rslearn 0.0.12__tar.gz → 0.0.13__tar.gz

{rslearn-0.0.12/rslearn.egg-info → rslearn-0.0.13}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rslearn
-Version: 0.0.12
+Version: 0.0.13
 Summary: A library for developing remote sensing datasets and models
 Author: OlmoEarth Team
 License:                                  Apache License
@@ -214,7 +214,7 @@ License-File: LICENSE
 License-File: NOTICE
 Requires-Dist: boto3>=1.39
 Requires-Dist: fiona>=1.10
-Requires-Dist: fsspec>=2025.9.0
+Requires-Dist: fsspec>=2025.10.0
 Requires-Dist: jsonargparse>=4.35.0
 Requires-Dist: lightning>=2.5.1.post0
 Requires-Dist: Pillow>=11.3

{rslearn-0.0.12 → rslearn-0.0.13}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "rslearn"
-version = "0.0.12"
+version = "0.0.13"
 description = "A library for developing remote sensing datasets and models"
 authors = [
     { name = "OlmoEarth Team" },
@@ -11,7 +11,7 @@ requires-python = ">=3.11"
 dependencies = [
     "boto3>=1.39",
     "fiona>=1.10",
-    "fsspec>=2025.9.0", # this is used both directly and indirectly (via universal_pathlib) in our code
+    "fsspec>=2025.10.0", # this is used both directly and indirectly (via universal_pathlib) in our code
     "jsonargparse>=4.35.0",
     "lightning>=2.5.1.post0",
     "Pillow>=11.3",

{rslearn-0.0.12 → rslearn-0.0.13}/rslearn/config/dataset.py

@@ -125,7 +125,8 @@ class BandSetConfig:
         self,
         config_dict: dict[str, Any],
         dtype: DType,
-        bands: list[str],
+        bands: list[str] | None = None,
+        num_bands: int | None = None,
         format: dict[str, Any] | None = None,
         zoom_offset: int = 0,
         remap: dict[str, Any] | None = None,
@@ -137,7 +138,10 @@ class BandSetConfig:
         Args:
             config_dict: the config dict used to configure this BandSetConfig
             dtype: the pixel value type to store tiles in
-            bands: list of band names in this BandSetConfig
+            bands: list of band names in this BandSetConfig. One of bands or num_bands
+                must be set.
+            num_bands: the number of bands in this band set. The bands will be named
+                B00, B01, B02, etc.
             format: the format to store tiles in, defaults to geotiff
             zoom_offset: store images at a resolution higher or lower than the window
                 resolution. This enables keeping source data at its native resolution,
@@ -155,6 +159,14 @@ class BandSetConfig:
                 materialization when creating mosaics, to determine which parts of the
                 source images should be copied.
         """
+        if (bands is None and num_bands is None) or (
+            bands is not None and num_bands is not None
+        ):
+            raise ValueError("exactly one of bands and num_bands must be set")
+        if bands is None:
+            assert num_bands is not None
+            bands = [f"B{idx}" for idx in range(num_bands)]
+
         if class_names is not None and len(bands) != len(class_names):
             raise ValueError(
                 f"the number of class lists ({len(class_names)}) does not match the number of bands ({len(bands)})"
@@ -187,9 +199,16 @@ class BandSetConfig:
         kwargs = dict(
             config_dict=config,
             dtype=DType(config["dtype"]),
-            bands=config["bands"],
         )
-        for k in ["format", "zoom_offset", "remap", "class_names", "nodata_vals"]:
+        for k in [
+            "bands",
+            "num_bands",
+            "format",
+            "zoom_offset",
+            "remap",
+            "class_names",
+            "nodata_vals",
+        ]:
             if k in config:
                 kwargs[k] = config[k]
         return BandSetConfig(**kwargs)  # type: ignore

{rslearn-0.0.12 → rslearn-0.0.13}/rslearn/data_sources/planetary_computer.py

@@ -827,3 +827,55 @@ class Sentinel1(PlanetaryComputer):
                 kwargs[k] = d[k]
 
         return Sentinel1(**kwargs)
+
+
+class Naip(PlanetaryComputer):
+    """A data source for NAIP data on Microsoft Planetary Computer.
+
+    See https://planetarycomputer.microsoft.com/dataset/naip.
+    """
+
+    COLLECTION_NAME = "naip"
+    ASSET_BANDS = {"image": ["R", "G", "B", "NIR"]}
+
+    def __init__(
+        self,
+        **kwargs: Any,
+    ):
+        """Initialize a new Naip instance.
+
+        Args:
+            band_names: list of bands to try to ingest.
+            kwargs: additional arguments to pass to PlanetaryComputer.
+        """
+        super().__init__(
+            collection_name=self.COLLECTION_NAME,
+            asset_bands=self.ASSET_BANDS,
+            **kwargs,
+        )
+
+    @staticmethod
+    def from_config(config: RasterLayerConfig, ds_path: UPath) -> "Naip":
+        """Creates a new Naip instance from a configuration dictionary."""
+        if config.data_source is None:
+            raise ValueError("config.data_source is required")
+        d = config.data_source.config_dict
+        kwargs = {}
+
+        if "timeout_seconds" in d:
+            kwargs["timeout"] = timedelta(seconds=d["timeout_seconds"])
+
+        if "cache_dir" in d:
+            kwargs["cache_dir"] = join_upath(ds_path, d["cache_dir"])
+
+        simple_optionals = [
+            "query",
+            "sort_by",
+            "sort_ascending",
+            "max_items_per_client",
+        ]
+        for k in simple_optionals:
+            if k in d:
+                kwargs[k] = d[k]
+
+        return Naip(**kwargs)

{rslearn-0.0.12 → rslearn-0.0.13}/rslearn/dataset/handler_summaries.py

@@ -20,6 +20,7 @@ class LayerPrepareSummary:
     # Counts
     windows_prepared: int
     windows_skipped: int
+    windows_rejected: int
     get_items_attempts: int
 
 

{rslearn-0.0.12 → rslearn-0.0.13}/rslearn/dataset/manage.py

@@ -118,6 +118,7 @@ def prepare_dataset_windows(
                     duration_seconds=time.monotonic() - layer_start_time,
                     windows_prepared=0,
                     windows_skipped=len(windows),
+                    windows_rejected=0,
                     get_items_attempts=0,
                 )
             )
@@ -141,6 +142,7 @@ def prepare_dataset_windows(
                     duration_seconds=time.monotonic() - layer_start_time,
                     windows_prepared=0,
                     windows_skipped=len(windows),
+                    windows_rejected=0,
                     get_items_attempts=0,
                 )
             )
@@ -181,6 +183,9 @@ def prepare_dataset_windows(
             attempts_counter=attempts_counter,
         )
 
+        windows_prepared = 0
+        windows_rejected = 0
+        min_matches = data_source_cfg.query_config.min_matches
         for window, result in zip(needed_windows, results):
             layer_datas = window.load_layer_datas()
             layer_datas[layer_name] = WindowLayerData(
@@ -191,13 +196,22 @@ def prepare_dataset_windows(
             )
             window.save_layer_datas(layer_datas)
 
+            # If result is empty and min_matches > 0, window was rejected due to min_matches
+            if len(result) == 0 and min_matches > 0:
+                windows_rejected += 1
+            else:
+                windows_prepared += 1
+
+        windows_skipped = len(windows) - len(needed_windows)
+
         layer_summaries.append(
             LayerPrepareSummary(
                 layer_name=layer_name,
                 data_source_name=data_source_cfg.name,
                 duration_seconds=time.monotonic() - layer_start_time,
-                windows_prepared=len(needed_windows),  # we assume all have succeeded
-                windows_skipped=len(windows) - len(needed_windows),
+                windows_prepared=windows_prepared,
+                windows_skipped=windows_skipped,
+                windows_rejected=windows_rejected,
                 get_items_attempts=attempts_counter.value,
             )
         )

{rslearn-0.0.12 → rslearn-0.0.13}/rslearn/models/olmoearth_pretrain/model.py

@@ -40,6 +40,7 @@ EMBEDDING_SIZES = {
     ModelID.OLMOEARTH_V1_NANO: 128,
     ModelID.OLMOEARTH_V1_TINY: 192,
     ModelID.OLMOEARTH_V1_BASE: 768,
+    ModelID.OLMOEARTH_V1_LARGE: 1024,
 }
 
 

{rslearn-0.0.12 → rslearn-0.0.13}/rslearn/train/prediction_writer.py

@@ -22,7 +22,11 @@ from rslearn.log_utils import get_logger
 from rslearn.utils.array import copy_spatial_array
 from rslearn.utils.feature import Feature
 from rslearn.utils.geometry import PixelBounds
-from rslearn.utils.raster_format import RasterFormat, load_raster_format
+from rslearn.utils.raster_format import (
+    RasterFormat,
+    adjust_projection_and_bounds_for_array,
+    load_raster_format,
+)
 from rslearn.utils.vector_format import VectorFormat, load_vector_format
 
 from .lightning_module import RslearnLightningModule
@@ -68,15 +72,18 @@ class VectorMerger(PatchPredictionMerger):
 class RasterMerger(PatchPredictionMerger):
     """Merger for raster data that copies the rasters to the output."""
 
-    def __init__(self, padding: int | None = None):
+    def __init__(self, padding: int | None = None, downsample_factor: int = 1):
         """Create a new RasterMerger.
 
         Args:
             padding: the padding around the individual patch outputs to remove. This is
                 typically used when leveraging overlapping patches. Portions of outputs
                 at the border of the window will still be retained.
+            downsample_factor: the factor by which the rasters output by the task are
+                lower in resolution relative to the window resolution.
         """
         self.padding = padding
+        self.downsample_factor = downsample_factor
 
     def merge(
         self, window: Window, outputs: Sequence[PendingPatchOutput]
@@ -87,8 +94,8 @@ class RasterMerger(PatchPredictionMerger):
         merged_image = np.zeros(
             (
                 num_channels,
-                window.bounds[3] - window.bounds[1],
-                window.bounds[2] - window.bounds[0],
+                (window.bounds[3] - window.bounds[1]) // self.downsample_factor,
+                (window.bounds[2] - window.bounds[0]) // self.downsample_factor,
             ),
             dtype=dtype,
         )
@@ -104,7 +111,10 @@ class RasterMerger(PatchPredictionMerger):
             # If the output is not on the left or top boundary, then we should apply
             # the padding (if set).
             src = output.output
-            src_offset = (output.bounds[0], output.bounds[1])
+            src_offset = (
+                output.bounds[0] // self.downsample_factor,
+                output.bounds[1] // self.downsample_factor,
+            )
             if self.padding is not None and output.bounds[0] != window.bounds[0]:
                 src = src[:, :, self.padding :]
                 src_offset = (src_offset[0] + self.padding, src_offset[1])
@@ -116,7 +126,10 @@ class RasterMerger(PatchPredictionMerger):
                 src=src,
                 dst=merged_image,
                 src_offset=src_offset,
-                dst_offset=(window.bounds[0], window.bounds[1]),
+                dst_offset=(
+                    window.bounds[0] // self.downsample_factor,
+                    window.bounds[1] // self.downsample_factor,
+                ),
             )
 
         return merged_image
@@ -330,9 +343,13 @@ class RslearnWriter(BasePredictionWriter):
                 self.output_layer, self.layer_config.band_sets[0].bands
             )
             assert isinstance(self.format, RasterFormat)
-            self.format.encode_raster(
-                raster_dir, window.projection, window.bounds, merged_output
+
+            # In case the merged_output is at a different resolution than the window,
+            # get adjusted projection and bounds for writing it.
+            projection, bounds = adjust_projection_and_bounds_for_array(
+                window.projection, window.bounds, merged_output
             )
+            self.format.encode_raster(raster_dir, projection, bounds, merged_output)
 
         elif self.layer_config.layer_type == LayerType.VECTOR:
             layer_dir = window.get_layer_dir(self.output_layer)

rslearn-0.0.13/rslearn/train/tasks/embedding.py

@@ -0,0 +1,116 @@
+"""Embedding task."""
+
+from typing import Any
+
+import numpy.typing as npt
+import torch
+from torchmetrics import MetricCollection
+
+from rslearn.utils import Feature
+
+from .task import Task
+
+
+class EmbeddingTask(Task):
+    """A dummy task for computing embeddings.
+
+    This task does not compute any targets or loss. Instead, it is just set up for
+    inference, to save embeddings from the configured model.
+    """
+
+    def process_inputs(
+        self,
+        raw_inputs: dict[str, torch.Tensor],
+        metadata: dict[str, Any],
+        load_targets: bool = True,
+    ) -> tuple[dict[str, Any], dict[str, Any]]:
+        """Processes the data into targets.
+
+        Args:
+            raw_inputs: raster or vector data to process
+            metadata: metadata about the patch being read
+            load_targets: whether to load the targets or only inputs
+
+        Returns:
+            tuple (input_dict, target_dict) containing the processed inputs and targets
+                that are compatible with both metrics and loss functions
+        """
+        return {}, {}
+
+    def process_output(
+        self, raw_output: Any, metadata: dict[str, Any]
+    ) -> npt.NDArray[Any] | list[Feature]:
+        """Processes an output into raster or vector data.
+
+        Args:
+            raw_output: the output from prediction head.
+            metadata: metadata about the patch being read
+
+        Returns:
+            either raster or vector data.
+        """
+        # Just convert the raw output to numpy array that can be saved to GeoTIFF.
+        return raw_output.cpu().numpy()
+
+    def visualize(
+        self,
+        input_dict: dict[str, Any],
+        target_dict: dict[str, Any] | None,
+        output: Any,
+    ) -> dict[str, npt.NDArray[Any]]:
+        """Visualize the outputs and targets.
+
+        Args:
+            input_dict: the input dict from process_inputs
+            target_dict: the target dict from process_inputs
+            output: the prediction
+
+        Returns:
+            a dictionary mapping image name to visualization image
+        """
+        # EmbeddingTask is only set up to support `model predict`.
+        raise NotImplementedError
+
+    def get_metrics(self) -> MetricCollection:
+        """Get the metrics for this task."""
+        return MetricCollection({})
+
+
+class EmbeddingHead(torch.nn.Module):
+    """Head for embedding task.
+
+    This picks one feature map from the input list of feature maps to output. It also
+    returns a dummy loss.
+    """
+
+    def __init__(self, feature_map_index: int | None = 0):
+        """Create a new EmbeddingHead.
+
+        Args:
+            feature_map_index: the index of the feature map to choose from the input
+                list of multi-scale feature maps (default 0). If the input is already
+                a single feature map, then set to None.
+        """
+        super().__init__()
+        self.feature_map_index = feature_map_index
+
+    def forward(
+        self,
+        features: torch.Tensor,
+        inputs: list[dict[str, Any]],
+        targets: list[dict[str, Any]] | None = None,
+    ) -> tuple[torch.Tensor, dict[str, Any]]:
+        """Select the desired feature map and return it along with a dummy loss.
+
+        Args:
+            features: list of BCHW feature maps (or one feature map, if feature_map_index is None).
+            inputs: original inputs (ignored).
+            targets: should contain classes key that stores the per-pixel class labels.
+
+        Returns:
+            tuple of outputs and loss dict
+        """
+        if self.feature_map_index is not None:
+            features = features[self.feature_map_index]
+
+        return features, {"loss": 0}

{rslearn-0.0.12 → rslearn-0.0.13}/rslearn/utils/raster_format.py

@@ -123,6 +123,44 @@ def get_transform_from_projection_and_bounds(
     )
 
 
+def adjust_projection_and_bounds_for_array(
+    projection: Projection, bounds: PixelBounds, array: npt.NDArray
+) -> tuple[Projection, PixelBounds]:
+    """Adjust the projection and bounds to correspond to the resolution of the array.
+
+    The returned projection and bounds cover the same spatial extent as the inputs, but
+    are updated so that the width and height match that of the array.
+
+    Args:
+        projection: the original projection.
+        bounds: the original bounds.
+        array: the CHW array for which to compute an updated projection and bounds. The
+            returned bounds will have the same width and height as this array.
+
+    Returns:
+        a tuple of adjusted (projection, bounds)
+    """
+    if array.shape[2] == (bounds[2] - bounds[0]) and array.shape[1] == (
+        bounds[3] - bounds[1]
+    ):
+        return (projection, bounds)
+
+    x_factor = array.shape[2] / (bounds[2] - bounds[0])
+    y_factor = array.shape[1] / (bounds[3] - bounds[1])
+    adjusted_projection = Projection(
+        projection.crs,
+        projection.x_resolution / x_factor,
+        projection.y_resolution / y_factor,
+    )
+    adjusted_bounds = (
+        round(bounds[0] * x_factor),
+        round(bounds[1] * y_factor),
+        round(bounds[0] * x_factor) + array.shape[2],
+        round(bounds[1] * y_factor) + array.shape[1],
+    )
+    return (adjusted_projection, adjusted_bounds)
+
+
 class RasterFormat:
     """An abstract class for writing raster data.
 

{rslearn-0.0.12 → rslearn-0.0.13/rslearn.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rslearn
-Version: 0.0.12
+Version: 0.0.13
 Summary: A library for developing remote sensing datasets and models
 Author: OlmoEarth Team
 License:                                  Apache License
@@ -214,7 +214,7 @@ License-File: LICENSE
 License-File: NOTICE
 Requires-Dist: boto3>=1.39
 Requires-Dist: fiona>=1.10
-Requires-Dist: fsspec>=2025.9.0
+Requires-Dist: fsspec>=2025.10.0
 Requires-Dist: jsonargparse>=4.35.0
 Requires-Dist: lightning>=2.5.1.post0
 Requires-Dist: Pillow>=11.3

{rslearn-0.0.12 → rslearn-0.0.13}/rslearn.egg-info/SOURCES.txt

@@ -131,6 +131,7 @@ rslearn/train/callbacks/peft.py
 rslearn/train/tasks/__init__.py
 rslearn/train/tasks/classification.py
 rslearn/train/tasks/detection.py
+rslearn/train/tasks/embedding.py
 rslearn/train/tasks/multi_task.py
 rslearn/train/tasks/per_pixel_regression.py
 rslearn/train/tasks/regression.py

{rslearn-0.0.12 → rslearn-0.0.13}/rslearn.egg-info/requires.txt

@@ -1,6 +1,6 @@
 boto3>=1.39
 fiona>=1.10
-fsspec>=2025.9.0
+fsspec>=2025.10.0
 jsonargparse>=4.35.0
 lightning>=2.5.1.post0
 Pillow>=11.3