[DOC] update intro notebook with material from ISF and EuroSciPy 2024 (…

…#7013) This PR updates the intro notebook with material from ISF 2024 and EuroSciPy 2024, and more generally brings it to a more up-to-date and consistent state. * reformat most existing images to scale with width of text * Updated "object and estimator types" table with links, add reference to `skpro` for "time" relevant objects * made basic example vignettes consistent with exposition slides in section 2 * added feature summary and marketplace slides from @fkiraly's talk at ISF 2024 to section 3 * added reference and link to estimator search in section 3 A few png are added for the slides in section 3.
sktime · Aug 27, 2024 · d0c294b · d0c294b
1 parent 7eabd75
commit d0c294b
Show file tree

Hide file tree

Showing 6 changed files with 94 additions and 41 deletions.
diff --git a/examples/00_sktime_intro.ipynb b/examples/00_sktime_intro.ipynb
@@ -98,7 +98,7 @@
     "2. **Fit** the instance of your model\n",
     "3. Use that fitted instance to **predict** new data!\n",
     "\n",
-    "![](./img/estimator-conceptual-model.jpg)"
+    "<img src=\"./img/estimator-conceptual-model.jpg\" alt=\"Estimator conceptual model\" style=\"width: 100%; max-width: 600px;\">"
    ]
   },
   {
@@ -423,7 +423,8 @@
     "= like a power plug adapter, it's plug&play if it conforms with the interface\n",
     "\n",
     "Pictorial summary:\n",
-    "![](./img/sklearn-unified-interface.jpg)"
+    "\n",
+    "<img src=\"./img/sklearn-unified-interface.jpg\" alt=\"Unified estimator interface\" style=\"width: 100%; max-width: 600px;\">"
    ]
   },
   {
@@ -498,54 +499,76 @@
    "metadata": {},
    "source": [
     "| Task | Status | Links |\n",
-    "|------|--------|-------|\n",
-    "| **Forecasting**     |  stable      |>>[Tutorial](https://www.sktime.net/en/latest/examples/01_forecasting.html) >> [API Reference](https://www.sktime.net/en/latest/api_reference/forecasting.html) >> [Extension Template](https://github.com/sktime/sktime/blob/main/extension_templates/forecasting.py)|\n",
-    "| **Time Series Classification** | stable |>>[Tutorial](https://github.com/sktime/sktime/blob/main/examples/02_classification.ipynb) >> [API Reference](https://www.sktime.net/en/latest/api_reference/classification.html) >> [Extension Template](https://github.com/sktime/sktime/blob/main/extension_templates/classification.py)|\n",
-    "| **Transformations** | stable |>>[Tutorial](https://github.com/sktime/sktime/blob/main/examples/03_transformers.ipynb) >> [API Reference](https://www.sktime.net/en/latest/api_reference/transformations.html) >> [Extension Template](https://github.com/sktime/sktime/blob/main/extension_templates/transformer.py)|\n",
-    "| **Parameter fitting** | maturing |>>[API Reference](https://www.sktime.net/en/latest/api_reference/param_est.html) >> [Extension Template](https://github.com/sktime/sktime/blob/main/extension_templates/transformer.py)|\n",
-    "| **Time Series Clustering** | maturing |>>[API Reference](https://www.sktime.net/en/latest/api_reference/clustering.html) >> [Extension Template](https://github.com/sktime/sktime/blob/main/extension_templates/clustering.py)|\n",
-    "| **Time Series Distances/Kernels** | maturing |>>[Tutorial](https://github.com/sktime/sktime/blob/main/examples/03_transformers.ipynb) >> [API Reference](https://www.sktime.net/en/latest/api_reference/dists_kernels.html) >> [Extension Template](https://github.com/sktime/sktime/blob/main/extension_templates/dist_kern_panel.py)|\n",
-    "| **Annotation** | experimental |>>[Extension Template](https://github.com/sktime/sktime/blob/main/extension_templates/annotation.py)|\n",
-    "|**Distributions and simulation** | experimental |  None  |"
+    "|---|---|---|\n",
+    "| **Forecasting** | stable | [Tutorial](https://www.sktime.net/en/latest/examples/01_forecasting.html) · [API Reference](https://www.sktime.net/en/latest/api_reference/forecasting.html) · [Extension Template](https://github.com/sktime/sktime/blob/main/extension_templates/forecasting.py)  |\n",
+    "| **Time Series Classification** | stable | [Tutorial](https://github.com/sktime/sktime/blob/main/examples/02_classification.ipynb) · [API Reference](https://www.sktime.net/en/latest/api_reference/classification.html) · [Extension Template](https://github.com/sktime/sktime/blob/main/extension_templates/classification.py) |\n",
+    "| **Time Series Regression** | stable | [Tutorial](https://github.com/sktime/sktime/blob/main/examples/02_classification.ipynb) · [API Reference](https://www.sktime.net/en/latest/api_reference/regression.html) |\n",
+    "| **Transformations** | stable | [Tutorial](https://github.com/sktime/sktime/blob/main/examples/03_transformers.ipynb) · [API Reference](https://www.sktime.net/en/latest/api_reference/transformations.html) · [Extension Template](https://github.com/sktime/sktime/blob/main/extension_templates/transformer.py)  |\n",
+    "| **Performance metrics for forecasts** | stable | [API Reference](https://www.sktime.net/en/latest/api_reference/performance_metrics.html) |\n",
+    "| **Time series splitting/resampling** | stable | [API Reference](https://www.sktime.net/en/latest/api_reference/split.html) |\n",
+    "| **Parameter fitting** | maturing | [API Reference](https://www.sktime.net/en/latest/api_reference/param_est.html) · [Extension Template](https://github.com/sktime/sktime/blob/main/extension_templates/transformer.py)  |\n",
+    "| **Time Series Alignment** | maturing | [API Reference](https://www.sktime.net/en/latest/api_reference/alignment.html) ·  [Extension Template](https://github.com/sktime/sktime/blob/main/extension_templates/alignment.py) |\n",
+    "| **Time Series Clustering** | maturing | [API Reference](https://www.sktime.net/en/latest/api_reference/clustering.html) ·  [Extension Template](https://github.com/sktime/sktime/blob/main/extension_templates/clustering.py) |\n",
+    "| **Time Series Distances/Kernels** | maturing | [Tutorial](https://github.com/sktime/sktime/blob/main/examples/03_transformers.ipynb) · [API Reference](https://www.sktime.net/en/latest/api_reference/dists_kernels.html) · [Extension Template](https://github.com/sktime/sktime/blob/main/extension_templates/dist_kern_panel.py) |\n",
+    "| **Anomalies, changepoints** | experimental | [Extension Template](https://github.com/sktime/sktime/blob/main/extension_templates/annotation.py) |\n",
+    "\n",
+    "In the [skpro](https://github.com/sktime/skpro) companion package:\n",
+    "\n",
+    "| Module | Status | Links |\n",
+    "|---|---|---|\n",
+    "| **Probabilistic tabular regression** | maturing | [Tutorial](https://github.com/sktime/skpro/blob/main/examples/01_skpro_intro.ipynb) · [API Reference](https://skpro.readthedocs.io/en/latest/api_reference/regression.html) · [Extension Template](https://github.com/sktime/skpro/blob/main/extension_templates/regression.py) |\n",
+    "| **Time-to-event (survival) prediction** | maturing |  [Tutorial](https://github.com/sktime/skpro/blob/main/examples/02_skpro_survival.ipynb) · [API Reference](https://skpro.readthedocs.io/en/latest/api_reference/survival.html) · [Extension Template](https://github.com/sktime/skpro/blob/main/extension_templates/survival.py) |\n",
+    "| **Performance metrics for proba predictions** | maturing | [API Reference](https://skpro.readthedocs.io/en/latest/api_reference/metrics.html) |\n",
+    "| **Probability distributions** | maturing | [Tutorial](https://github.com/sktime/skpro/blob/main/examples/03_skpro_distributions.ipynb) · [API Reference](https://skpro.readthedocs.io/en/latest/api_reference/distributions.html) · [Extension Template](https://github.com/sktime/skpro/blob/main/extension_templates/distributions.py) |\n"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "**Some examples below - details in the respective module introduction notebooks!**"
+    "### Example - forecasting"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Example - forecasting"
+    "<img src=\"./img/forecasting.png\" alt=\"Forecasting\" style=\"width: 100%; max-width: 600px;\">"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
+    "# get the data\n",
     "from sktime.datasets import load_airline\n",
-    "from sktime.forecasting.naive import NaiveForecaster\n",
     "\n",
-    "# step 1: data specification\n",
-    "y = load_airline()\n",
+    "y = load_airline()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
     "\n",
-    "# step 2: specifying forecasting horizon\n",
-    "fh = range(1, 37)\n",
+    "from sktime.forecasting.naive import NaiveForecaster\n",
     "\n",
-    "# step 3: specifying the forecasting algorithm\n",
+    "# step 1: specify the forecasting algorithm\n",
     "forecaster = NaiveForecaster(strategy=\"last\", sp=12)\n",
     "\n",
-    "# step 4: fitting the forecaster\n",
-    "forecaster.fit(y)\n",
+    "# step 2: specify forecasting horizon\n",
+    "fh = np.arange(1, 37)  # we want to predict the next 36 months\n",
     "\n",
-    "# step 5: querying predictions\n",
-    "y_pred = forecaster.predict(fh)"
+    "# step 3: fit the forecaster\n",
+    "forecaster.fit(y, fh=fh)\n",
+    "\n",
+    "# step 4: make the forecast\n",
+    "y_pred = forecaster.predict()"
    ]
   },
   {
@@ -574,7 +597,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Example - classification"
+    "### Example - time series classification"
    ]
   },
   {
@@ -586,30 +609,39 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
-    "from sktime.classification.distance_based import KNeighborsTimeSeriesClassifier\n",
+    "# get the data\n",
     "from sktime.datasets import load_osuleaf\n",
-    "from sktime.dists_kernels import ScipyDist\n",
-    "from sktime.dists_kernels.compose_tab_to_panel import FlatDist\n",
     "\n",
-    "# step 1 - specify training data\n",
+    "# for training\n",
     "X_train, y_train = load_osuleaf(split=\"train\", return_type=\"numpy3D\")\n",
     "\n",
-    "# step 2 - specify data to predict labels for\n",
+    "# for prediction\n",
     "X_new, _ = load_osuleaf(split=\"test\", return_type=\"numpy3D\")\n",
-    "X_new = X_new[:2]\n",
+    "X_new = X_new[:2]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from sktime.classification.distance_based import KNeighborsTimeSeriesClassifier\n",
+    "from sktime.dists_kernels import ScipyDist\n",
+    "from sktime.dists_kernels.compose_tab_to_panel import AggrDist\n",
     "\n",
-    "# step 3 - specify the classifier\n",
-    "mean_eucl_dist = FlatDist(ScipyDist())\n",
+    "# step 1 - specify the classifier\n",
+    "mean_eucl_dist = AggrDist(ScipyDist())\n",
     "clf = KNeighborsTimeSeriesClassifier(n_neighbors=3, distance=mean_eucl_dist)\n",
     "\n",
-    "# step 4 - fitting the classifier\n",
+    "# step 2 - fit the classifier\n",
     "clf.fit(X_train, y_train)\n",
     "\n",
-    "# step 5 - predict labels on new data\n",
+    "# step 3 - predict labels on new data\n",
     "y_pred = clf.predict(X_new)"
    ]
   },
@@ -703,15 +735,36 @@
     "\n",
     "- lots of great implementations and methods out there!\n",
     "- but many different interfaces, not composable like `sklearn`\n",
+    "\n",
+    "<img src=\"./img/ts-fragmentation.png\" alt=\"The fragmented time series ecosystem\" style=\"width: 100%; max-width: 600px;\">"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
     "\n",
     "`sktime` integrates the ecosystem - in friendly collaboration with all the packages out there!\n",
     "\n",
-    "- unified interface standard\n",
-    "- highly composable\n",
-    "- mini-package manager on estiator/module level\n",
-    "- easily extensible - 3rd party plugins, other packages\n",
+    "<img src=\"./img/sktime-interoperable.png\" alt=\"sktime integration\" style=\"width: 100%; max-width: 600px;\">"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<img src=\"./img/sktime-composable.png\" alt=\"sktime composition\" style=\"width: 100%; max-width: 600px;\">"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "easy search for plug&play components across the ecosystem!\n",
+    "\n",
+    "Try the [`sktime` estimator search](https://www.sktime.net/en/latest/estimator_overview.html)\n",
     "\n",
-    "![](./img/unified_framework.png)"
+    "<img src=\"./img/estimator-search.png\" alt=\"sktime estimator search\" style=\"width: 100%; max-width: 600px;\">"
    ]
   },
   {

diff --git a/examples/img/estimator-search.png b/examples/img/estimator-search.png
diff --git a/examples/img/sktime-composable.png b/examples/img/sktime-composable.png
diff --git a/examples/img/sktime-doc-estimator-overview.png b/examples/img/sktime-doc-estimator-overview.png
diff --git a/examples/img/sktime-interoperable.png b/examples/img/sktime-interoperable.png
diff --git a/examples/img/ts-fragmentation.png b/examples/img/ts-fragmentation.png