | \n| Herring | **65.62** | 60.94 | 62.50 | 60.94 | \n| InlineSkate | **40.55** | 38.36 | 38.00 | 38.55 | \n| InsectWingbeatSound | 39.49 | 59.75 | **65.91** | 52.73 | \n| ItalyPowerDemand | 96.11 | 96.50 | **96.89** | 96.40 | \n| LargeKitchenAppliances | 89.60 | **90.93** | 83.20 | 89.60 | \n| Lighting2 | **80.33** | **80.33** | 77.05 | 78.69 | \n| Lighting7 | 89.04 | **90.41** | 83.56 | 83.56 | \n| MALLAT | 96.93 | 96.93 | 95.44 | **97.40** | \n| Meat | 91.67 | 95.00 | 93.33 | **96.67** | \n| MedicalImages | 78.29 | 79.74 | **80.13** | 78.42 | \n| MiddlePhalanxOutlineAgeGroup | 53.90 | 59.09 | **60.39** | 59.09 | \n| MiddlePhalanxOutlineCorrect | 81.10 | 83.51 | **83.85** | 83.51 | \n| MiddlePhalanxTW | 51.95 | 51.95 | **55.19** | 49.35 | \n| MoteStrain | 93.37 | **93.93** | 93.45 | 93.05 | \n| NonInvasiveFatalECG_Thorax1 | **96.44** | 96.39 | 95.88 | 95.01 | \n| NonInvasiveFatalECG_Thorax2 | 95.73 | 96.18 | 96.54 | 95.01 | \n| OSULeaf | 97.52 | **98.76** | 78.51 | 98.35 | \n| OliveOil | **86.67** | **86.67** | **86.67** | **86.67** | \n| PhalangesOutlinesCorrect | 83.57 | 84.27 | 83.57 | **84.97** | \n| Phoneme | 32.65 | **35.13** | 30.91 | 34.81 | \n| Plane | **100.00** | **100.00** | 99.05 | **100.00** | \n| ProximalPhalanxOutlineAgeGroup | 84.39 | 84.88 | **85.85** | 85.37 | \n| ProximalPhalanxOutlineCorrect | **92.10** | 91.75 | 90.38 | **92.10** | \n| ProximalPhalanxTW | 79.51 | 77.56 | **80.98** | 78.54 | \n| RefrigerationDevices | 50.40 | 53.07 | **53.33** | 52.80 | \n| ScreenType | **65.07** | 62.13 | 52.27 | 62.13 | \n| ShapeletSim | 86.11 | 81.11 | 70.56 | **93.89** | \n| ShapesAll | 90.00 | **92.83** | 89.17 | 92.00 | \n| SmallKitchenAppliances | 79.47 | **82.13** | 77.60 | 78.93 | \n| SonyAIBORobotSurface | 95.84 | 94.68 | 78.04 | **96.17** | \n| SonyAIBORobotSurfaceII | **98.22** | 97.69 | 88.88 | 98.11 | \n| StarLightCurves | 96.78 | **97.92** | 97.79 | 97.38 | \n| Strawberry | 97.84 | **98.11** | 97.57 | **98.11** | \n| SwedishLeaf | **97.28** | **97.28** | 96.16 | 96.48 | \n| Symbols | 95.68 | **95.88** | 91.06 | 91.56 | \n| ToeSegmentation1 | 96.49 | **98.25** | 81.58 | 96.05 | \n| ToeSegmentation2 | 90.77 | 92.31 | **93.08** | 91.54 | \n| Trace | **100.00** | **100.00** | 98.00 | **100.00** | \n| TwoLeadECG | 99.91 | **100.00** | 97.72 | **100.00** | \n| Two_Patterns | 87.62 | **100.00** | **100.00** | **100.00** | \n| UWaveGestureLibraryAll | 82.86 | 92.27 | **96.26** | 87.16 | \n| Wine | 77.78 | 87.04 | **90.74** | 83.33 | \n| WordsSynonyms | 55.96 | 66.93 | **68.97** | 62.85 | \n| Worms | 76.62 | 81.82 | 62.34 | **83.12** | \n| WormsTwoClass | 74.03 | **77.92** | 63.64 | **77.92** | \n| synthetic_control | 98.67 | **100.00** | **100.00** | **100.00** | \n| uWaveGestureLibrary_X | 76.13 | 82.10 | **83.28** | 79.51 | \n| uWaveGestureLibrary_Y | 64.82 | 73.20 | **75.38** | 68.68 | \n| uWaveGestureLibrary_Z | 73.12 | **78.03** | 77.41 | 76.19 | \n| wafer | 99.61 | 99.84 | 99.81 | **99.90** | \n| yoga | 87.10 | **89.33** | 88.57 | 88.17 | \n| **Wins** | 18 | **38** | 29 | 27 | \n\n\n## Critical difference diagrams\nIf you would like to generate these diagrams, take a look at [this code](https://github.com/hfawaz/cd-diagram)!\n\n![cd-diagram-resnets](https://github.com/hfawaz/ijcnn19ensemble/blob/master/png/cd-diagram-resnets.png)\n![cd-diagram-all](https://github.com/hfawaz/ijcnn19ensemble/blob/master/png/cd-diagram-all.png)\n![cd-diagram-nne](https://github.com/hfawaz/ijcnn19ensemble/blob/master/png/cd-diagram-nne.png)\n \n\n## Reference\n\nIf you re-use this work, please cite:\n\n```\n@InProceedings{IsmailFawaz2019deep,\n Title = {Deep Neural Network Ensembles for Time Series Classification},\n Author = {Ismail Fawaz, Hassan and Forestier, Germain and Weber, Jonathan and Idoumghar, Lhassane and Muller, Pierre-Alain},\n booktitle = {IEEE International Joint Conference on Neural Networks},\n Year = {2019}\n}\n```\n\n## Acknowledgement\n\nWe would like to thank NVIDIA Corporation for the Quadro P6000 grant and the Mésocentre of Strasbourg for providing access to the cluster.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhfawaz%2Fijcnn19ensemble","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fhfawaz%2Fijcnn19ensemble","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhfawaz%2Fijcnn19ensemble/lists"}