Add datacube variables support as MLM input and output dimensions

.remarkignore

@@ -16,3 +16,7 @@ env/
 node_modules/
 docs/_build/
 docs/build/
+
+# potentially conflicting dev installs
+stac-mlm/
+mlm/

CHANGELOG.md

@@ -9,6 +9,14 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Added
 
+- Add `ModelDataVariable` to `stac_model` for corresponding `mlm:input` and `mlm:output` definitions as the JSON schema.
+- Add `variables` properties to [Model Input Object](README.md#model-input-object)
+  to allow specifying the relevant data variables used by the model,
+  with cross-references to the [datacube](https://github.com/stac-extensions/datacube) extension
+  (relates to [#90](https://github.com/stac-extensions/mlm/issues/90)).
+- Add `bands` and `variables` properties to [Model Output Object](README.md#model-output-object)
+  to allow specifying the relevant bands or variables produced by the model if any applies.
+- Add `downscaling` to [Tasks](./README.md#task-enum) as common operation for climate variable models.
 - Add [ML-Model Legacy](./docs/legacy/ml-model.md) document providing migration guidance
   from the deprecated [ML-Model](https://github.com/stac-extensions/ml-model) extension
   (relates to [stac-extensions/ml-model#16](https://github.com/stac-extensions/ml-model/pull/16)).
@@ -24,7 +32,18 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   (fixes [#69](https://github.com/stac-extensions/mlm/issues/69)).
 
 ### Changed
-
+
+- Update `stac-model==0.4.0` to provide corresponding additions for `variables` reference.
+- Refactor `ModelInput` and `ModelOutput` objects to use a new `ModelBandsOrVariablesReferences` definition
+  combining the `ModelBand` and `ModelDataVariable` lists.
+- Moved `ModelBand` from `stac_model.input` to `stac_model.base` since it is now required
+  by both `ModelInput` and `ModelOutput` objects.
+- Refactor the JSON schema to check for `bands` and `variables` references within both `mlm:input` and `mlm:output`.
+  If either location detects that either `bands` or `variables` is provided, their corresponding sets of extensions
+  providing relevant descriptions are verified.
+- Refactor the JSON schema `mlm:output` property to employ a `ModelOutput` object definition
+  rather than directly provided properties nested under the array.
+- Refactor the JSON schema to allow the omission of `bands` under `mlm:input` if the `variables` property is provided.
 - Update `stac-model==0.3.0` to provide `ValueScalingObject` from installed package.
 
 ### Deprecated
@@ -37,7 +56,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Fixed
 
-- n/a
+- Fix `stac_model.output.ModelOutput` enforcing the need to specify `classification:classes` or `classes`.
+  The property can now be omitted if the model does not need to indicate that it produces a classification output.
+- Fix missing ``encoding="utf-8"`` parameters in `open` calls leading to failing parsing of example JSON STAC Item
+  when they contain non-ASCII characters.
 
 ## [v1.4.0](https://github.com/stac-extensions/mlm/tree/v1.4.0)
 

Makefile

@@ -3,7 +3,15 @@ SHELL ?= /usr/bin/env bash
 
 # use the directory rather than the python binary to allow auto-discovery, which is more cross-platform compatible
 PYTHON_PATH := $(shell which python)
-PYTHON_ROOT := $(shell dirname $(dir $(PYTHON_PATH)))
+# handle whether running on Windows or Unix-like systems
+ifneq ($(findstring $(PYTHON_PATH),bin/python),)
+	PYTHON_ROOT := $(shell dirname $(dir $(PYTHON_PATH)))
+else
+	PYTHON_ROOT := $(shell dirname $(PYTHON_PATH))
+endif
+ifeq ($(patsubst %/bin,,$(lastword $(PYTHON_ROOT))),)
+  PYTHON_ROOT := $(dir $(PYTHON_ROOT))
+endif
 UV_PYTHON_ROOT ?= $(PYTHON_ROOT)
 
 # to actually reuse an existing virtual/conda environment, the 'UV_PROJECT_ENVIRONMENT' variable must be set to it
@@ -54,7 +62,7 @@ format: codestyle
 #* Linting
 .PHONY: test
 test: setup
-	$(UV_COMMAND) run --python "$(UV_PYTHON_ROOT)" pytest -c pyproject.toml --cov-report=html --cov=stac_model tests/
+	$(UV_COMMAND) run --python "$(UV_PYTHON_ROOT)" pytest -c pyproject.toml -v --cov-report=html --cov=stac_model tests/
 
 .PHONY: check
 check: check-examples check-markdown check-lint check-mypy check-safety check-citation

best-practices.md

@@ -8,6 +8,7 @@ models or creating tools to work with STAC.
 
 - [Using STAC Common Metadata Fields for the MLM Extension](#using-stac-common-metadata-fields-for-the-mlm-extension)
 - [Recommended Extensions to Compose with the MLM Extension](#recommended-extensions-to-compose-with-the-mlm-extension)
+  - [STAC Bands, EO, Raster and DataCube Extensions](#stac-bands-eo-raster-and-datacube-extensions)
   - [Processing Extension](#processing-extension)
   - [ML-AOI and Label Extensions](#ml-aoi-and-label-extensions)
   - [Classification Extension](#classification-extension)
@@ -68,6 +69,28 @@ information regarding these references, see the [ML-AOI and Label Extensions](#m
 
 ## Recommended Extensions to Compose with the MLM Extension
 
+### STAC Bands, EO, Raster and DataCube Extensions
+
+When a model takes as input or produces as output raster bands, spatio-temporal data variables or leverages 
+Earth Observation data, it is **STRONGLY RECOMMENDED** to use a corresponding STAC Core or STAC Extension specification
+to describe them. This ensures that common metadata fields can be used both for search and retrieval of available
+data provided by catalogues, as well as for models using or producing them.
+
+Relevant core fields or extensions include:
+
+- [STAC Common Metadata Bands](https://github.com/radiantearth/stac-spec/blob/v1.1.0/commons/common-metadata.md#bands)
+- [STAC EO Extension](https://github.com/stac-extensions/eo)
+- [STAC Raster Extension](https://github.com/stac-extensions/raster)
+- [STAC DataCube Extension](https://github.com/stac-extensions/datacube)
+
+For further details on how to use these extensions and their implications in the context of
+the Machine Learning Model extension, please refer to the corresponding sections:
+
+- [Model Input Object](./README.md#model-input-object)
+- [Model Output Object](./README.md#model-output-object)
+- [Bands and Statistics](./README.md#bands-and-statistics)
+- [Data Variables](./README.md#data-variables)
+
 ### Processing Extension
 
 It is recommended to use at least the `processing:lineage` and `processing:level` fields from

examples/collection.json

@@ -56,6 +56,10 @@
       "href": "item_bands_expression.json",
       "rel": "item"
     },
+    {
+      "href": "item_datacube_variables.json",
+      "rel": "item"
+    },
     {
       "href": "item_eo_bands.json",
       "rel": "item"