diff --git a/aeon/regression/deep_learning/_inception_time.py b/aeon/regression/deep_learning/_inception_time.py index 8e40beb649..5e6c6a56e9 100644 --- a/aeon/regression/deep_learning/_inception_time.py +++ b/aeon/regression/deep_learning/_inception_time.py @@ -136,8 +136,7 @@ class InceptionTimeRegressor(BaseRegressor): Notes ----- - Adapted from the implementation from Fawaz et. al - https://github.com/hfawaz/InceptionTime/blob/master/regressors/inception.py + Adapted from the implementation from Fawaz et. al ..[1] and Ismail-Fawaz et al. https://github.com/MSD-IRIMAS/CF-4-TSC diff --git a/docs/api_reference/data_format.rst b/docs/api_reference/data_format.rst index 28ed911f56..6cc20b2989 100644 --- a/docs/api_reference/data_format.rst +++ b/docs/api_reference/data_format.rst @@ -203,7 +203,7 @@ This section provides full set of instructions to create a format specification for your dataset that is compatible with ``aeon``. Remember that this begins with the assumption that you have the dataset readily available in -expected `format `_. +expected `format `_. Few points to keep in mind while creating the dataset: diff --git a/docs/examples.md b/docs/examples.md index 3dddf67909..7b4b269b2f 100644 --- a/docs/examples.md +++ b/docs/examples.md @@ -186,7 +186,7 @@ Overview of Transformations :::{grid-item-card} :img-top: examples/transformations/img/tsfresh.png :class-img-top: aeon-card-image-m -:link: /examples/transformations/feature_extraction_with_tsfresh.ipynb +:link: /examples/transformations/tsfresh.ipynb :link-type: ref :text-align: center diff --git a/examples/datasets/data_loading.ipynb b/examples/datasets/data_loading.ipynb index fe97bdad28..c3b1bb49a7 100644 --- a/examples/datasets/data_loading.ipynb +++ b/examples/datasets/data_loading.ipynb @@ -8,7 +8,7 @@ "[Provided datasets](provided_data.ipynb). Downloading data is described in\n", "[Downloading and loading benchmarking datasets](load_data_from_web.ipynb). You\n", "can of course load and format the data so that it conforms to the input types described\n", - "in [Data structures and containers for aeon estimators](data_structures.ipynb). `aeon`\n", + "in [Data structures and containers for aeon estimators](datasets.ipynb). `aeon`\n", "also provides data formats for time series for both forecasting and machine learning.\n", "These are all text files with a particular structure. Both formats store a single time\n", "series per row.\n", @@ -33,7 +33,7 @@ " ).\n", "\n", "The baked in datasets are described [here](provided_data.ipynb). Data\n", - "structures to store the data are described [here](data_structures.ipynb)." + "structures to store the data are described [here](datasets.ipynb)." ], "metadata": { "collapsed": false @@ -276,7 +276,7 @@ "source": [ "Train and test partitions of the ArrowHead problem have been loaded into 3D numpy\n", "arrays with an associated array of class values. Further info on data structures is\n", - "given in [this notebook](data_structures.ipynb). Datasets that are shipped with aeon\n", + "given in [this notebook](datasets.ipynb). Datasets that are shipped with aeon\n", "(like ArrowHead, BasicMotions and PLAID) can be more simply loaded with bespoke\n", "functions. More details [here](provided_data.ipynb)" ] @@ -435,8 +435,7 @@ "\n", "A further option is to load data into aeon from tab separated value (`.tsv`) files.\n", "Researchers at the University of Riverside, California make a variety of timeseries\n", - "data available in this format at [Eamonn Keogh's website](https://www.cs.ucr\n", - ".edu/~eamonn/time_series_data_2018). Each row is a time series, and the class value\n", + "data available in this format at [Eamonn Keogh's website](https://www.cs.ucr.edu/~eamonn/time_series_data_2018). Each row is a time series, and the class value\n", "is the first one.\n", "\n", "The `load_from_tsv_file` method in `aeon.datasets` supports reading\n", diff --git a/examples/datasets/load_data_from_web.ipynb b/examples/datasets/load_data_from_web.ipynb index 71ea7561e8..7dd4c4bca4 100644 --- a/examples/datasets/load_data_from_web.ipynb +++ b/examples/datasets/load_data_from_web.ipynb @@ -19,7 +19,7 @@ "numpy if `n_timepoints` is different for different cases. Forecasting data are loaded\n", "into pd.DataFrame. Anomaly detection dataset are loaded into 2D numpy arrays of shape\n", "`(n_timepoints, n_channels)`. For more information on aeon data types see the\n", - "[data structures notebook](data_structures.ipynb).\n", + "[data structures notebook](datsets.ipynb).\n", "\n", "Note that this notebook is dependent on external websites, so will not function if\n", "you are not online or the associated website is down. We use the following four\n", diff --git a/examples/distances/sklearn_distances.ipynb b/examples/distances/sklearn_distances.ipynb index f28ad3d0be..e22579828c 100644 --- a/examples/distances/sklearn_distances.ipynb +++ b/examples/distances/sklearn_distances.ipynb @@ -63,8 +63,8 @@ "but it is treating the data as vector rather than as time series.\n", "\n", "If we try and use with an `aeon` style 3D numpy\n", - "`(n_cases, n_channels, n_timepoints)`, they will crash as `scikit-learn` expect a 2D \n", - "numpy array. See the [data_formats](../datasets/data_structures.ipynb) for details on \n", + "`(n_cases, n_channels, n_timepoints)`, they will crash as `scikit-learn` expect a 2D\n", + "numpy array. See the [data_formats](../datasets/datasets.ipynb) for details on\n", "data storage." ] }, @@ -121,8 +121,8 @@ "collapsed": false }, "source": [ - "We can use `KNeighborsClassifier` with a callable `aeon` distance function, but the \n", - "input must still be 2D numpy array. " + "We can use `KNeighborsClassifier` with a callable `aeon` distance function, but the\n", + "input must still be 2D numpy array." ] }, { @@ -240,19 +240,19 @@ "collapsed": false }, "source": [ - "Also note that using an `aeon` distance function as callable does not will not work with \n", - "some kNN options such as [`KDTree`](https://scikit-learn.org/stable/modules/generated/sklearn.neighbors.KDTree.html) \n", + "Also note that using an `aeon` distance function as callable does not will not work with\n", + "some kNN options such as [`KDTree`](https://scikit-learn.org/stable/modules/generated/sklearn.neighbors.KDTree.html)\n", "class or [`BallTree`](https://scikit-learn.org/stable/modules/generated/sklearn.neighbors.BallTree.html),\n", "as stated in the scikit-learn doc of these classes:\n", "\n", "_Note: Callable functions in the metric parameter are NOT supported for KDTree_\n", "_and Ball Tree. Function call overhead will result in very poor performance._\n", "\n", - "Because of these problems, we have implemented a KNN classifier and regressor to use \n", + "Because of these problems, we have implemented a KNN classifier and regressor to use\n", "with our distance functions.\n", "\n", - "The `aeon` kNN classifier using a 3D numpy array achieves the same performance than the \n", - "`sklearn` one using the 2D numpy array even using time series specific distance \n", + "The `aeon` kNN classifier using a 3D numpy array achieves the same performance than the\n", + "`sklearn` one using the 2D numpy array even using time series specific distance\n", "functions. The results achieved are the same as time series are univariate and hence,\n", "the data can be formatted as a 2D array:" ] @@ -307,7 +307,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "However, if the time series dataset is a multivariate one, data has to be represented \n", + "However, if the time series dataset is a multivariate one, data has to be represented\n", "using a 3D numpy array. In this case, to use the `sklearn` knn approach, channels have\n", "to be concatenated, and therefore, specific edit time series distances may compute the\n", "distance between values of different channels, and the results may be biased by these\n", @@ -398,7 +398,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Similar conclusions can be drawn for the kNN regressor. First of all, we load the \n", + "Similar conclusions can be drawn for the kNN regressor. First of all, we load the\n", "TSER dataset." ] }, @@ -426,7 +426,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Now, we compare the prediction of the `aeon` and `scikit-learn` versions. As the \n", + "Now, we compare the prediction of the `aeon` and `scikit-learn` versions. As the\n", "Covid3Month dataset is univariate, the results of both libraries should be the same." ] }, @@ -547,7 +547,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Same conclusions can be obtained when dealing with a TSER dataset. " + "Same conclusions can be obtained when dealing with a TSER dataset." ] }, { @@ -623,8 +623,8 @@ "collapsed": false }, "source": [ - "The SVM estimators in `scikit-learn` can be used with pairwise distance matrices. Please \n", - "note that not all elastic distance functions are kernels, and it is desirable that they \n", + "The SVM estimators in `scikit-learn` can be used with pairwise distance matrices. Please\n", + "note that not all elastic distance functions are kernels, and it is desirable that they\n", "are for SVM. DTW is not a metric, but MSM and TWE are." ] }, @@ -715,7 +715,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "SVR and NuSVR also allow to use the distance function as callable as previously \n", + "SVR and NuSVR also allow to use the distance function as callable as previously\n", "aforementioned. As can be observed, the results are the same:" ] }, @@ -860,7 +860,7 @@ "collapsed": false }, "source": [ - "You can use pairwise distance functions within the `scikit-learn` FunctionTransformer \n", + "You can use pairwise distance functions within the `scikit-learn` FunctionTransformer\n", "wrapper" ] }, diff --git a/examples/pydata/Amsterdam-2023/Lets do the time warp again.ipynb b/examples/pydata/Amsterdam-2023/Lets_do_the_time_warp_again.ipynb similarity index 99% rename from examples/pydata/Amsterdam-2023/Lets do the time warp again.ipynb rename to examples/pydata/Amsterdam-2023/Lets_do_the_time_warp_again.ipynb index 8abe8dce6a..0d2007db3b 100644 --- a/examples/pydata/Amsterdam-2023/Lets do the time warp again.ipynb +++ b/examples/pydata/Amsterdam-2023/Lets_do_the_time_warp_again.ipynb @@ -26,7 +26,7 @@ }, { "cell_type": "code", - "execution_count": 2, + "execution_count": 1, "metadata": { "ExecuteTime": { "end_time": "2023-09-24T21:09:32.118338976Z", @@ -37,11 +37,9 @@ "outputs": [ { "data": { - "text/plain": [ - "108.0" - ] + "text/plain": "108.0" }, - "execution_count": 2, + "execution_count": 1, "metadata": {}, "output_type": "execute_result" } diff --git a/examples/transformations/minirocket.ipynb b/examples/transformations/minirocket.ipynb index bdc32b4e72..91619248a0 100644 --- a/examples/transformations/minirocket.ipynb +++ b/examples/transformations/minirocket.ipynb @@ -68,7 +68,7 @@ "source": [ "### 1.2 Load the Training Data\n", "\n", - "For more details on the data set, see the [univariate time series classification notebook](https://github.com/aeon-toolkit/aeon/blob/main/examples/02_classification_univariate.ipynb).\n", + "For more details on the data set, see the [classification notebook](../classification/classification.ipynb).\n", "\n", "**Note**: Input time series must be *at least* of length 9. Pad shorter time series\n", "using, e.g., `Padder` (`aeon.transformers.collection`)." diff --git a/examples/transformations/resizing.ipynb b/examples/transformations/resizing.ipynb index efd23cc5ee..80f2254c7d 100644 --- a/examples/transformations/resizing.ipynb +++ b/examples/transformations/resizing.ipynb @@ -130,7 +130,7 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 5, "outputs": [ { "name": "stdout", @@ -165,7 +165,7 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 6, "metadata": { "execution": { "iopub.execute_input": "2020-12-19T14:32:01.245270Z", @@ -208,13 +208,13 @@ }, { "cell_type": "code", - "execution_count": 12, + "execution_count": 7, "outputs": [ { "data": { - "text/plain": "0.8212290502793296" + "text/plain": "0.8268156424581006" }, - "execution_count": 12, + "execution_count": 7, "metadata": {}, "output_type": "execute_result" }