Updated code to use stacchip 0.1.33, simplifying bbox calculations

nbs/v1-inference-simsearch-naip-stacchip.ipynb

@@ -46,7 +46,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "%pip install stacchip==0.1.31"
+    "%pip install stacchip==0.1.33"
    ]
   },
   {
@@ -68,12 +68,10 @@
     "import numpy as np\n",
     "import pandas as pd\n",
     "import pystac_client\n",
-    "import rasterio\n",
     "import shapely\n",
     "import torch\n",
     "import yaml\n",
     "from box import Box\n",
-    "from pyproj import Transformer\n",
     "from stacchip.chipper import Chipper\n",
     "from stacchip.indexer import NoStatsChipIndexer\n",
     "from stacchip.processors.prechip import normalize_timestamp\n",
@@ -136,48 +134,14 @@
     "random.shuffle(items_list)"
    ]
   },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "ef3daea0-4e97-426a-ba9e-9f05ef0a37e5",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def get_bounds_centroid(url: str):\n",
-    "    \"\"\"\n",
-    "    Retrieve the bounds and centroid of an image from its URL.\n",
-    "\n",
-    "    Parameters:\n",
-    "    url (str): The URL of the image.\n",
-    "\n",
-    "    Returns:\n",
-    "    tuple: Bounds coordinates and centroid coordinates.\n",
-    "    \"\"\"\n",
-    "    with rasterio.open(url) as rst:\n",
-    "        bounds = rst.bounds\n",
-    "        transformer = Transformer.from_crs(rst.crs, 4326)\n",
-    "\n",
-    "        centroid_x = (bounds.left + bounds.right) / 2\n",
-    "        centroid_y = (bounds.top + bounds.bottom) / 2\n",
-    "\n",
-    "        centroid_x, centroid_y = transformer.transform(centroid_x, centroid_y)\n",
-    "\n",
-    "        bounds_b, bounds_l = transformer.transform(bounds.left, bounds.bottom)\n",
-    "        bounds_t, bounds_r = transformer.transform(bounds.right, bounds.top)\n",
-    "\n",
-    "        return [bounds_b, bounds_l, bounds_t, bounds_r], [centroid_x, centroid_y]"
-   ]
-  },
   {
    "cell_type": "code",
    "execution_count": null,
    "id": "ccb089e6-36d7-41a4-8297-f5517aa42065",
    "metadata": {},
    "outputs": [],
    "source": [
-    "chips = []  # List to hold chip metadata\n",
     "chip_images = []  # List to hold chip pixels\n",
-    "chip_centroids = []  # List to hold chip centroids\n",
     "chip_bounds = []  # List to hold chip bounds"
    ]
   },
@@ -194,10 +158,6 @@
     "    # Index the chips in the item\n",
     "    indexer = NoStatsChipIndexer(item)\n",
     "\n",
-    "    # Obtain the item bounds and centroid\n",
-    "    bounds, centroid = get_bounds_centroid(item.assets[\"image\"].href)\n",
-    "    print(f\"Bounds coordinates: {bounds}, centroid coordinates: {centroid}\")\n",
-    "\n",
     "    # Instantiate the chipper\n",
     "    chipper = Chipper(\n",
     "        indexer, asset_blacklist=[\"thumbnail\", \"tilejson\", \"rendered_preview\"]\n",
@@ -206,10 +166,9 @@
     "    # Get 5 randomly sampled chips from the total\n",
     "    # number of chips within this item's entire image\n",
     "    for chip_id in random.sample(range(0, len(chipper)), 25):\n",
-    "        chips.append(chipper[chip_id])\n",
-    "        chip_images.append(chipper[chip_id][\"image\"])\n",
-    "        chip_bounds.append(bounds)\n",
-    "        chip_centroids.append(centroid)"
+    "        x_index, y_index, chip = chipper[chip_id]\n",
+    "        chip_images.append(chip[\"image\"])\n",
+    "        chip_bounds.append(indexer.get_chip_bbox(x_index, y_index))"
    ]
   },
   {
@@ -464,12 +423,12 @@
    "source": [
     "embeddings = []\n",
     "i = 0\n",
-    "for tile, centroid, box in zip(chip_images, chip_centroids, chip_bounds):\n",
-    "    lon, lat = centroid[0], centroid[1]\n",
-    "\n",
+    "for tile, box in zip(chip_images, chip_bounds):\n",
     "    date = datetime.datetime.strptime(f\"{YEAR}-06-01\", \"%Y-%m-%d\")\n",
     "    gsd = 0.6\n",
     "\n",
+    "    lon, lat = chip_bounds[0].centroid.coords[0]\n",
+    "\n",
     "    datacube = prep_datacube(\n",
     "        np.array(tile), lat, lon, pd.to_datetime(f\"{YEAR}-06-01\"), gsd, model.device\n",
     "    )\n",
@@ -482,12 +441,9 @@
     "        \"embeddings\": [np.ascontiguousarray(embeddings_.squeeze())],\n",
     "        \"image\": [np.ascontiguousarray(np.array(tile.transpose(1, 2, 0)).flatten())],\n",
     "    }\n",
-    "    # Define the bounding box as a Polygon (xmin, ymin, xmax, ymax)\n",
-    "    box_emb = shapely.geometry.box(box[0], box[1], box[2], box[3])\n",
-    "    print(box_emb)\n",
     "\n",
     "    # Create the GeoDataFrame\n",
-    "    gdf = gpd.GeoDataFrame(data, geometry=[box_emb], crs=\"EPSG:4326\")\n",
+    "    gdf = gpd.GeoDataFrame(data, geometry=[box], crs=\"EPSG:4326\")\n",
     "\n",
     "    outpath = f\"{outdir_embeddings}/{i}.gpq\"\n",
     "    gdf.to_parquet(path=outpath, compression=\"ZSTD\", schema_version=\"1.0.0\")\n",