Skip to content

Releases: descarteslabs/descarteslabs-python

Client release v1.4.1

08 Sep 21:13
Compare
Choose a tag to compare

Fixed

No more irrelevant DeprecationWarnings when importing the descarteslabs package (#235). Deprecated functionality in the package will now show FutureWarnings instead.

Workflows (channel v0-16) - Fixed

  • wf.map.geocontext doesn't raise an error about the CRS of the map
  • wf.flows doesn't raise an error about versions from incompatible channels

Client release v1.4.0

20 Aug 19:44
Compare
Choose a tag to compare

Catalog client

  • Example code has been cleaned up.

Workflows (channel v0-16) - Added

  • Sharing of any Workflows object as a Workflow with version and access control. Browse through shared Workflows with the wf.flows browser widget.
  • Upload images to the DL catalog from Workflows jobs. Usage details can be found in the docs.
  • wf.np.median
  • Job.cancel() to cancel running jobs.
  • Transient failures in Jobs are automatically retried, resulting in fewer errors.
  • Search widget on wf.map by default.

Workflows - Fixed

  • Bitwise operations on imagery no longer fail
  • wf.np.linspace no longer fails when being called correctly
  • .median is slightly less prone to OOM errors

Workflows - Changed

  • Array.to_imagery now accepts KnownDict for bandinfo and properties.
  • Numbers can now be constructed from Strs

Client release v1.3.0

16 Jun 15:08
Compare
Choose a tag to compare

Workflows (channel v0-15) - Added

  • Output formats for .compute including GeoTIFF, JSON, PyArrow, and MessagePack. Usage details can be found in the docs.
  • Destinations for Job results: download and email. Usage details can be found in the docs.
  • Save .compute outputs to a file with the file= argument.
  • Pixel value inspector: click in the map widget to view pixel values.
  • wf.ifelse for simple conditional logic.
  • NumPy functions including hypot, bitwise_and, bitwise_or, bitwise_xor, bitwise_not, invert, and ldexp
  • Bitwise Array and MaskedArray operations
  • size attribute on Array and MaskedArray
  • astype function on Array and MaskedArray for changing the dtype
  • flatten function on Array and MaskedArray for flattening into a 1D array
  • MaskedArray.compressed for getting all unmasked data as a 1D array
  • get function on Dict and KnownDict for providing a default value if a key does not exist
  • nbands attribute on Image and ImageCollection
  • proxify can handle scenes.GeoContexts
  • Dict.contains, Dict.length

Workflows - Fixed

  • Fewer failures and hanging calls when connecting to the Workflows backend (like .compute, .visualize, Job.get, etc.)
  • wf.numpy.histogram works correctly with computed values for range and bins (such as range=[arr.min(), arr.max()])
  • More consistent throughput when a large number of jobs are submitted
  • Arrays can now be constructed from proxy Lists
  • MaskedArray.filled works correctly when passed Python values
  • Long-running sessions (like Jupyter kernels) refresh credentials instead of failing with auth errors after many hours of use
  • wf.numpy.dot and wf.numpy.einsum no longer fail when being called correctly
  • Occasional errors like ('array-89199362e9a5d598fb5c82805136834d', 0, 0) when calling wf.compute() with multiple values are resolved

Workflows - Changed

  • pick_bands accepts duplicate band names. Enjoy easier Sentinel-1 "vv vh vv" visualizations!
  • ImageCollection.from_id is always ordered by date
  • wf.numpy.percentile no longer accepts an axis argument
  • breaking wf.Job construction and interface changes:
    • Use a single wf.Job(..) call instead of wf.Job.build(...).execute() to create and launch a Job
    • New Job.result_to_file method
    • Job.status is removed in favor of a single Job.stage
    • wf.TimeoutError renamed to wf.JobTimeoutError

Client release v1.2.0

23 Apr 22:17
Compare
Choose a tag to compare

Workflows (channel v0-14) - Added

  • 191 functions from NumPy are available for Workflows Arrays, including parts of the numpy.linalg and numpy.ma submodules. See the full list on the docs.
  • index_to_coords and coords_to_index methods on Image/ImageCollection/GeoContext for converting between geospatial and array coordinates
  • value_at function on Image and ImageCollection for extracting single pixel values at spatial coordinates.

Workflows - Fixed

  • Using datetimes as parameters to visualize behaves correctly.

Client release v1.1.3

02 Apr 18:26
Compare
Choose a tag to compare

Catalog client

  • Fixed a bug that prevented uploading ndarrays of type uint8

Workflows (channel v0-13) - Added

  • Array support for argmin, argmax, any, all
  • pick_bands supports an allow_missing kwarg to drop band names that may be missing from the data without an error.
  • wf.compute supports passing lists or tuples of items to compute at the same time. Passing multiple items to wf.compute, rather than calling obj.compute for each separately, is usually faster.
  • Casting from Bool to Int: wf.Int(True)
  • Experimental .inspect() method for small computations during interactive use.

Workflows - Changed

  • [breaking] Array no longer uses type parameters: now you construct an Array with wf.Array([1, 2, 3]), not wf.Array[wf.Int, 1]([1, 2, 3]). Remember, Array is an experimental API and will continue to make frequent breaking changes!
  • Workflows now reuses the same gRPC client by default---so repeated or parallel calls to .compute, etc. will be faster. Calling .compute within a thread pool will also be significantly more efficient.

Workflows - Fixed

  • wf.numpy.histogram correctly accepts a List[Float] as the range argument

Client release v1.1.2

13 Mar 14:55
Compare
Choose a tag to compare

1.1.2 fixes a bug which caused Workflows map layers to behave erratically when changing colormaps.

Client release v1.1.1

12 Mar 00:13
Compare
Choose a tag to compare

1.1.1 fixes a packaging issue that caused import descarteslabs.workflows to fail.

It also makes NumPy an explicit dependency. NumPy was already a transitive dependency, so this shouldn't cause any changes.

You should NOT install version 1.1.0; 1.1.1 should be used instead in all circumstances.

Client release v1.1.0

11 Mar 18:55
Compare
Choose a tag to compare

[1.1.0] - 2020-03-11

Catalog client

  • Image.upload() now emits a deprecation warning if the image has a cs_code or projection property.
    The projection defined in the uploaded file is always used and applied to the resulting image in the Catalog.
  • Image.upload_ndarray() now emits a deprecation warning if the image has both a cs_code and a projection
    property. Only one of them may be supplied, and cs_code is given preference.

Scenes

  • SceneCollection.download_mosaic has new default behavior for mask_alpha wherein the alpha band will be
    used as a mask by default if it is available for all scenes in the collection, even if it is not specified in
    the list of bands.

Workflows (channel v0-12) - Added

  • Experimental Array API following the same syntax as NumPy arrays. It supports vectorized operations, broadcasting,
    and multidimensional indexing.
    • ndarray attribute of Image and ImageCollection will return a MaskedArray.
    • Over 60 NumPy ufuncs are now callable with Workflows Array.
    • Includes other useful Array functions like min(), median(), transpose(), concatenate(), stack(), histogram(), and reshape().
  • ImageCollection.sortby_composite() for creating an argmin/argmax composite of an ImageCollection.
  • Slicing of List, Tuple, Str, and ImageCollection.
  • wf.range for generating a sequence of numbers between start and stop values.
  • ImageCollectionGroupby.mosaic() for applying ImageCollection.mosaic to each group.
  • wf.exp(), wf.square(), wf.log1p(), wf.arcsin(), wf.arccos(), and wf.arctan()
  • Datetime.is_between() for checking if a Datetime falls within a specified date range
  • FeatureCollection.contains()
  • Container operations on GeometryCollection including:
    • GeometryCollection.contains()
    • GeometryCollection.sorted()
    • GeometryCollection.map()
    • GeometryCollection.filter()
    • GeometryCollection.reduce()
  • List and Tuple can now be compared with other instances of their type via __lt__(), __eq__() etc.
  • List.__add__() and List.__mul__() for concatenating and duplicating Lists.

Workflows - Changed

  • Products without alpha band and nodata value are rejected, instead of silently producing unwanted behavior.
  • ImageCollection.concat_bands now throws a better error when trying to concatenate bands from another ImageCollection that is not the same length.
  • Any is now promotable to all other types automatically.
  • Better error when trying to iterate over Proxytypes.
  • Interactive map: calls to visualize now clear layer errors.
  • Interactive map: when setting scales, invalid values are highlighted in red.

Workflows - Fixed

  • Better errors when specifying invalid type parameters for Proxytypes that require them.
  • Field access on Feature, FeatureCollection, Geometry, and GeomeryCollection no longer fails.
  • In from_id, processing level 'cubespline' no longer fails.

Client release v1.0.0

21 Jan 03:53
Compare
Choose a tag to compare
As of January 1st, 2020, the client library no longer supports Python 2. For more information, please contact [email protected]. For help with porting to Python 3, please visit https://docs.python.org/3/howto/pyporting.html.

Catalog client

  • There is an entirely new backend supporting asynchronous uploads of image files and ndarrays with
    the catalog client. There are minor changes to the ImageUpload class (a new events field has subsumed
    errors, and the job_id field has been removed) but the basic interface is unchanged so most
    code will keep functioning without any changes.
  • It is now possible to cancel image uploads.
  • Errors messages are now easier to read.
  • Many improvements to the documentation.
  • You can now create or retrieve an existing object using the get_or_create method.
  • Retrieving a Band or Image by name is now possible by calling get_band or get_image on the
    Product instance. You can also use the Product's named_id function to get a complete id for
    images and bands.
  • A new convenience function make_valid_name on Image and Band classes will return a sanitized
    name without invalid characters.
  • A new property ATTRIBUTES enumerates which attributes are available for a specific catalog object.
  • Trying to set an attribute that does not exist will now raise AttributeError.
  • update_related_objects_permissions() should no longer fail with a JSON serialization error.
  • Setting a read-only attribute will now raise an AttributeValidationError.
  • Saving a new object while one with the same id already exists will now raise a ConflictError
    instead of BadRequestError.
  • If a retrieved object has since been deleted from the catalog, saving any changes or trying to
    reload it will now raise a DeletedObjectError.
  • Resolution fields now accept string values such as "10m" or "0.008 degrees". If the value cannot
    be parsed, an AttributeValidationError will be raised.
  • Changes to the extra_properties attribute are now tracked correctly.

Packaging

  • This release no longer supports Python 2.
  • This package is now distributed as a Python 3 wheel which will speed up installation.

Workflows (channel v0-11) - Added

  • Handling of missing data via empty ImageCollections
    • ImageCollection.from_id returns an empty ImageCollection if no data exist for the given time/place, rather than an error
    • ImageCollection.filter returns an empty ImageCollection if the predicate is False for every Image, rather than an error
    • Image.replace_empty_with and ImageCollection.replace_empty_with for explicitly filling in missing data
    • See the Workflows guide for more information
  • Docstrings and examples on every class and function!
  • Assigning new metadata to Image properties & bandinfo: Image.with_properties(), Image.with_bandinfo()
  • Interactive map: colorbar legends on layers with colormaps (requires matplotlib)
  • Dict.from_pairs: construct a Dict from a sequence of key-value pairs
  • Map displays a fullscreen button by default ([breaking] if your code adds one, you'll now get two)
  • wf.concat for concatentating Image and ImageCollection objects
    • ImageCollection.concat now accepts Image objects; new Image.concat accepts Image or ImageCollection
  • ImageCollection.mosaic()
  • FeatureCollection.sorted(), FeatureCollection.length(), FeatureCollection.__reversed__()
  • GeometryCollection.length(), GeometryCollection.__reversed__()

Workflows - Changed

  • wf.zip now supports ImageCollection, FeatureCollection, GeometryCollection as well as List and Str
  • Get a GeoContext for the current bounds of the map in any resolution, shape, or CRS (including "utm", which automatically picks the right UTM zone for you) with wf.map.geocontext. Also now returns a Scenes GeoContext for better introspection and use with Raster.
  • Better backend type-checking displays the possible arguments for most functions if called incorrectly
  • arr_shape included when calling wf.GeoContext.compute()
  • More readable errors when communication with the backend fails
  • Interactive map: layout handles being resized, for example setting wf.map.layout.height = '1000px'
  • Any is no longer callable; Any.cast encouraged
  • remove_layer and clear_layers moved from wf.interactive.MapApp class to wf.interactive.Map (non-breaking change)
  • [possibly breaking] band renaming in binary operators only occurs when broadcasting: red + red is just red, rather than red_add_red. red + blue is still red_add_blue. Code which depends on accessing bands by name may need to change.

Workflows - Fixed

  • wf.where propagates masks correctly, and handles metadata correctly with multi-band inputs
  • processing_level="surface" actually returns surface-reflectance-processed imagery
  • ImageCollection.sorted() works properly
  • Viewing global-extent WGS84 images on the Workflows map no longer causes errors
  • List proxytype no longer infinitely iterable in Python
  • Repeated use of axis="bands" works correctly
  • ImageCollection.from_images correctly aligns the bands of the inputs
  • Numeric casting (wf.Int(wf.Float(2.2))) works as expected
  • More descriptive error when constructing an invalid wf.Datetime
  • Computing a single Bool value derived from imagery works correctly

Client release v0.28.1

10 Dec 23:20
Compare
Choose a tag to compare

Changed

  • Update workflows client channel
  • Workflows map UI is more stable: errors and layers won't fill the screen