|
88 | 88 | "#### [`arcgis.learn`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html)\n",
|
89 | 89 | "* New Table of Contents heading:\n",
|
90 | 90 | " * Replace [`Tabular and Feature Models`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#tabular-and-feature-data-models) with [`Feature, Tabular and Timeseries Models`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#feature-tabular-and-timeseries-models)\n",
|
91 |
| - "* New models:\n", |
| 91 | + "* New Table of Contents entry and models:\n", |
92 | 92 | " * [`Object Tracking Models`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#object-tracking-models)\n",
|
93 | 93 | " * [`SiamMask`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#siammask)\n",
|
94 | 94 | " * [`ObjectTracker`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#objecttracker)\n",
|
95 | 95 | " * [`Track`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#tracker)\n",
|
96 |
| - "* Adds cell size ranges and output units to trainied models (`Esri Model Definitions`)\n", |
| 96 | + "* Adds cell size ranges and output units to trained models (`Esri Model Definitions`)\n", |
97 | 97 | "* Adds `monitor` parameter to `fit()` methods of models for use with `early_stopping` and `checkpoint`\n",
|
| 98 | + "* Adds documentation to `path` parameter for [`prepare_data()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#prepare-data)\n", |
| 99 | + "* Adds keyword parameters and documentation to [`prepare_data()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#prepare-data):\n", |
| 100 | + " * `min_points`\n", |
| 101 | + " * `classes_of_interest`\n", |
| 102 | + " * `extra_features`\n", |
| 103 | + " * `remap_classes`\n", |
| 104 | + " * `background_classcode`\n", |
98 | 105 | "* [`Data Preparation Methods`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#data-preparation-methods)\n",
|
99 | 106 | " * Adds ability to use a folder or list of paths (multi-folder support) in `path` parameter for [`prepare_data()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#prepare-data)\n",
|
100 | 107 | " * Adds `working_dir` parameter to data prepartion methods which sets a default path as a prefix for saving trained models and checkpoints\n",
|
|
119 | 126 | " * Adds cabability to [`save()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#arcgis.learn.MLModel.save) method to allow saving model without data input\n",
|
120 | 127 | " * [`TimeSeriesModel`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#timeseriesmodel)\n",
|
121 | 128 | " * Enhances vizualization to show separate graph for each dependent variable when using `show_batch()` on `TabularData` object\n",
|
| 129 | + " * Adds warning message when using [`predict()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#arcgis.learn.TimeSeriesModel.predict) if [`openpyxl`](https://openpyxl.readthedocs.io/en/stable/) dependency not installed\n", |
122 | 130 | "* [`Pixel Classification Models`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#pixel-classification-models)\n",
|
123 | 131 | " * Adds [`ConnectNet`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#connectnet) Model\n",
|
124 |
| - " * Adds inference and exports support to:\n", |
| 132 | + " * Adds `dice_loss_average` keyword parameter to initialization options\n", |
| 133 | + " * Adds inference and exports support to:\n", |
125 | 134 | " * [`ChangeDetector`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#changedetector)\n",
|
126 |
| - " * Adds threshholding functionality to:\n", |
| 135 | + " * Adds threshholding functionality to:\n", |
127 | 136 | " * [`BDCNEdgeDetector`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html?highlight=retinanet#bdcnedgedetector)\n",
|
128 | 137 | " * [`HEDEdgeDectector`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html?highlight=retinanet#hededgedetector)\n",
|
129 | 138 | " * [`MultiTaskRoadExtractor`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html?highlight=retinanet#multitaskroadextractor)\n",
|
|
133 | 142 | " * Adds inference and export support to:\n",
|
134 | 143 | " * [`Pix2Pix`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#pix2pix)\n",
|
135 | 144 | " * [`CycleGAN`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#cyclegan)\n",
|
| 145 | + " * Adds additional checks for whether multispectral images were used in initialization for:\n", |
| 146 | + " * [`Pix2Pix`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#pix2pix)\n", |
| 147 | + " * [`CycleGAN`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#cyclegan)\n", |
136 | 148 | "* [`Model Management`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#model-management)\n",
|
137 | 149 | " * Adds [`train_model()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#train-model) method"
|
138 | 150 | ]
|
|
207 | 219 | "metadata": {},
|
208 | 220 | "source": [
|
209 | 221 | "#### [`arcgis.learn`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html)\n",
|
| 222 | + "* Fixes to various functions and model methods for `ArcGIS Pro Notebooks`\n", |
| 223 | + " * `show_batch()` \n", |
| 224 | + " * `lr_find()` \n", |
| 225 | + " * `model.show_results()`\n", |
| 226 | + "* Fixes `WARNING 003054: Input Raster does not support PIXEL_SPACE (raw image space). Running in MAP_SPACE` warning message when running the [Classify Pixels Using Deep Learning](https://pro.arcgis.com/en/pro-app/latest/tool-reference/image-analyst/classify-pixels-using-deep-learning.htm) tool in `ArcGIS Pro`\n", |
210 | 227 | "* [`Data Preparation Methods`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#data-preparation-methods)\n",
|
211 | 228 | " * Fixes `expected scalar type Float` Exception with [`prepare_data()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#prepare-data) when using `pytorch 1.7.0`\n",
|
212 | 229 | "* [`Pixel Classification Models`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#pixel-classification-models)\n",
|
| 230 | + " * Improves `per_class_metrics()` results on models initialized from unbalanced datasets\n", |
213 | 231 | " * Fixes `fit()` method in segmentation models returning `NaN` values for certain attributes: [`UNetClassifier`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#unetclassifier), [`PSPNetClassifier`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#pspnetclassifier), [`DeepLab`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#deeplab)\n",
|
214 | 232 | " * [`MultiTaskRoadExtractor`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html?highlight=retinanet#multitaskroadextractor) \n",
|
215 | 233 | " * Fixes issue for [`show_results()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html?highlight=retinanet#arcgis.learn.MultiTaskRoadExtractor.show_results) plotting images incorrectly\n",
|
216 | 234 | " * Fixes issue causing [`save()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html?highlight=retinanet#arcgis.learn.MultiTaskRoadExtractor.save) only to work after calling [`fit()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html?highlight=retinanet#arcgis.learn.MultiTaskRoadExtractor.fit)\n",
|
217 | 235 | " * Fixes [`load()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html?highlight=retinanet#arcgis.learn.MultiTaskRoadExtractor.load) so it only needs the model name\n",
|
218 | 236 | " * Fixes errors with [`fit()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html?highlight=retinanet#arcgis.learn.MultiTaskRoadExtractor.fit) when using monitor functionality\n",
|
| 237 | + " * Fixes parameter table for [`show_results()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#arcgis.learn.MultiTaskRoadExtractor.show_results) documentation\n", |
219 | 238 | "* [`Unstructured Text Models`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#unstructured-text-models)\n",
|
220 | 239 | " * [`arcgis.learn.text`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#arcgis-learn-text-module) module\n",
|
221 | 240 | " * Adds capability to use `HuggingFace` [pretrained models](https://huggingface.co/transformers/pretrained_models.html) for:\n",
|
|
237 | 256 | " * Fixes issue with [`lr_find()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html?highlight=tabulardata#arcgis.learn.FeatureClassifier.lr_find):\n",
|
238 | 257 | " * with object created with [`mixup=True`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html?highlight=tabulardata#arcgis.learn.FeatureClassifier)\n",
|
239 | 258 | " * with `MultiLabeled_Tiles` dataset type object and [`mixup=True`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html?highlight=tabulardata#arcgis.learn.FeatureClassifier)\n",
|
240 |
| - " * Fixes issue with using [`predict()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html?highlight=tabulardata#arcgis.learn.FeatureClassifier.predict) on a trained model\n", |
241 |
| - " \n", |
| 259 | + " * Fixes issue with using [`predict()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html?highlight=tabulardata#arcgis.learn.FeatureClassifier.predict) on a trained model \n", |
242 | 260 | "* [`Feature, Tabular and Timeseries Models`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#feature-tabular-and-timeseries-models)\n",
|
243 | 261 | " * [`MLModel`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#mlmodel)\n",
|
244 | 262 | " * Fixes inaccurate [`predict()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#arcgis.learn.MLModel.predict) results because training transformations were not incorporating training statistics\n",
|
|
248 | 266 | " * [`PointCNN`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#pointcnn)\n",
|
249 | 267 | " * Fixes bug with [`predict_las()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#arcgis.learn.PointCNN.predict_las) returning class codes from the model instead of the actual class codes in specific scenarios\n",
|
250 | 268 | " * Fixes error message raised with [`from_model()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#arcgis.learn.PointCNN.from_model) if library dependencies are missing\n",
|
| 269 | + " * Fixes [`show_results()`](file:///Users/john3092/Repos/geosaurus/docs/api_ref/build/html/arcgis.learn.toc.html?highlight=dice%20coefficient#arcgis.learn.PointCNN.show_results) and `show_batch()` when run against data object created with a `remap_classes` argument in [`prepare_data()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#prepare-data)\n", |
251 | 270 | "* [`Object Detection Models`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#object-detection-models)\n",
|
252 | 271 | " * [`SingleShotDetector`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#singleshotdetector)\n",
|
253 | 272 | " * Fixes out of memory error when using [`show_results()`](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#arcgis.learn.SingleShotDetector.show_results) on model with multispectral data\n",
|
|
0 commit comments