wip

difference.ipynb

@@ -10,39 +10,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
    "id": "e84709ab-1b47-49ee-8cbd-8aa69744b6c3",
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "a4103c257b7b4691a0a86d817390e9ca",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "FloatProgress(value=0.0, layout=Layout(width='auto'), style=ProgressStyle(bar_color='black'))"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "31245f0a73634185b297464775ba4c13",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "FloatProgress(value=0.0, layout=Layout(width='auto'), style=ProgressStyle(bar_color='black'))"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "outputs": [],
    "source": [
     "import ibis\n",
     "from ibis import _\n",
@@ -54,11 +25,10 @@
     "    .rename(geom = \"SHAPE\", gid = \"OBJECTID\")\n",
     "   # .filter(_.UNIT_NAME == \"Angeles National Forest\")\n",
     "    .filter(_.reGAP < 3) \n",
-    "    #.mutate(geom = _.geom.convert(\"epsg:3310\",\"epsg:4326\"))\n",
     ")\n",
-    "\n",
-    "conn.create_table(\"t1\", tbl.filter(_.Release_Year == 2024))\n",
-    "conn.create_table(\"t2\", tbl.filter(_.Release_Year == 2023))\n",
+    "lvl = \"State\"\n",
+    "conn.create_table(\"t1\", tbl.filter(_.Release_Year == 2024, _.MNG_AG_LEV == lvl), overwrite = True)\n",
+    "conn.create_table(\"t2\", tbl.filter(_.Release_Year == 2023, _.MNG_AG_LEV == lvl), overwrite = True)\n",
     "\n",
     "conn.disconnect()"
    ]
@@ -68,7 +38,22 @@
    "execution_count": null,
    "id": "121c3cad-680c-4f3a-9075-638711ea1634",
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ba5fb88fdfbc4c43a4df911a9b958cfd",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "FloatProgress(value=0.0, layout=Layout(width='auto'), style=ProgressStyle(bar_color='black'))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
    "source": [
     "%%time\n",
     "import duckdb\n",

preprocess.ipynb

@@ -5,7 +5,7 @@
    "id": "4b4adc2a-bf0c-4ace-87be-dbaf90be0125",
    "metadata": {},
    "source": [
-    "# Preporcessing"
+    "# Pre-processing"
    ]
   },
   {
@@ -98,7 +98,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 1,
    "id": "55afe07c-8681-4308-bbb9-e460f7380f86",
    "metadata": {},
    "outputs": [],
@@ -111,28 +111,6 @@
     "ca2024 = conn.read_parquet(\"ca2024.parquet\")"
    ]
   },
-  {
-   "cell_type": "code",
-   "execution_count": 5,
-   "id": "9d4cd1c4-288b-4d1c-907c-ca76ccbdb1d6",
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "geopandas.geodataframe.GeoDataFrame"
-      ]
-     },
-     "execution_count": 5,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "gdf = ca2024.execute()\n",
-    "gdf.__class__"
-   ]
-  },
   {
    "cell_type": "markdown",
    "id": "ebbb2650-4442-4e54-8467-5e681d6fab9e",