What's new in version 1.8.1

New

New Guides

• Labeling text using Doccano
• Point Cloud Segmentation using PointCNN
• Working with Multispectral Data

New Samples

• Shipwrecks detection using bathymetric data
• Visualize monthly changes in Hirakund reservoir using video
• Data Visualization: Construction permits, part 1/2
• Data Summarization: Construction permits, part 2/2
• Analyzing United States tornadoes
• Which college district has the fewest low-income families?
• Data Preparation: Hurricane analysis, part 1/3
• Exploratory Statistics: Hurricane analysis, part 2/3
• Correlation: Hurricane analysis, part 3/3
• Raster Analytics: Calculate wildfire landslide risk

arcgis.gis

• Adds arcgis.gis.tasks submodule to work with scheduling Notebook Items:
• TaskManager
• Task
• Run
• Adds GroupMigrationManager class
• Supports content migration betweeen Enterprise 10.8.1 portals Group to Group, preserving itemId values:
  • hosted feature services, web maps, web mapping applications or any other text-based items
• Adds migration property to Group to access GroupMigrationManager
• Enables UserManager.create() to use New Member Defaults when set if corresponding arguments not provided
• Adds ability to specify an item_id value for:
• ContentManager.add()
• ContentManager.create_service()
• ContentManager.import_data()
• Item.publish()
• Adds support for adding WMTS Services as items - (OGC WMTS Specification)
• Enhancements to field name mapping in clone_items() for Web Applications, Dashboards, and Surveys
• Adds PortalDataStore class

arcgis.gis.admin

• Adds NotebookServer, MissionServer, and WorkflowManager as function value options to Federation.update() method
• Adds AGOLUsageReports documentation to API Reference

arcgis.gis.server

• Adds ItemInformationManager documentation to API Reference
• Adds code sample to connect to stand-alone server (See Usage Example #2)

arcgis.gis.nb

• Adds recent_statistics property to the object returned from the NotebookServer.system property
• Adds support for expiration value configured from the NotebookServer

arcgis.features

• Adds transformations parameter to SyncManager.create() to support 2-Way replication
• Adds attributes to Spatially Enabled Dataframe
• sanitize_column_names() method to ensure columns names are valid for to_featurelayer() and/or to_featureclass() methods
• renderer property to access dataframe renderer
• Adds geometry_column parameter to pd.Dataframe.spatial.from_df() method
• Adds support for as_df parameter on query() operation against layers that do not suppport pagination

arcgis.raster

• Adds RasterCollection class
• Adds ImageryLayerCacheManager documentation to API Reference
• Adds properties to Raster class:
• vmin
• vmax
• Adds slices() method to ImageryLayer class
• Adds slice_id parameter to the ImageryLayer.export_image() method

arcgis.raster.analytics

• Adds functions:
• optimal_path_as_line
• optimal_region_connetions
• analyze_changes_using_ccdc
• detect_change_using_change_analysis_raster
• manage_multidimensional_raster
• sample
• Adds new parameters to generate_trend_raster
• cycle_length
• cycle_unit
• rmse
• r2
• slope_p_value
• Adds functionality to summarize_raster_within
• percentile_value and process_as_multidimensional parameters
• PERCENTILE option for statistic_type parameter
• Adds functionality to generate_multidimensional_anomaly
• New reference_mean_raster parameter
• EXTERNAL_RASTER as option for calculation_interval parameter
• Deprecates functions -
• Sofware raises warning message when older tool is used and directions on which tool to use as replacement:
  • optimum_travel_cost_network
  • determine_travel_costpath_as_polyline
  • costpath_as_polyline

arcgis.raster.functions

• Adds new functions:
• generate_trend
• predict_using_trend
• linear_spectral_unmixing
• multidimensional_filter
• Adds sentinel functions:
• s1_radiometric_calibration
• s1_thermal_noise_removals
• interpolate_irregular_data
• Adds parameters to shaded_relief()
  • colorramp
  • hillshade_type

arcgis.raster.functions.gbl

• Adds parameters to distance_accumulation
• output_back_direction_raster
• output_source_direction_raster_name
• output_source_location_raster_name
• output_distance-accumulation_raster_name
• Adds percentile_value parameter to zonal_statisitics
• Adds PERCENTILE statistic_type to zonal_statisitics
• Removes parameter from distance_accumulation
• generate_back_direction_band
• Enhances distance_accumulation to generate multiple outputs
• Enhances distance_allocation to generate multiple outputs
• Removes parameter from [distance_allocation](distance_accumulation
• generate_source_row_column_bands
• Adds optimal_path_as_raster function
• Deprecates functions:
• Sofware raises warning message when older tool is used and directions on which tool to use as replacement:
  • Calculate Distance
  • Calculate Travel Cost
  • Cost Allocation
  • Cost Back Link
  • Cost Distance
  • Cost Path
  • Euclidean Allocation
  • Euclidean Back Direction
  • Euclidian Direction
  • Euclidean Distance
  • Path Distance
  • Path Distance Allocation
  • Path Distance Back Link

arcgis.raster.orthomapping

• Enables tools to run in asynchronous manner (See future parameter on tools)

arcgis.geoanalytics

• Adds new function to analyze_patterns
• gwr
• Adds documentation in find_locations:
• parameters to find_dwell_locations
  • time_boundary_split
  • time_split_unit
  • time_reference

arcgis.geocoding

• Adds missing parameter documentation for geocode() function
• match_out_of_range
• location_type
• lang_code
• source_country
• Adds missing parameter decumentation for reverse_geocode() function
• feature_types
• location_type
• lang_code
• Adds missing parameter documentation for suggest()
• search_extent
• max_suggestions
• country_code

arcgis.mapping

• New arcgis.mapping.ogc submodule to support OGC Layers:
  • CSVLayer
  • GeoJSONLayer
  • GeoRSSLayer
  • WMSLayer
  • WMTSLayer
• Adds ability to set the WebMap basemap to instances of:
  • WebMap
  • MapImageLayer
  • ImageryLayer
  • MapView

    NOTE: See WebMap.basemap Usage Example /#3

arcgis.widgets

• Updates the map widget to use JavaScript API 4.15
• Adds support for snychronized navigation between mutiple MapViews
  • sync_navigation(_mapview_) - See USAGE EXAMPLE
  • unsync_navigation(_mapview_) - unlinks navitation between two MapView instances
• Adds properties to map widget:
• scale
• snap_to_zoom

arcgis.apps

• workforce

  • New web_app_link property on Workforce Assignment class
  • Sets assigned_date argument value if none is entered when Assignment created

• tracker

  • Allows administrators to use custom parameters when enabling location tracking
  • Sets description and snippet of the created item

arcgis.learn

• Adds models to module:
• FullyConnectedNetwork - fully connected neural network for tabular/feature data
• MLModel - integrates Scikit-learn's machine learning models
• SuperResolution
• YOLOv3
• FasterRCNN
• Adds ModelExtension class - for customized sources
• Adds prepare_tabulardata() function for data preperation
• Adds Transform3d class for data augmentation to PointCloud data
• Adds TensorFlow backend support to the following models, enabling tf-lite model creation for deployment to mobile devices and integration with AppStudio and Survey123 (private beta support):
• Unet
• FeatureClassifier
• SingleShotDetector
• Adds load_model() function for direct loading of any model
• Adds visualization, data augmentation, and 3d transformation support to PointCNN model
• Enhances UnetClassifier and PSPNetClassifier models by adding:
• per class precision recall and f1score
• balanced loss
• focal loss
• mixup aug
• Adds Multiple-GPU support to:
• MaskRCNN
• Unet
• DeepLab
• PSPNetClassifier
• Adds Model Extension support
• Adds PointCNN.show_results() documentation to API Reference
• documents max_sampling_points parameter for raster rendering
• Adds predict_h5() method to PointCNN model
• Adds support for non continous class mappings when using prepare_data() for PointCNN
• Adds iters_per_epoch parameter documentation to PointCNN.fit()
• Enhances prepare_data() with Keyword Arguments table to document parameters pertinent to multispectral imagery input
• Adds performance metric functions to EntityRecognizer
• metrics_per_label()
• precision_score()
• recall_score()
• f1_score()
• Adds NoData support to image segmentation models:
• Unet
• DeepLab
• PSPNetClassifier
• Enhances visualizations when using PointCNN workflows

Fixes

Guides

• Updates to Working with different authentication schemes
• clarifies non-interactive login experience
• Updates to Geospatial deep learning with [arcgis.learn](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#)
• clarifies installation steps
• Updates to Named Entity Extraction Workflow with [arcgis.learn](https://developers.arcgis.com/python/api-reference/arcgis.learn.toc.html#)
• clarifies prerequisites

Samples

• Updates to Finding suitable spots for placing heart defibrillator equipments in public
• formatting changes
• use of filter property to query FeatureLayer
• Updates to Safe streets to schools
• zoom levels adjusted for maps
• Updates to Land Cover Classification using Satellite Imagery and Deep Learning
• refines search for data
• Updates to Detecting Swimming Pools using Satellite Imagery and Deep Learning
• adjusts library imports
• Updates to Madison city crime analysis
• clarifies prerequisites

arcgis.gis

• Removes incorrect Exception message on UserManager.create() when ., @ and/or - in username argument
• Removes incorrect Exception messge on User.update_license_type() and [User.update_role()] when ., @ and/or - in username
• Fixes error when invalid input passed to get() method on ContentManager, UserManager, or GroupManager
• Fixes 301/redirect error with PKI authentication schemes not using the web adaptor
• Fixes Item.copy() method failing to copy the url property to the new item
• Fixes input format when using assign_to_items() for Group categories

arcgis.gis.admin

• Fixes issue where updating the Web GIS name did not update Home Page

arcgis.features

• Fixes to_featureclass() method on the Spatially Enabled Dataframe dropping Z values from geometries
• Fixes from_featureclass() method on the Spatially Enabled Dataframe nullifying zero-length polyline geometries
• Fixes issue with geometry operations on Spatially Enabled Dataframe.SHAPE column not preserving spatial reference
• Improves performance of Spatially Enabled Dataframe from_xy() method
• Updates standard feature analysis functions reference documentation to refer to Feature Input
• Fixes error with true_centroid() on the Spatially Enabled DataFrame in ArcPy environments

arcgis.features.analysis

• summarize_data
• Adds parameters to aggregate_points()
• bin_type
• bin_size
• bin_size_unit
• future

arcgis.raster.functions.gbl

• Enhances region_group parameter documentation

arcgis.geometry

• Enhancements to is_valid property on Geometry instances. (See Example.)
• Fixes code to compare whether two SpatialReferences are equal

arcgis.raster.functions

• Fixes issue with classify() when second raster not specified

arcgis.widgets

• Fixes issue with zoom_to_layer() not working with spatial reference in wkt formatt

arcgis.apps

• tracker

  • Fixes support service name conflicts

arcgis.learn

• Improves overall Exception handling
• Increases flexibility of from_model() methods in ArcGIS Models to allow for different backends
• Fixes metric calculation error in PointCNN.compute_precision_recall()
• Fixes error when using prepare_data() for PSPNetClassifier when balancing statistics are not present
• Fixes error with PSPNetClassifer.lr_find() with all backbones
• Fixes display issues with lr.find() in DeepLab and PSPNetClassifier
• Fixes show_batch() error with multispectral models
• Fixes prepare_data() errors when the training data was created with Output No Feature Data option checked
• Fixes error with export_point_dataset() on PointCNN when working with data that has non-contiguous classes
• Fixes lr_find() and fit() functions when used with export_point_dataset() output created when max_points argument exceeds default
• Fixes errors with RetinaNet.predict() on objects created with resnet101 or resnet152 backbones
• Fixes error with show_results() when using multispectral data
• Multiple improvements to the predict_las() function using the PointCNN model
• preserves point cloud formatting and version of input and output
• fixes output and messaging when print_metrics is True
• fixes loss of information in the las file
• Removal of 'non-zero' classcodes specific accuracy at completion of epoch when using PointCNN
• Change default behavior of export_point_dataset( ) extra_features to empty