|
| 1 | +(pycaret)= |
| 2 | +# PyCaret |
| 3 | + |
| 4 | +```{div} |
| 5 | +:style: "float: right; margin-left: 1em" |
| 6 | +[{w=180px}](https://pycaret.org/) |
| 7 | +``` |
| 8 | +```{div} |
| 9 | +:style: "clear: both" |
| 10 | +``` |
| 11 | + |
| 12 | +:::{rubric} About |
| 13 | +::: |
| 14 | + |
| 15 | +[PyCaret] is an open-source, low-code machine learning library for Python that |
| 16 | +automates machine learning workflows. |
| 17 | + |
| 18 | +It is a high-level interface and AutoML wrapper on top of your loved machine learning |
| 19 | +libraries like scikit-learn, xgboost, ray, lightgbm, and many more. PyCaret provides a |
| 20 | +universal interface to utilize these libraries without needing to know the details |
| 21 | +of the underlying model architectures and parameters. |
| 22 | + |
| 23 | +:::{rubric} Concept |
| 24 | +::: |
| 25 | + |
| 26 | +The general concept of PyCaret - and for the matter of fact for AutoML in general - |
| 27 | +is rather simple: One takes the raw data, splits it into a training and a test set |
| 28 | +and then trains a number of different models on the training set. The models are |
| 29 | +then evaluated on the test set and the best performing model is selected. |
| 30 | + |
| 31 | +:::{rubric} Hyperparameter tuning |
| 32 | +::: |
| 33 | + |
| 34 | +This process gets repeated for tuning the hyperparameters of the best models. |
| 35 | +Again, this process is highly empirical. The parameters are changed, the model is |
| 36 | +retrained and evaluated again. This process is repeated until the best performing |
| 37 | +parameters are found. |
| 38 | + |
| 39 | +Modern algorithms for executing all these experiments are - amongst others - |
| 40 | +GridSearch, RandomSearch and BayesianSearch. For a quick introduction into |
| 41 | +these methods, see [Introduction to hyperparameter tuning]. |
| 42 | + |
| 43 | +:::{rubric} Benefits |
| 44 | +::: |
| 45 | + |
| 46 | +In the past, all these trial-and-error experiments had to be done manually, |
| 47 | +which is a tedious and time-consuming task. PyCaret automates this process |
| 48 | +and provides a simple interface to execute all these experiments in a |
| 49 | +straight-forward way. The notebooks referenced below demonstrate how this works. |
| 50 | + |
| 51 | +:::{rubric} Learn |
| 52 | +::: |
| 53 | + |
| 54 | +Tutorials and Notebooks about using [PyCaret] together with CrateDB. |
| 55 | + |
| 56 | +::::{info-card} |
| 57 | +:::{grid-item} |
| 58 | +:columns: 9 |
| 59 | +**Notebook: AutoML classification with PyCaret** |
| 60 | + |
| 61 | +Explore the PyCaret framework and show how to use it to train different |
| 62 | +classification models. |
| 63 | + |
| 64 | +[][AutoML with PyCaret and CrateDB] |
| 65 | +[][automl-classify-github] |
| 66 | +[][automl-classify-colab] |
| 67 | +::: |
| 68 | +:::{grid-item} |
| 69 | +:columns: 3 |
| 70 | +{tags-primary}`Fundamentals` \ |
| 71 | +{tags-secondary}`Time Series` \ |
| 72 | +{tags-secondary}`Anomaly Detection` \ |
| 73 | +{tags-secondary}`Prediction / Forecasting` |
| 74 | +::: |
| 75 | +:::: |
| 76 | + |
| 77 | +::::{info-card} |
| 78 | +:::{grid-item} |
| 79 | +:columns: 9 |
| 80 | +**Notebook: Train time series forecasting models** |
| 81 | + |
| 82 | +How to train time series forecasting models using PyCaret and CrateDB. |
| 83 | + |
| 84 | +[][AutoML with PyCaret and CrateDB] |
| 85 | +[][automl-forecasting-github] |
| 86 | +[][automl-forecasting-colab] |
| 87 | +::: |
| 88 | +:::{grid-item} |
| 89 | +:columns: 3 |
| 90 | +{tags-primary}`Fundamentals` \ |
| 91 | +{tags-secondary}`Time Series` \ |
| 92 | +{tags-secondary}`Training` \ |
| 93 | +{tags-secondary}`Classification` \ |
| 94 | +{tags-secondary}`Forecasting` |
| 95 | +::: |
| 96 | +:::: |
| 97 | + |
| 98 | + |
| 99 | +[AutoML with PyCaret and CrateDB]: https://github.com/crate/cratedb-examples/tree/main/topic/machine-learning/pycaret |
| 100 | +[automl-classify-github]: https://github.com/crate/cratedb-examples/blob/main/topic/machine-learning/pycaret/automl_classification_with_pycaret.py |
| 101 | +[automl-classify-colab]: https://colab.research.google.com/github/crate/cratedb-examples/blob/main/topic/machine-learning/pycaret/automl_classification_with_pycaret.ipynb |
| 102 | +[automl-forecasting-github]: https://github.com/crate/cratedb-examples/blob/main/topic/machine-learning/pycaret/automl_timeseries_forecasting_with_pycaret.ipynb |
| 103 | +[automl-forecasting-colab]: https://colab.research.google.com/github/crate/cratedb-examples/blob/main/topic/machine-learning/pycaret/automl_timeseries_forecasting_with_pycaret.ipynb |
| 104 | +[Introduction to hyperparameter tuning]: https://medium.com/analytics-vidhya/comparison-of-hyperparameter-tuning-algorithms-grid-search-random-search-bayesian-optimization-5326aaef1bd1 |
| 105 | +[PyCaret]: https://www.pycaret.org |
0 commit comments