rslearn 0.0.1__py3-none-any.whl → 0.0.21__py3-none-any.whl

rslearn/tile_stores/file.py

@@ -1,242 +0,0 @@
-"""FileTileStore TileStore implementation."""
-
-import json
-from typing import Any
-
-import numpy.typing as npt
-from upath import UPath
-
-from rslearn.config import RasterFormatConfig, TileStoreConfig, VectorFormatConfig
-from rslearn.utils import Feature, PixelBounds, Projection
-from rslearn.utils.fsspec import open_atomic
-from rslearn.utils.raster_format import (
-    GeotiffRasterFormat,
-    RasterFormat,
-    load_raster_format,
-)
-from rslearn.utils.vector_format import (
-    GeojsonVectorFormat,
-    VectorFormat,
-    load_vector_format,
-)
-
-from .tile_store import LayerMetadata, TileStore, TileStoreLayer
-
-
-class FileTileStoreLayer(TileStoreLayer):
-    """A layer in a FileTileStore."""
-
-    def __init__(
-        self,
-        path: UPath,
-        projection: Projection | None = None,
-        raster_format: RasterFormat = GeotiffRasterFormat(),
-        vector_format: VectorFormat = GeojsonVectorFormat(),
-    ):
-        """Creates a new FileTileStoreLayer.
-
-        The root directory is a subfolder of the FileTileStore's root directory.
-
-        Args:
-            fs: filesystem where this layer is stored
-            path: path on the filesystem corresponding to this layer
-            projection: the projection of this layer
-            raster_format: the RasterFormat to use for reading/writing raster data
-            vector_format: the VectorFormat to use for reading/writing vector data
-        """
-        self.path = path
-        self.raster_format = raster_format
-        self.vector_format = vector_format
-        self.projection = projection
-
-        if not self.projection:
-            self.projection = self.get_metadata().projection
-
-    def read_raster(self, bounds: PixelBounds) -> npt.NDArray[Any] | None:
-        """Read raster data from the store.
-
-        Args:
-            bounds: the bounds within which to read
-
-        Returns:
-            the raster data
-        """
-        return self.raster_format.decode_raster(self.path, bounds)
-
-    def write_raster(self, bounds: PixelBounds, array: npt.NDArray[Any]) -> None:
-        """Write raster data to the store.
-
-        Args:
-            bounds: the bounds of the raster
-            array: the raster data
-        """
-        self.raster_format.encode_raster(self.path, self.projection, bounds, array)
-
-    def get_raster_bounds(self) -> PixelBounds:
-        """Gets the bounds of the raster data in the store."""
-        return self.raster_format.get_raster_bounds(self.path)
-
-    def read_vector(self, bounds: PixelBounds) -> list[Feature]:
-        """Read vector data from the store.
-
-        Args:
-            bounds: the bounds within which to read
-
-        Returns:
-            the vector data
-        """
-        return self.vector_format.decode_vector(self.path, bounds)
-
-    def write_vector(self, data: list[Feature]) -> None:
-        """Save vector tiles to the store.
-
-        Args:
-            data: the vector data
-        """
-        self.vector_format.encode_vector(self.path, self.projection, data)
-
-    def get_metadata(self) -> LayerMetadata:
-        """Get the LayerMetadata associated with this layer."""
-        with (self.path / "metadata.json").open("r") as f:
-            return LayerMetadata.deserialize(json.load(f))
-
-    def set_property(self, key: str, value: Any) -> None:
-        """Set a property in the metadata for this layer.
-
-        Args:
-            key: the property key
-            value: the property value
-        """
-        metadata = self.get_metadata()
-        metadata.properties[key] = value
-        self.save_metadata(metadata)
-
-    def save_metadata(self, metadata: LayerMetadata) -> None:
-        """Save the LayerMetadata associated with this layer."""
-        self.path.mkdir(parents=True, exist_ok=True)
-        metadata_path = self.path / "metadata.json"
-        with open_atomic(metadata_path, "w") as f:
-            json.dump(metadata.serialize(), f)
-
-
-class FileTileStore(TileStore):
-    """A TileStore that stores data on the local filesystem."""
-
-    def __init__(
-        self,
-        path: UPath,
-        raster_format: RasterFormat = GeotiffRasterFormat(),
-        vector_format: VectorFormat = GeojsonVectorFormat(),
-    ):
-        """Initialize a new FileTileStore.
-
-        Args:
-            fs: the filesystem for this tile store.
-            path: the path on the fs containing root of this tile store.
-            raster_format: the RasterFormat (defaults to Geotiff)
-            vector_format: the VectorFormat (defaults to GeoJSON)
-        """
-        self.path = path
-        self.raster_format = raster_format
-        self.vector_format = vector_format
-
-    def _get_layer_dir(self, layer_id: tuple[str, ...]) -> UPath:
-        """Get the path of the specified layer ID."""
-        cur = self.path
-        for part in layer_id:
-            if "/" in part or part.startswith("."):
-                raise ValueError(f"Invalid layer_id part {part}")
-            cur /= part
-        return cur
-
-    def create_layer(
-        self, layer_id: tuple[str, ...], metadata: LayerMetadata
-    ) -> TileStoreLayer:
-        """Create a layer in the tile store (or get matching existing layer).
-
-        Args:
-            layer_id: the id of the layer to create
-            metadata: metadata about the layer
-
-        Returns:
-            a TileStoreLayer corresponding to the new or pre-existing layer
-        """
-        layer_dir = self._get_layer_dir(layer_id)
-        layer = FileTileStoreLayer(
-            layer_dir,
-            projection=metadata.projection,
-            raster_format=self.raster_format,
-            vector_format=self.vector_format,
-        )
-        if not (layer_dir / "metadata.json").exists():
-            layer.save_metadata(metadata)
-        return layer
-
-    def get_layer(self, layer_id: tuple[str, ...]) -> TileStoreLayer | None:
-        """Get a layer in the tile store.
-
-        Args:
-            layer_id: the id of the layer to get
-
-        Returns:
-            the layer, or None if it does not exist yet.
-        """
-        layer_dir = self._get_layer_dir(layer_id)
-        if not (layer_dir / "metadata.json").exists():
-            return None
-        return FileTileStoreLayer(
-            layer_dir,
-            raster_format=self.raster_format,
-            vector_format=self.vector_format,
-        )
-
-    def list_layers(self, prefix: tuple[str, ...] = tuple()) -> list[str]:
-        """List options for next part of layer ID with the specified prefix.
-
-        Args:
-            prefix: the prefix to match
-
-        Returns:
-            available options for next part of the layer ID
-        """
-        layer_dir = self._get_layer_dir(prefix)
-        if not layer_dir.is_dir():
-            return []
-        layer_names = []
-        for p in layer_dir.iterdir():
-            layer_names.append(p.name)
-        return layer_names
-
-    @staticmethod
-    def from_config(config: TileStoreConfig, ds_path: UPath) -> "FileTileStore":
-        """Initialize a FileTileStore from configuration.
-
-        Args:
-            config: the TileStoreConfig.
-            ds_path: the dataset root path.
-
-        Returns:
-            the FileTileStore
-        """
-        d = config.config_dict
-
-        path_str = d.get("path", "tiles")
-        # See if this is relative to ds_path or an absolute path.
-        if "://" in path_str:
-            fs_options = d.get("path_options", {})
-            path = UPath(path_str, **fs_options)
-        else:
-            path = ds_path / path_str
-
-        kwargs = dict(
-            path=path,
-        )
-        if "raster_format" in d:
-            kwargs["raster_format"] = load_raster_format(
-                RasterFormatConfig.from_config(d["raster_format"])
-            )
-        if "vector_format" in d:
-            kwargs["vector_format"] = load_vector_format(
-                VectorFormatConfig.from_config(d["vector_format"])
-            )
-        return FileTileStore(**kwargs)

rslearn/utils/mgrs.py

@@ -1,24 +0,0 @@
-"""Utilities related to MGRS cells."""
-
-from collections.abc import Generator
-
-import numpy as np
-from mgrs import MGRS
-
-
-def for_each_cell(
-    bounds: tuple[float, float, float, float],
-) -> Generator[str, None, None]:
-    """Yields each MGRS cell in the specified WGS84 bounds.
-
-    Args:
-        bounds: (minx, miny, maxx, maxy) in WGS84 coordinates
-    """
-    # Determine MGRS cell by converting to WGS84 and iterating over several
-    # longitude/latitudes.
-    # MGRS is at most 0.9-ish degrees (at equator) so we can iterate in 0.5
-    # degree increments.
-    mgrs_db = MGRS()
-    for lon in np.arange(bounds[0], bounds[2], 0.5):
-        for lat in np.arange(bounds[1], bounds[3], 0.5):
-            yield mgrs_db.toMGRS(lat, lon, MGRSPrecision=0)

rslearn/utils/utils.py

@@ -1,22 +0,0 @@
-"""Miscellaneous utility functions."""
-
-import os
-from contextlib import contextmanager
-from typing import Any
-
-
-@contextmanager
-def open_atomic(filepath: str, *args: list[Any], **kwargs: dict[str, Any]):
-    """Open a file for atomic writing.
-
-    Will write to a temporary file, and rename it to the destination upon success.
-
-    Args:
-        filepath: the file path to be opened
-        *args: ny valid arguments for :code:`open`
-        **kwargs: any valid keyword arguments for :code:`open`
-    """
-    tmppath = filepath + ".tmp." + str(os.getpid())
-    with open(tmppath, *args, **kwargs) as file:
-        yield file
-    os.rename(tmppath, filepath)

rslearn-0.0.1.dist-info/RECORD

@@ -1,88 +0,0 @@
-rslearn/__init__.py,sha256=fFmAen3vxZyosEfPbG0W46IttujYGVxzrGkJ0YutmmY,73
-rslearn/const.py,sha256=qEe6gFMmHPfm1mDRPo8pERicx1g2qgLlgB-GOFvq8uc,681
-rslearn/main.py,sha256=KfxqSw13Mulb4XX7n6B-W8yY3UaTGpC-l2fsQ3FcPPQ,20993
-rslearn/config/__init__.py,sha256=GmNFnNWMI54vx4C_ic3LX_KI7oywd9z0XhCRdOd5PYs,634
-rslearn/config/dataset.py,sha256=ssmhyK3LeCW2YBFJj-ZbkYprrv6Jrs_Bc3lEGnKDiUI,17633
-rslearn/data_sources/__init__.py,sha256=vWBfZnvGhSGHVdJZHshtjRfpgh_Owm51gRtUmpcuA0g,1186
-rslearn/data_sources/aws_landsat.py,sha256=FhU3YgzjVzdgy7hNd9toKGnAtKfQ8cT95t3JdOW7JOg,15576
-rslearn/data_sources/aws_open_data.py,sha256=lNCRwFZsZ7nSb2_tMnmNFQ7ejaZcDBCzPMsBzq8j-wY,30632
-rslearn/data_sources/copernicus.py,sha256=Jq7NAt_95d8GlHGMtn01YhLGqCJZ_4YEcmmttO2NzlI,1178
-rslearn/data_sources/data_source.py,sha256=zn7wSRcsAW9c1OO7iiJnt9_QzqlHaotTzZSmriv-LMQ,3756
-rslearn/data_sources/gcp_public_data.py,sha256=SMrIDYVOFYfcOgVczxXAbsQUHzfICE4WEMuea6KFJj4,20689
-rslearn/data_sources/geotiff.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-rslearn/data_sources/google_earth_engine.py,sha256=rU3s1NeigX7WnHECWaVXteEYtHCPTrGy_CfZqiKPu_Q,11381
-rslearn/data_sources/local_files.py,sha256=67-YRzVS4f-Xru43L5ICadkP4gNB6yuRF4fH6LeNdws,16248
-rslearn/data_sources/openstreetmap.py,sha256=0lCWOXGTIjy-7lorMvqT5IyLbir7lcXdJbETF1G7rYE,19291
-rslearn/data_sources/planet.py,sha256=t2vqKSMqw3ybpJRXRiZD8_4UKYAqWKUZj_ibmuVxWvM,10825
-rslearn/data_sources/raster_source.py,sha256=VvhosEn8CtDV53lGD_P1NTysdnxelEYaYtzYa16fE8A,10807
-rslearn/data_sources/usgs_landsat.py,sha256=QjQxkD8e9xQTkYQ7EKRvJcU4Rpr7LQFe7Bz1gfT3egA,17644
-rslearn/data_sources/utils.py,sha256=CKZqXvLeO5BSgBvDETLhQtmmcWHih1kw4Y0buNB5NOA,4475
-rslearn/data_sources/vector_source.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-rslearn/data_sources/xyz_tiles.py,sha256=4UOqNknYlWB9USuERCZPkLxpyl72jlWav0mXqrldbTU,12197
-rslearn/dataset/__init__.py,sha256=rG7AzBSV-9b66FYqsYKe53-hc_Iwu-iv3i652LCLor0,172
-rslearn/dataset/add_windows.py,sha256=jE81bnM0Vqpl7jQe8RYZpjOYbfE89Q1BRWAnT8CBzJQ,8478
-rslearn/dataset/dataset.py,sha256=LkSEUhtJR_LgGN8nGd21cdx8R2A8mE7L1p28f5iPUuc,3456
-rslearn/dataset/manage.py,sha256=4VU2Vv_-TBj86E7G2qViWzfvOw8ZiHAukIEna76S-oQ,10816
-rslearn/dataset/materialize.py,sha256=10-qBKrbVh4LZ6mttoEIU_6kx7T1piJ1jvyjd3zzR0Q,9376
-rslearn/dataset/remap.py,sha256=HF5Kn5z6dEPbCpSxaBmYVDSzCuBjG2w8csPsk0RRlv4,2088
-rslearn/dataset/window.py,sha256=C6Q0TYambdjdMaiiXsaurRNbVUuqFWi0PiCJeMAFPQ0,6016
-rslearn/models/__init__.py,sha256=_vWoF9d2Slah8-6XhYhdU4SRsy_CNxXjCGQTD2yvu3Q,22
-rslearn/models/faster_rcnn.py,sha256=cCpe88Rh8MQ3S7ETWi-03tkFTOG9ecx0oCaa_c43oGY,8249
-rslearn/models/fpn.py,sha256=BNybJCqW7vABuvpCzU1Ha4lXM4xd4i9wPkcom8VKO9o,1581
-rslearn/models/multitask.py,sha256=BTRKFJEHblwrdSueMV-xv9nF3LZ_Tw4Y2GfzvHsppw8,2136
-rslearn/models/pick_features.py,sha256=uY5raEobeX4PquFbONqkoubVA2RNOnZENqVEw1Tm6vs,867
-rslearn/models/pooling_decoder.py,sha256=wkID0enokrbRsdtlpfnKJIF8N9TowiSN3cHNXxMXXLE,2689
-rslearn/models/registry.py,sha256=0j6jY-rmB_wKckHEUeCYWXtjviDOEr5c7WhIMxxDwBk,90
-rslearn/models/sam2_enc.py,sha256=gNlPokr7eNxO2KvnzDMXNxYM2WRO0YkQPjR4110n6cw,3508
-rslearn/models/satlaspretrain.py,sha256=9kfTRFoAZNnn73fYnXYAvbQKtqhdPrPvIPqMfWEo_Yw,2470
-rslearn/models/simple_time_series.py,sha256=ubxjQ6CvxrXnB6IdTftsquvIkYm-otUwJ55vMKGGJYY,9626
-rslearn/models/singletask.py,sha256=t-6rwPEESxNqbWaS0pHU597sMQuCFUpQkjX-vbYhbmc,1457
-rslearn/models/ssl4eo_s12.py,sha256=mm9RxQv69OKv9wDXak52mCNxAqlD0DCeXF5lcWwCMQQ,3566
-rslearn/models/swin.py,sha256=vCRh0pBtDg5azx7dHRhlEm_kJO6jACdBlkmmFzkRxcE,5517
-rslearn/models/unet.py,sha256=vgJBTv0XG0kKCbE5d6WeXHAzBGa9MuXKtoJpRDdxJX8,4907
-rslearn/tile_stores/__init__.py,sha256=sW4Xz3aiHOEeVoELd_GYl8XdKgNZKNknjZluaX7me_g,826
-rslearn/tile_stores/file.py,sha256=ISyV-UBgArJJwaPNKG_UC8HHixyiwbXcxba8oF9CLds,7905
-rslearn/tile_stores/tile_store.py,sha256=HxnRVyyhnmIZARmR9VNzgv5YCJHuUOP4hxruLm1AUYI,6560
-rslearn/train/__init__.py,sha256=fnJyY4aHs5zQqbDKSfXsJZXY_M9fbTsf7dRYaPwZr2M,30
-rslearn/train/data_module.py,sha256=y7cnq2jH8kHoODIbFg_n2FxBaOk2wVQyErolbr5oZs0,5805
-rslearn/train/dataset.py,sha256=-YAjX1h7xOlC-D7IMToRC6r7kXt2qxekXXp9XbAv4Vc,23538
-rslearn/train/lightning_module.py,sha256=SO0LmZOsTPZdFnDCJrixAkkLs1nNylZsc9p6EYJE7S8,10682
-rslearn/train/prediction_writer.py,sha256=iU4r1vsueXfMIjImeZmqpT6qKVbFTnK-ivS43u32elo,6591
-rslearn/train/callbacks/__init__.py,sha256=VNV0ArZyYMvl3dGK2wl6F046khYJ1dEBlJS6G_SYNm0,47
-rslearn/train/callbacks/freeze_unfreeze.py,sha256=lWzQ3ktRM1lMLpCboDCCr04b68gFx5s2zLHacfGID_Q,3019
-rslearn/train/tasks/__init__.py,sha256=dag1u72x1-me6y0YcOubUo5MYZ0Tjf6-dOir9UeFNMs,75
-rslearn/train/tasks/classification.py,sha256=QBsUrRDUGekpqnbHnVkyLNa-aJA1G1ytNUy3W8XCcao,12944
-rslearn/train/tasks/detection.py,sha256=iluKIMPui2FITLFSPsG-HoLA0RkAlT6MKN4VMmx2Z9c,19492
-rslearn/train/tasks/multi_task.py,sha256=wJJNNpW-abjw4dYy6fi2mVIndqBvkRbxilfU1isaK4E,5373
-rslearn/train/tasks/regression.py,sha256=bXV426Je4RhMUgN_flZm6ClUOCO54aFZUu8ZCe-k8iU,8170
-rslearn/train/tasks/segmentation.py,sha256=pugxC4ZZdUZDBIDsRgduIm6nmeUF7cMc-yn2HdC4D3E,7236
-rslearn/train/tasks/task.py,sha256=MsIbpa5adZml8UkGjvkNBKReiPIIIy7oyWZdW4owmZA,3841
-rslearn/train/transforms/__init__.py,sha256=iqolxzIibL5jx-PRwAmgGt3e6tsfbzJUl7dXKgyKvYI,765
-rslearn/train/transforms/concatenate.py,sha256=i-B5X-TzpawM2bLBByZKhQaw13aEAUhhPB8ep9khZ_M,1414
-rslearn/train/transforms/crop.py,sha256=ggJDp48OAnnQoYFHUY0-2Xwk7TDtMy2CaBzc9prwoI8,4022
-rslearn/train/transforms/flip.py,sha256=HuF9R_8D7kAcn6HzsKhkwb_HCLicdMYhPWcc4iySvJ8,3369
-rslearn/train/transforms/normalize.py,sha256=rvVrQ9OvzqJesM_OVUkarxDolbHwZns48HPvaY8l56k,2367
-rslearn/train/transforms/pad.py,sha256=NYQSLiB2etAOJOrwN3q8DjyZMDM6i1f4zK1dat7NKto,4679
-rslearn/train/transforms/transform.py,sha256=aQC_HLtGqW_RnI5u-pa7j7iqlsEEVaE1JVsBdC7wYJ4,3881
-rslearn/utils/__init__.py,sha256=vY3WUOfmkS_PfNbyPL0zw8qp0LMbNi2ODnG1dddsLSM,667
-rslearn/utils/array.py,sha256=G8F5Yqzzh-YcJ1qqmMfOMkuToWtq0TA1DM43vDdwS9w,2351
-rslearn/utils/feature.py,sha256=kAti0I84kYppzszWpXd9tQafPoBhKYb99Om_hIVTnGo,1643
-rslearn/utils/fsspec.py,sha256=WfSFgf8YRKkfxMWE0FaZfT1yWv3Qu4O8bT_Fzo8KF1Y,2863
-rslearn/utils/geometry.py,sha256=CSGhbtDLx2NIFphzkQMgwd_cH_gcih7rO_qjISn3i_w,8745
-rslearn/utils/get_utm_ups_crs.py,sha256=dOiCXD2naUtuIF95r8YmwaQdmy2LejGxBtK_zXnnNts,3617
-rslearn/utils/grid_index.py,sha256=I50R9gPzEhxR8g56CPBPt0oaNW2XU-jsA_IZxl_CvaU,2270
-rslearn/utils/mgrs.py,sha256=S_FFps6jhstJqV9VpyM_ppRV22Yf1VlKIXfNUvsSB-o,757
-rslearn/utils/mp.py,sha256=GdQCDn7EY63yjX996D9rewxvd9fOZdZcTHIywzUNlUY,1371
-rslearn/utils/raster_format.py,sha256=tKbon1nSuUWaNpMk4W-W0bNSHSV_BcTY1POQLjQ0kPY,20823
-rslearn/utils/rtree_index.py,sha256=TvJE0eaRvhZmwaXGQJbPyV3vekOfIFhsUBmuXvQbVkk,4676
-rslearn/utils/spatial_index.py,sha256=eomJAUgzmjir8j9HZnSgQoJHwN9H0wGTjmJkMkLLfsU,762
-rslearn/utils/sqlite_index.py,sha256=4MqDnMoo48_-zvqa0CXEh7-8q-GKujBHevV4wqqE9sA,2747
-rslearn/utils/time.py,sha256=2ilSLG94_sxLP3y5RSV5L5CG8CoND_dbdzYEHVtN-I8,387
-rslearn/utils/utils.py,sha256=xcrxzAFT3hQMzoXRnhWxHUTneN_EbnSAM0IeoK4DL6I,659
-rslearn/utils/vector_format.py,sha256=al5UyZt_r9SLFRgrmS3PLp056EkJoo0D1pPmp3aL8jY,8514
-rslearn-0.0.1.dist-info/LICENSE,sha256=_99ZWPoLdlUbqZoSC5DF4ihiNwl5rTEmBaq2fACecdg,11352
-rslearn-0.0.1.dist-info/METADATA,sha256=OYg-VNguo0cDZ9TdZWWQ9AwOs2BhO8M7Fn3NnXEaaq0,30640
-rslearn-0.0.1.dist-info/WHEEL,sha256=OVMc5UfuAQiSplgO0_WdW7vXVGAt9Hdd6qtN4HotdyA,91
-rslearn-0.0.1.dist-info/entry_points.txt,sha256=doTBQ57NT7nq-dgYGgTTw6mafcGWb_4PWYtYR4rGm50,46
-rslearn-0.0.1.dist-info/top_level.txt,sha256=XDKo90WBH8P9RQumHxo0giLJsoufT4r9odv-WE6Ahk4,8
-rslearn-0.0.1.dist-info/RECORD,,