rslearn 0.0.24__py3-none-any.whl → 0.0.25__py3-none-any.whl

rslearn/vis/vis_server.py

@@ -0,0 +1,574 @@
+"""Visualization server for rslearn datasets.
+
+This module provides a web server to visualize rslearn datasets using the Dataset/Window APIs.
+"""
+
+import argparse
+import json
+import random
+from pathlib import Path
+from typing import Any
+
+from flask import Flask, Response
+from flask import render_template as flask_render_template
+from PIL import Image
+from upath import UPath
+
+from rslearn.config import LayerType
+from rslearn.dataset import Dataset, Window
+from rslearn.log_utils import get_logger
+
+from .render_raster_label import read_raster_layer, render_raster_label
+from .render_sensor_image import render_sensor_image
+from .render_vector_label import get_vector_label_by_property, render_vector_label_image
+from .utils import array_to_bytes, format_window_info, generate_label_colors
+
+logger = get_logger(__name__)
+
+
+def generate_image_as_bytes(
+    window: Window,
+    layer_name: str,
+    dataset: Dataset,
+    bands: dict[str, list[str]],
+    normalization: dict[str, str],
+    task_type: str,
+    label_colors: dict[str, tuple[int, int, int]] | None,
+    label_colors_dict: dict[str, dict[str, tuple[int, int, int]]] | None = None,
+    group_idx: int = 0,
+    label_layers: list[str] | None = None,
+) -> bytes:
+    """Generate an image for a window/layer combination as PNG bytes.
+
+    Args:
+        window: Window object
+        layer_name: Layer name to visualize
+        dataset: Dataset object
+        bands: Dictionary mapping layer_name -> list of band names
+        normalization: Dictionary mapping layer_name -> normalization method
+        task_type: Task type
+        label_colors: Dictionary mapping label class names to RGB colors
+        label_colors_dict: Dictionary mapping layer_name -> label_colors
+        group_idx: Item group index
+        label_layers: List of layer names that are labels
+
+    Returns:
+        PNG image bytes
+    """
+    label_layers = label_layers or []
+    layer_config = dataset.layers[layer_name]
+
+    # Render raster sensor image
+    if layer_config.type == LayerType.RASTER and layer_name not in label_layers:
+        if layer_name in bands and layer_name in normalization:
+            array = read_raster_layer(
+                window, layer_name, layer_config, bands[layer_name], group_idx=group_idx
+            )
+            image_array = render_sensor_image(array, normalization[layer_name])
+            return array_to_bytes(image_array)
+        else:
+            raise ValueError(
+                f"Bands or normalization not specified for layer {layer_name}"
+            )
+
+    # Render labels
+    elif layer_name in label_layers:
+        layer_label_colors = (
+            label_colors_dict.get(layer_name) if label_colors_dict else label_colors
+        )
+        if not layer_label_colors:
+            raise ValueError(f"No label colors available for layer {layer_name}")
+
+        # Render raster label
+        if layer_config.type == LayerType.RASTER:
+            band_set = layer_config.band_sets[0]
+            label_array = read_raster_layer(
+                window,
+                layer_name,
+                layer_config,
+                [band_set.bands[0]],
+                group_idx=group_idx,
+            )
+            image_array = render_raster_label(
+                label_array, layer_label_colors, layer_config
+            )
+            return array_to_bytes(image_array, resampling=Image.Resampling.NEAREST)
+
+        # Render vector label
+        elif layer_config.type == LayerType.VECTOR:
+            image_array = render_vector_label_image(
+                window,
+                layer_name,
+                layer_config,
+                task_type,
+                layer_label_colors,
+                dataset,
+                label_layers,
+                group_idx,
+                bands,
+                normalization,
+            )
+            # Use NEAREST for segmentation labels, LANCZOS for detection (which overlays on reference image)
+            resampling = (
+                Image.Resampling.NEAREST
+                if task_type == "segmentation"
+                else Image.Resampling.LANCZOS
+            )
+            return array_to_bytes(image_array, resampling=resampling)
+
+    raise ValueError(f"Layer {layer_name} is not a raster sensor image or label layer")
+
+
+# Global state (set during initialization)
+_app_state: dict[str, Any] = {}
+
+
+def prepare_visualization_data(
+    sampled_windows: list[Window],
+    dataset: Dataset,
+    layers: list[str],
+    label_layers: list[str],
+    task_type: str,
+    label_colors_dict: dict[str, dict[str, tuple[int, int, int]]],
+    group_idx: int,
+) -> dict[str, Any]:
+    """Prepare data for visualization template.
+
+    Args:
+        sampled_windows: List of windows to display
+        dataset: Dataset object
+        layers: List of all layers
+        label_layers: List of label layer names
+        task_type: Task type
+        label_colors_dict: Dictionary mapping layer_name -> label_colors
+        group_idx: Item group index
+
+    Returns:
+        Dictionary with template context data
+    """
+    window_data: list[dict[str, Any]] = []
+    for idx, window in enumerate(sampled_windows):
+        time_range, lat, lon = format_window_info(window)
+        maps_link = (
+            f"https://www.google.com/maps?q={lat},{lon}"
+            if lat is not None and lon is not None
+            else None
+        )
+
+        available_layers = set()
+        mask_layers = []
+        label_texts = {}
+
+        for layer_name in layers:
+            if layer_name not in dataset.layers:
+                continue
+            layer_config = dataset.layers[layer_name]
+            try:
+                if (
+                    layer_config.type == LayerType.RASTER
+                    and layer_name not in label_layers
+                ):
+                    if window.is_layer_completed(layer_name, group_idx=group_idx):
+                        available_layers.add(layer_name)
+                elif layer_name in label_layers:
+                    if layer_config.type == LayerType.VECTOR:
+                        if task_type == "classification":
+                            try:
+                                if not window.is_layer_completed(
+                                    layer_name, group_idx=group_idx
+                                ):
+                                    logger.debug(
+                                        f"Layer {layer_name} not marked as completed for window {window.name}, attempting to read anyway"
+                                    )
+                                label_text = get_vector_label_by_property(
+                                    window,
+                                    layer_config,
+                                    layer_name,
+                                    group_idx=group_idx,
+                                )
+                                if label_text is not None:
+                                    label_texts[layer_name] = label_text
+                            except Exception as e:
+                                logger.debug(
+                                    f"Failed to get label text for {layer_name} in window {window.name}: {e}"
+                                )
+                        else:
+                            if window.is_layer_completed(
+                                layer_name, group_idx=group_idx
+                            ):
+                                mask_layers.append(layer_name)
+                    elif layer_config.type == LayerType.RASTER:
+                        if window.is_layer_completed(layer_name, group_idx=group_idx):
+                            mask_layers.append(layer_name)
+            except Exception as e:
+                logger.debug(
+                    f"Error processing layer {layer_name} for window {window.name}: {e}"
+                )
+                continue
+
+        # Format time range for template
+        time_range_formatted = None
+        if time_range:
+            time_range_formatted = (
+                time_range[0].isoformat()[:10],
+                time_range[1].isoformat()[:10],
+            )
+
+        window_data.append(
+            {
+                "idx": idx,
+                "name": window.name,
+                "time_range": time_range,
+                "time_range_formatted": time_range_formatted,
+                "lat": lat,
+                "lon": lon,
+                "maps_link": maps_link,
+                "available_layers": available_layers,
+                "mask_layers": mask_layers,
+                "label_texts": label_texts,
+            }
+        )
+
+    label_colors = None
+    sorted_label_keys = None
+    if label_colors_dict:
+        first_label_layer = list(label_colors_dict.keys())[0]
+        label_colors = label_colors_dict[first_label_layer]
+        sorted_label_keys = sorted(label_colors.keys())
+
+    return {
+        "windows": window_data,
+        "layers": layers,
+        "label_colors": label_colors,
+        "sorted_label_keys": sorted_label_keys,
+        "task_type": task_type,
+    }
+
+
+def create_app(
+    dataset: Dataset,
+    windows: list[Window],
+    layers: list[str],
+    bands: dict[str, list[str]],
+    normalization: dict[str, str],
+    task_type: str,
+    label_colors_dict: dict[str, dict[str, tuple[int, int, int]]],
+    group_idx: int,
+    label_layers: list[str],
+    max_samples: int,
+) -> Flask:
+    """Create and configure Flask app.
+
+    Args:
+        dataset: Dataset object
+        windows: List of all windows
+        layers: List of all layers
+        bands: Dictionary mapping layer_name -> list of band names
+        normalization: Dictionary mapping layer_name -> normalization method
+        task_type: Task type
+        label_colors_dict: Dictionary mapping layer_name -> label_colors
+        group_idx: Item group index
+        label_layers: List of label layer names
+        max_samples: Maximum number of windows to sample
+
+    Returns:
+        Configured Flask app
+    """
+    # Set template folder explicitly to ensure Flask can find templates
+    template_folder = Path(__file__).parent / "templates"
+    app = Flask(__name__, template_folder=str(template_folder))
+
+    @app.route("/")
+    def index() -> str:
+        """Render the main visualization page."""
+        if len(windows) > max_samples:
+            sampled_windows = random.sample(windows, max_samples)
+        else:
+            sampled_windows = windows
+
+        template_data = prepare_visualization_data(
+            sampled_windows,
+            dataset,
+            layers,
+            label_layers,
+            task_type,
+            label_colors_dict,
+            group_idx,
+        )
+        return flask_render_template("visualization.html", **template_data)
+
+    @app.route("/images/<int:window_idx>/<layer_name>")
+    def get_image(window_idx: int, layer_name: str) -> Response:
+        """Generate and serve an image for a specific window/layer.
+
+        Args:
+            window_idx: Index of the window in the windows list
+            layer_name: Name of the layer to visualize
+
+        Returns:
+            PNG image response or error response
+        """
+        if window_idx < 0 or window_idx >= len(windows):
+            return Response(
+                "Window index out of range", status=404, mimetype="text/plain"
+            )
+
+        window = windows[window_idx]
+
+        layer_label_colors = None
+        if label_colors_dict and layer_name in label_colors_dict:
+            layer_label_colors = label_colors_dict[layer_name]
+        elif label_colors_dict:
+            first_label_layer = list(label_colors_dict.keys())[0]
+            layer_label_colors = label_colors_dict[first_label_layer]
+
+        image_bytes = generate_image_as_bytes(
+            window,
+            layer_name,
+            dataset,
+            bands,
+            normalization,
+            task_type,
+            layer_label_colors,
+            label_colors_dict,
+            group_idx,
+            label_layers,
+        )
+
+        return Response(
+            image_bytes,
+            mimetype="image/png",
+            headers={"Content-Length": str(len(image_bytes))},
+        )
+
+    return app
+
+
+def run(
+    dataset_path: str | Path | UPath,
+    layers: list[str] | None = None,
+    bands: dict[str, list[str]] | None = None,
+    normalization: dict[str, str] | None = None,
+    task_type: str | None = None,
+    max_samples: int = 100,
+    port: int = 8000,
+    host: str = "0.0.0.0",
+    group_idx: int = 0,
+    label_layers: list[str] | None = None,
+) -> None:
+    """Run the visualization server.
+
+    Args:
+        dataset_path: Path to dataset directory (containing config.json)
+        layers: List of layer names to visualize
+        bands: Dictionary mapping layer_name -> list of band names
+        normalization: Dictionary mapping layer_name -> normalization method
+        task_type: Task type - "classification", "regression", "detection", or "segmentation"
+        max_samples: Maximum number of windows to sample
+        port: Port to serve on
+        host: Host to bind to
+        group_idx: Item group index (default 0)
+        label_layers: List of layer names that are labels
+    """
+    dataset_path = UPath(dataset_path)
+    dataset = Dataset(dataset_path)
+
+    label_layers = label_layers or []
+
+    if layers is None:
+        raise ValueError("--layers is required")
+    all_layers = list(set(layers + label_layers))
+    raster_image_layers = [name for name in all_layers if name not in label_layers]
+    label_layers_in_list = [name for name in all_layers if name in label_layers]
+
+    bands = bands or {}
+    normalization = normalization or {}
+    for layer_name in raster_image_layers:
+        if layer_name not in bands:
+            raise ValueError(
+                f"Bands not specified for layer {layer_name}. Please provide --bands {layer_name}:band1,band2,band3"
+            )
+        if layer_name not in normalization:
+            raise ValueError(
+                f"Normalization not specified for layer {layer_name}. Please provide --normalization {layer_name}:method"
+            )
+
+    label_colors_dict = {}
+    if task_type in ("segmentation", "detection"):
+        for label_layer_name in label_layers_in_list:
+            label_config = dataset.layers[label_layer_name]
+            if not label_config.class_names:
+                raise ValueError(
+                    f"class_names must be specified in the config for label layer '{label_layer_name}'. "
+                    "Auto-detection of class names is not supported."
+                )
+            else:
+                label_classes = set(label_config.class_names)
+                label_colors = generate_label_colors(label_classes)
+                label_colors_dict[label_layer_name] = label_colors
+                logger.info(
+                    f"Found {len(label_classes)} label classes for {label_layer_name}: {sorted(label_classes)}"
+                )
+
+    logger.info(f"Loading all windows from dataset {dataset_path}")
+    windows = dataset.load_windows()
+    logger.info(f"Loaded {len(windows)} windows from dataset")
+    logger.info(f"Layers: {all_layers}")
+    logger.info(f"Bands: {bands}")
+    logger.info(f"Normalization: {normalization}")
+    logger.info(f"Task type: {task_type}")
+
+    if task_type is None:
+        raise ValueError("--task_type is required")
+
+    app = create_app(
+        dataset,
+        windows,
+        all_layers,
+        bands,
+        normalization,
+        task_type,
+        label_colors_dict,
+        group_idx,
+        label_layers,
+        max_samples,
+    )
+
+    logger.info(f"Serving on http://{host}:{port}")
+    logger.info(f"Open http://localhost:{port} in your browser")
+    logger.info(
+        f"Loaded {len(windows)} windows - refreshing the page will show a different random sample"
+    )
+
+    app.run(host=host, port=port, debug=False)
+
+
+def parse_bands_arg(bands_str: str | None) -> dict[str, list[str]]:
+    """Parse --bands argument as JSON.
+
+    Args:
+        bands_str: JSON string mapping layer_name -> list of band names, e.g. '{"sentinel2": ["B04", "B03", "B02"]}'
+
+    Returns:
+        Dictionary mapping layer_name -> list of band names
+    """
+    if not bands_str:
+        return {}
+    try:
+        bands_dict = json.loads(bands_str)
+        if not isinstance(bands_dict, dict):
+            raise ValueError("Bands must be a JSON object/dictionary")
+        for layer_name, band_list in bands_dict.items():
+            if not isinstance(band_list, list):
+                raise ValueError(f"Bands for layer '{layer_name}' must be a list")
+            if not all(isinstance(band, str) for band in band_list):
+                raise ValueError(f"All bands for layer '{layer_name}' must be strings")
+        return bands_dict
+    except json.JSONDecodeError as e:
+        raise ValueError(f"Invalid JSON for bands: {e}") from e
+
+
+def parse_normalization_arg(norm_str: str | None) -> dict[str, str]:
+    """Parse --normalization argument as JSON.
+
+    Args:
+        norm_str: JSON string mapping layer_name -> normalization method, e.g. '{"sentinel2": "sentinel2_rgb"}'
+
+    Returns:
+        Dictionary mapping layer_name -> normalization method
+    """
+    if not norm_str:
+        return {}
+    try:
+        norm_dict = json.loads(norm_str)
+        if not isinstance(norm_dict, dict):
+            raise ValueError("Normalization must be a JSON object/dictionary")
+        for layer_name, method in norm_dict.items():
+            if not isinstance(method, str):
+                raise ValueError(
+                    f"Normalization method for layer '{layer_name}' must be a string"
+                )
+        return norm_dict
+    except json.JSONDecodeError as e:
+        raise ValueError(f"Invalid JSON for normalization: {e}") from e
+
+
+def main() -> None:
+    """Main entry point for the visualization server CLI."""
+    parser = argparse.ArgumentParser(
+        description="Visualize rslearn dataset in a web browser"
+    )
+    parser.add_argument(
+        "dataset_path",
+        type=str,
+        help="Path to dataset directory (containing config.json)",
+    )
+    parser.add_argument(
+        "--layers",
+        type=str,
+        required=True,
+        nargs="+",
+        help="List of layer names to visualize (default: all raster layers)",
+    )
+    parser.add_argument(
+        "--label_layers",
+        type=str,
+        nargs="+",
+        help="List of layer names that are labels (same format as --layers)",
+    )
+    parser.add_argument(
+        "--bands",
+        type=str,
+        required=True,
+        help='Bands to visualize per layer as JSON. Example: \'{"sentinel2": ["B04", "B03", "B02"]}\'',
+    )
+    parser.add_argument(
+        "--normalization",
+        type=str,
+        required=True,
+        help='Normalization method per layer as JSON. Example: \'{"sentinel2": "sentinel2_rgb"}\'',
+    )
+    parser.add_argument(
+        "--task_type",
+        type=str,
+        required=True,
+        choices=["classification", "regression", "detection", "segmentation"],
+        help="Task type (default: auto-detect, but user should specify)",
+    )
+    parser.add_argument(
+        "--max_samples",
+        type=int,
+        default=100,
+        help="Maximum number of windows to sample",
+    )
+    parser.add_argument(
+        "--port", type=int, default=8000, help="Port to serve on (default: 8000)"
+    )
+    parser.add_argument(
+        "--host", type=str, default="0.0.0.0", help="Host to bind to (default: 0.0.0.0)"
+    )
+    parser.add_argument(
+        "--group_idx", type=int, default=0, help="Item group index (default: 0)"
+    )
+
+    args = parser.parse_args()
+
+    bands_dict = parse_bands_arg(args.bands)
+    normalization_dict = parse_normalization_arg(args.normalization)
+
+    run(
+        dataset_path=args.dataset_path,
+        layers=args.layers,
+        bands=bands_dict if bands_dict else None,
+        normalization=normalization_dict if normalization_dict else None,
+        task_type=args.task_type,
+        max_samples=args.max_samples,
+        port=args.port,
+        host=args.host,
+        group_idx=args.group_idx,
+        label_layers=args.label_layers,
+    )
+
+
+if __name__ == "__main__":
+    main()

{rslearn-0.0.24.dist-info → rslearn-0.0.25.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rslearn
-Version: 0.0.24
+Version: 0.0.25
 Summary: A library for developing remote sensing datasets and models
 Author: OlmoEarth Team
 License:                                  Apache License
@@ -214,6 +214,7 @@ License-File: LICENSE
 License-File: NOTICE
 Requires-Dist: boto3>=1.39
 Requires-Dist: fiona>=1.10
+Requires-Dist: flask>=3.0.0
 Requires-Dist: fsspec>=2025.10.0
 Requires-Dist: jsonargparse>=4.35.0
 Requires-Dist: lightning>=2.5.1.post0
@@ -482,6 +483,18 @@ We can visualize both the GeoTIFFs together in qgis:
 qgis $DATASET_PATH/windows/default/seattle_54912_-527360/layers/*/*/geotiff.tif
 ```
 
+We can also visualize samples using the visualization module:
+```
+python -m rslearn.vis.vis_server \
+    $DATASET_PATH \
+    --layers sentinel2 \  # image modality layers
+    --label_layers label_raster \  # layer layers
+    --bands '{"sentinel2": ["B04", "B03", "B02"]}' \  # specify bands wanted for each image modality
+    --normalization '{"sentinel2": "sentinel2_rgb"}'  \  # specify normalization wanted for each image modality
+    --task_type segmentation \  # segmentation, detection, or classification
+    --max_samples 100 \  # number of datapoints to randomly sample and visualize
+    --port 8000
+```
 
 ### Training a Model
 

{rslearn-0.0.24.dist-info → rslearn-0.0.25.dist-info}/RECORD

@@ -115,7 +115,7 @@ rslearn/tile_stores/tile_store.py,sha256=9AeYduDYPp_Ia2NMlq6osptpz_AFGIOQcLJrqZ_
 rslearn/train/__init__.py,sha256=fnJyY4aHs5zQqbDKSfXsJZXY_M9fbTsf7dRYaPwZr2M,30
 rslearn/train/all_patches_dataset.py,sha256=EVoYCmS3g4OfWPt5CZzwHVx9isbnWh5HIGA0RBqPFeA,21145
 rslearn/train/data_module.py,sha256=pgut8rEWHIieZ7RR8dUvhtlNqk0egEdznYF3tCvqdHg,23552
-rslearn/train/dataset.py,sha256=Jy1jU3GigfHaFeX9rbveX9bqy2Pd5Wh_vquD6_aFnS8,36522
+rslearn/train/dataset.py,sha256=gMOUmhaKc9whTFJLCeTQC0J6F-pkphEUVR6Cp42vM-0,38659
 rslearn/train/lightning_module.py,sha256=V4YoEg9PrwrgG4q9Dmv_9OBrSIK-SRPzjWtZRIfmPFg,15366
 rslearn/train/model_context.py,sha256=6o66BY6okBK-D5e0JUwPd7fxD_XehVaqxdQkJJKmQ3E,2580
 rslearn/train/optimizer.py,sha256=EKSqkmERalDA0bF32Gey7n6z69KLyaUWKlRsGJfKBmE,927
@@ -128,12 +128,12 @@ rslearn/train/callbacks/gradients.py,sha256=4YqCf0tBb6E5FnyFYbveXfQFlgNPyxIXb2FC
 rslearn/train/callbacks/peft.py,sha256=wEOKsS3RhsRaZTXn_Kz2wdsZdIiIaZPdCJWtdJBurT8,4156
 rslearn/train/tasks/__init__.py,sha256=dag1u72x1-me6y0YcOubUo5MYZ0Tjf6-dOir9UeFNMs,75
 rslearn/train/tasks/classification.py,sha256=72ZBcbunMsdPYQN53S-4GfiLIDrr1X3Hni07dBJ0pu0,14261
-rslearn/train/tasks/detection.py,sha256=B0tfB7UGIbRtjnye3PhzLmfeQ4X7ImO3A-_LeNhBA54,21988
+rslearn/train/tasks/detection.py,sha256=8nxa90ht0-xlcO_gsI-zR02nL82-wJXq4LpXNFvg5Lw,21730
 rslearn/train/tasks/embedding.py,sha256=NdJEAaDWlWYzvOBVf7eIHfFOzqTgavfFH1J1gMbAMVo,3891
 rslearn/train/tasks/multi_task.py,sha256=32hvwyVsHqt7N_M3zXsTErK1K7-0-BPHzt7iGNehyaI,6314
 rslearn/train/tasks/per_pixel_regression.py,sha256=Clrod6LQGjgNC0IAR4HLY7eCGWMHj2mk4d4moZCl4Qc,10209
 rslearn/train/tasks/regression.py,sha256=bVS_ApZSpbL0NaaM8Mu5Bsu4SBUyLpVtrPslulvvZHs,12695
-rslearn/train/tasks/segmentation.py,sha256=LZeuveHhMQsjNOQfMcwqSI4Ux3k9zfa58A2eZHSif8Y,29391
+rslearn/train/tasks/segmentation.py,sha256=ZAAal6VShK4OPd-zArDmKdFH58JijMA6aGx_kp3eHPs,29094
 rslearn/train/tasks/task.py,sha256=nMPunl9OlnOimr48saeTnwKMQ7Du4syGrwNKVQq4FL4,4110
 rslearn/train/transforms/__init__.py,sha256=BkCAzm4f-8TEhPIuyvCj7eJGh36aMkZFYlq-H_jkSvY,778
 rslearn/train/transforms/concatenate.py,sha256=hVVBaxIdk1Cx8JHPirj54TGpbWAJx5y_xD7k1rmGmT0,3166
@@ -148,6 +148,7 @@ rslearn/train/transforms/sentinel1.py,sha256=PMMFo4RysMpZS7PBZ3yHGzJ5JADPZAOZpCb
 rslearn/train/transforms/transform.py,sha256=n1Qzqix2dVvej-Q7iPzHeOQbqH79IBlvqPoymxhNVpE,4446
 rslearn/utils/__init__.py,sha256=GNvdTUmXakiEMnLdje7k1fe5aC7SFVqP757kbpN6Fzw,558
 rslearn/utils/array.py,sha256=RC7ygtPnQwU6Lb9kwORvNxatJcaJ76JPsykQvndAfes,2444
+rslearn/utils/colors.py,sha256=ELY9_buH06TOVPLrDAyf2S0G--ZiOxnnP8Ujim6_3ig,369
 rslearn/utils/feature.py,sha256=lsg0WThZDJzo1mrbaL04dXYI5G3x-n5FG9aEjj7uUaI,1649
 rslearn/utils/fsspec.py,sha256=h3fER_bkewzR9liEAULXguTIvXLUXA17pC_yZoWN5Tk,5902
 rslearn/utils/geometry.py,sha256=S60uEvCQ3SyR2-3TJ1yUucFmGRM_rcoIT_gH83PN3RE,21485
@@ -162,10 +163,17 @@ rslearn/utils/sqlite_index.py,sha256=YGOJi66544e6JNtfSft6YIlHklFdSJO2duxQ4TJ2iu4
 rslearn/utils/stac.py,sha256=c8NwOCKWvUwA-FSKlxZn-t7RZYweuye53OufT0bAK4A,5996
 rslearn/utils/time.py,sha256=2ilSLG94_sxLP3y5RSV5L5CG8CoND_dbdzYEHVtN-I8,387
 rslearn/utils/vector_format.py,sha256=4ZDYpfBLLxguJkiIaavTagiQK2Sv4Rz9NumbHlq-3Lw,15041
-rslearn-0.0.24.dist-info/licenses/LICENSE,sha256=_99ZWPoLdlUbqZoSC5DF4ihiNwl5rTEmBaq2fACecdg,11352
-rslearn-0.0.24.dist-info/licenses/NOTICE,sha256=wLPr6rwV_jCg-xEknNGwhnkfRfuoOE9MZ-lru2yZyLI,5070
-rslearn-0.0.24.dist-info/METADATA,sha256=gV5mgeYPYiKWrEu7D8acOubWvg76Nn_4ICvlD7iTpcs,37936
-rslearn-0.0.24.dist-info/WHEEL,sha256=qELbo2s1Yzl39ZmrAibXA2jjPLUYfnVhUNTlyF1rq0Y,92
-rslearn-0.0.24.dist-info/entry_points.txt,sha256=doTBQ57NT7nq-dgYGgTTw6mafcGWb_4PWYtYR4rGm50,46
-rslearn-0.0.24.dist-info/top_level.txt,sha256=XDKo90WBH8P9RQumHxo0giLJsoufT4r9odv-WE6Ahk4,8
-rslearn-0.0.24.dist-info/RECORD,,
+rslearn/vis/__init__.py,sha256=1tqz1zuWorous4UqFB_x2kvzIbywX-qqHFYA37tsDFc,49
+rslearn/vis/normalization.py,sha256=Dv_H_XOUwF_LEQHzIhBDHYn9aBTjoPWoPtXBnKSCowQ,3678
+rslearn/vis/render_raster_label.py,sha256=3ZN6o0taymENKeBCkB9BtesmdEOi2_H2ziQ79BbuDI4,3104
+rslearn/vis/render_sensor_image.py,sha256=D0ynK6ABPV046970lIKwF98klpSCtrsUvZTwtZaNQeI,767
+rslearn/vis/render_vector_label.py,sha256=ncwgRKCYCJCK1-wTpjgksOiDDebku37LpAyq6wsg4jg,14939
+rslearn/vis/utils.py,sha256=Zop3dEmyaXUYhPiGdYzrTO8BRXWscP2dEZy2myQUnNk,2765
+rslearn/vis/vis_server.py,sha256=kIGnhTy-yfu5lBOVCoo8VVG259i974JPszudCePbzfI,20157
+rslearn-0.0.25.dist-info/licenses/LICENSE,sha256=_99ZWPoLdlUbqZoSC5DF4ihiNwl5rTEmBaq2fACecdg,11352
+rslearn-0.0.25.dist-info/licenses/NOTICE,sha256=wLPr6rwV_jCg-xEknNGwhnkfRfuoOE9MZ-lru2yZyLI,5070
+rslearn-0.0.25.dist-info/METADATA,sha256=L_Ilp6WKln27vT4ja6ky8qft_1pkm2aql-uXOtghpTo,38576
+rslearn-0.0.25.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
+rslearn-0.0.25.dist-info/entry_points.txt,sha256=doTBQ57NT7nq-dgYGgTTw6mafcGWb_4PWYtYR4rGm50,46
+rslearn-0.0.25.dist-info/top_level.txt,sha256=XDKo90WBH8P9RQumHxo0giLJsoufT4r9odv-WE6Ahk4,8
+rslearn-0.0.25.dist-info/RECORD,,

{rslearn-0.0.24.dist-info → rslearn-0.0.25.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (80.10.1)
+Generator: setuptools (80.10.2)
 Root-Is-Purelib: true
 Tag: py3-none-any