|
| 1 | +{ |
| 2 | + "cells": [ |
| 3 | + { |
| 4 | + "cell_type": "markdown", |
| 5 | + "id": "a986e1d5", |
| 6 | + "metadata": {}, |
| 7 | + "source": [ |
| 8 | + "## Creating the SQLite Databases Necessary for DES Sorcha\n", |
| 9 | + "\n", |
| 10 | + "This notebook creates the SQLite databases needed to run **DES** in Sorcha. This code obtains the nescessory data from https://github.com/bernardinelli/DESTNOSIM/tree/master/data and converts it to sqlite databases. It creates two files:\n", |
| 11 | + "\n", |
| 12 | + "1. **Visits database** from `y6a1c.ccdcorners.fits.gz`\n", |
| 13 | + "2. **Pointing database** from `y6a1c.exposures.positions.fits`\n" |
| 14 | + ] |
| 15 | + }, |
| 16 | + { |
| 17 | + "cell_type": "code", |
| 18 | + "execution_count": null, |
| 19 | + "id": "70dc0f7f", |
| 20 | + "metadata": {}, |
| 21 | + "outputs": [], |
| 22 | + "source": [ |
| 23 | + "import numpy as np\n", |
| 24 | + "from astropy.io import fits\n", |
| 25 | + "from sqlalchemy import create_engine\n", |
| 26 | + "import sqlite3\n", |
| 27 | + "from astropy.time import Time\n", |
| 28 | + "import pandas as pd\n", |
| 29 | + "import requests\n", |
| 30 | + "from io import BytesIO\n", |
| 31 | + "import gzip\n", |
| 32 | + "\n", |
| 33 | + "\n", |
| 34 | + "\n", |
| 35 | + "def process_pointings_to_sqlite(\n", |
| 36 | + " use_url,\n", |
| 37 | + " fits_path=\"y6a1c.exposures.positions.fits\",\n", |
| 38 | + " db_path=\"DES_TNO.db\",\n", |
| 39 | + " url=\"https://github.com/bernardinelli/DESTNOSIM/raw/refs/heads/master/data/y6a1c.exposures.positions.fits\"\n", |
| 40 | + "):\n", |
| 41 | + " \"\"\"\n", |
| 42 | + " Processes a FITS file containing DES pointing data and stores \n", |
| 43 | + " data into a SQLite database with indexing.\n", |
| 44 | + "\n", |
| 45 | + " Parameters\n", |
| 46 | + " ----------\n", |
| 47 | + " use_url : bool\n", |
| 48 | + " If True, the FITS file will be downloaded from the specified `url`. If False, the local file path\n", |
| 49 | + " specified by `fits_path` will be used.\n", |
| 50 | + "\n", |
| 51 | + " fits_path : str, optional\n", |
| 52 | + " Path to the local FITS file (default is \"y6a1c.exposures.positions.fits\"). Only used if `use_url` is False.\n", |
| 53 | + "\n", |
| 54 | + " db_path : str, optional\n", |
| 55 | + " Path where the SQLite database will be saved (default is \"DES_TNO.db\").\n", |
| 56 | + "\n", |
| 57 | + " url : str, optional\n", |
| 58 | + " Direct URL to the raw FITS file hosted on GitHub.\n", |
| 59 | + "\n", |
| 60 | + " \n", |
| 61 | + " \"\"\"\n", |
| 62 | + " if use_url == True:\n", |
| 63 | + " response = requests.get(url)\n", |
| 64 | + " response.raise_for_status()\n", |
| 65 | + " with fits.open(BytesIO(response.content)) as hdul:\n", |
| 66 | + " df = hdul[1].data\n", |
| 67 | + " else:\n", |
| 68 | + " with fits.open(fits_path) as HDUl:\n", |
| 69 | + " df = HDUl[1].data\n", |
| 70 | + "\n", |
| 71 | + " # Define column names\n", |
| 72 | + " cov_1 = \"cov_xx\"\n", |
| 73 | + " cov_2 = \"cov_yy\"\n", |
| 74 | + " cov_3 = \"cov_xy\"\n", |
| 75 | + " observatory_1 = \"observatory_1\"\n", |
| 76 | + " observatory_2 = \"observatory_2\"\n", |
| 77 | + " observatory_3 = \"observatory_3\"\n", |
| 78 | + " velocity_1 = \"velocity_1\"\n", |
| 79 | + " velocity_2 = \"velocity_2\"\n", |
| 80 | + " velocity_3 = \"velocity_3\"\n", |
| 81 | + "\n", |
| 82 | + " # Create dictionary for DataFrame construction\n", |
| 83 | + " temp = {\n", |
| 84 | + " cov_1: np.array(df[\"cov\"][:, 0], dtype=\"<f8\"),\n", |
| 85 | + " cov_2: np.array(df[\"cov\"][:, 1], dtype=\"<f8\"),\n", |
| 86 | + " cov_3: np.array(df[\"cov\"][:, 2], dtype=\"<f8\"),\n", |
| 87 | + " \"covwarn\": df[\"covwarn\"],\n", |
| 88 | + " \"fieldDec\": np.array(df[\"dec\"], dtype=\"<f8\"),\n", |
| 89 | + " \"ecl_lat\": np.array(df[\"ecl_lat\"], dtype=\"<f8\"),\n", |
| 90 | + " \"ecl_lon\": np.array(df[\"ecl_lon\"], dtype=\"<f8\"),\n", |
| 91 | + " \"observationId\": np.array(df[\"expnum\"], \"<i4\"),\n", |
| 92 | + " \"filter\": df[\"filter\"],\n", |
| 93 | + " \"observationMidpointMJD\": np.array(df[\"mjd_mid\"], dtype=\"<f8\"),\n", |
| 94 | + " observatory_1: np.array(df[\"observatory\"][:, 0], dtype=\"<f8\"),\n", |
| 95 | + " observatory_2: np.array(df[\"observatory\"][:, 1], dtype=\"<f8\"),\n", |
| 96 | + " observatory_3: np.array(df[\"observatory\"][:, 2], dtype=\"<f8\"),\n", |
| 97 | + " \"obs_ecl_lon\": np.array(df[\"obs_ecl_lon\"], \"<f8\"),\n", |
| 98 | + " \"fieldRA\": np.array(df[\"ra\"], dtype=\"<f8\"),\n", |
| 99 | + " \"fiveSigmaDepth\": np.array(df[\"m50\"], dtype=\"<f8\"),\n", |
| 100 | + " \"k\": np.array(df[\"k\"], dtype=\"<f8\"),\n", |
| 101 | + " \"c\": np.array(df[\"c\"], dtype=\"<f8\"),\n", |
| 102 | + " velocity_1: np.array(df[\"velocity\"][:, 0], dtype=\"<f8\"),\n", |
| 103 | + " velocity_2: np.array(df[\"velocity\"][:, 1], dtype=\"<f8\"),\n", |
| 104 | + " velocity_3: np.array(df[\"velocity\"][:, 2], dtype=\"<f8\"),\n", |
| 105 | + " }\n", |
| 106 | + "\n", |
| 107 | + " # Set exposure times\n", |
| 108 | + " exo_time_s = np.full(len(df), 90)\n", |
| 109 | + " mask = (temp[\"observationMidpointMJD\"] < 57447) & (temp[\"filter\"] == \"Y\")\n", |
| 110 | + " exo_time_s[mask] = 45\n", |
| 111 | + "\n", |
| 112 | + " # Convert TDB to TAI\n", |
| 113 | + " time = Time(temp[\"observationMidpointMJD\"], format=\"mjd\", scale=\"utc\")\n", |
| 114 | + " time_TAI = time.tai\n", |
| 115 | + " temp[\"observationMidpointMJD\"] = time_TAI.value\n", |
| 116 | + "\n", |
| 117 | + " # Check for NaNs\n", |
| 118 | + " if np.any(pd.isnull(temp[\"observationMidpointMJD\"])):\n", |
| 119 | + " print(\"Warning: NaN values found in observationMidpointMJD\")\n", |
| 120 | + "\n", |
| 121 | + " # Create DataFrame and insert exposure times\n", |
| 122 | + " df_hdl1 = pd.DataFrame(temp)\n", |
| 123 | + " df_hdl1.insert(8, \"visitExposureTime\", exo_time_s)\n", |
| 124 | + "\n", |
| 125 | + " # Save to SQLite database\n", |
| 126 | + " engine = create_engine(f\"sqlite:///{db_path}\")\n", |
| 127 | + " df_hdl1.to_sql(\"observations\", engine, if_exists=\"replace\", index=False)\n", |
| 128 | + "\n", |
| 129 | + " # Create indexes efficiently\n", |
| 130 | + " index_queries = [\n", |
| 131 | + " \"CREATE INDEX idx_lat_long ON observations(ecl_lat,ecl_lon)\",\n", |
| 132 | + " \"CREATE INDEX idx_filter ON observations(filter)\",\n", |
| 133 | + " \"CREATE INDEX idx_dec_ra ON observations(fieldDec,fieldRA)\",\n", |
| 134 | + " \"CREATE INDEX idx_dec_ra_mjd ON observations(fieldDec,fieldRA,observationMidpointMJD)\",\n", |
| 135 | + " \"CREATE INDEX idx_mjd ON observations(observationMidpointMJD)\",\n", |
| 136 | + " \"CREATE INDEX idx_m50_c_k ON observations(fiveSigmaDepth,c,k)\"\n", |
| 137 | + " ]\n", |
| 138 | + "\n", |
| 139 | + " with sqlite3.connect(db_path) as db:\n", |
| 140 | + " cursor = db.cursor()\n", |
| 141 | + " for query in index_queries:\n", |
| 142 | + " cursor.execute(query)\n", |
| 143 | + "\n", |
| 144 | + " print(f\"DES pointing data processed and saved to: {db_path}\")\n", |
| 145 | + "\n", |
| 146 | + "\n", |
| 147 | + "\n", |
| 148 | + "def process_ccd_visits_to_sqlite(\n", |
| 149 | + " use_url,\n", |
| 150 | + " pointings_db_path,\n", |
| 151 | + " fits_path=\"y6a1c.ccdcorners.fits.gz\",\n", |
| 152 | + " db_path=\"DES_visits.db\",\n", |
| 153 | + " url=\"https://github.com/bernardinelli/DESTNOSIM/raw/refs/heads/master/data/y6a1c.ccdcorners.fits.gz\"\n", |
| 154 | + " ):\n", |
| 155 | + " \"\"\"\n", |
| 156 | + " Processes a FITS file containing DES ccd visits and stores \n", |
| 157 | + " data into a SQLite database with indexing.\n", |
| 158 | + "\n", |
| 159 | + " Parameters\n", |
| 160 | + " ----------\n", |
| 161 | + " use_url : bool\n", |
| 162 | + " If True, the FITS file will be downloaded from the specified `url`. If False, the local file path\n", |
| 163 | + " specified by `fits_path` will be used.\n", |
| 164 | + "\n", |
| 165 | + " fits_path : str, optional\n", |
| 166 | + " Path to the local FITS file (default is \"y6a1c.ccdcorners.fits.gz\"). \n", |
| 167 | + "\n", |
| 168 | + " db_path : str, optional\n", |
| 169 | + " Path where the SQLite database will be saved (default is \"DES_TNO.db\").\n", |
| 170 | + "\n", |
| 171 | + " url : str, optional\n", |
| 172 | + " Direct URL to the raw FITS file hosted on GitHub.\n", |
| 173 | + "\n", |
| 174 | + " \n", |
| 175 | + " \"\"\"\n", |
| 176 | + " if use_url == True:\n", |
| 177 | + " response = requests.get(url)\n", |
| 178 | + " response.raise_for_status()\n", |
| 179 | + " with gzip.open(BytesIO(response.content), 'rb') as gz:\n", |
| 180 | + " with fits.open(gz) as hdul:\n", |
| 181 | + " df = hdul[1].data\n", |
| 182 | + " else:\n", |
| 183 | + " with fits.open(fits_path) as HDUl:\n", |
| 184 | + " df = HDUl[1].data\n", |
| 185 | + "\n", |
| 186 | + "\n", |
| 187 | + " ra = df[\"ra\"]\n", |
| 188 | + " dec = df[\"dec\"]\n", |
| 189 | + "\n", |
| 190 | + " temp = {\n", |
| 191 | + " \"visitId\": df[\"expnum\"],\n", |
| 192 | + " \"detectorID\": df[\"ccdnum\"],\n", |
| 193 | + " \"llcra\": ra[:, 0],\n", |
| 194 | + " \"llcdec\": dec[:, 0],\n", |
| 195 | + " \"lrcra\": ra[:, 1],\n", |
| 196 | + " \"lrcdec\": dec[:, 1],\n", |
| 197 | + " \"urcra\": ra[:, 2],\n", |
| 198 | + " \"urcdec\": dec[:, 2],\n", |
| 199 | + " \"ulcra\": ra[:, 3],\n", |
| 200 | + " \"ulcdec\": dec[:, 3],\n", |
| 201 | + " \"ra\": ra[:, 4], # center RA\n", |
| 202 | + " \"dec\": dec[:, 4], # center Dec\n", |
| 203 | + " }\n", |
| 204 | + "\n", |
| 205 | + " # Create DataFrame\n", |
| 206 | + " df_hdl1 = pd.DataFrame(temp)\n", |
| 207 | + " \n", |
| 208 | + " engine_pointings = create_engine(f\"sqlite:///{pointings_db_path}\")\n", |
| 209 | + " df_depth = pd.read_sql(\"SELECT observationId, fiveSigmaDepth FROM observations\", engine_pointings)\n", |
| 210 | + " df_depth = df_depth.set_index(\"observationId\")\n", |
| 211 | + " df_depth = pd.read_sql(\"SELECT observationId, fiveSigmaDepth FROM observations\", engine_pointings)\n", |
| 212 | + " df_depth = df_depth.astype({\"fiveSigmaDepth\": \"<f8\"}) # just to be safe on dtype\n", |
| 213 | + " df_hdl1[\"magLim\"] = df_hdl1[\"visitId\"].map(df_depth[\"fiveSigmaDepth\"])\n", |
| 214 | + " \n", |
| 215 | + "\n", |
| 216 | + " # Save to SQLite\n", |
| 217 | + " engine = create_engine(f\"sqlite:///{db_path}\")\n", |
| 218 | + " df_hdl1.to_sql(\"observations\", engine, if_exists=\"replace\", index=False, chunksize=10000)\n", |
| 219 | + "\n", |
| 220 | + " # Create spatial index\n", |
| 221 | + " index_sql = \"\"\"\n", |
| 222 | + " CREATE INDEX IF NOT EXISTS idx_obs \n", |
| 223 | + " ON observations(visitId)\n", |
| 224 | + " \"\"\"\n", |
| 225 | + "\n", |
| 226 | + " with sqlite3.connect(db_path) as db:\n", |
| 227 | + " db.execute(index_sql)\n", |
| 228 | + "\n", |
| 229 | + " print(f\"DES CCD visits data processed and saved to {db_path}\")\n", |
| 230 | + "\n", |
| 231 | + "\n" |
| 232 | + ] |
| 233 | + }, |
| 234 | + { |
| 235 | + "cell_type": "code", |
| 236 | + "execution_count": 2, |
| 237 | + "id": "f09d3b0b", |
| 238 | + "metadata": {}, |
| 239 | + "outputs": [ |
| 240 | + { |
| 241 | + "name": "stdout", |
| 242 | + "output_type": "stream", |
| 243 | + "text": [ |
| 244 | + "DES pointing data processed and saved to: DES_TNO.db\n", |
| 245 | + "DES CCD visits data processed and saved to DES_visits.db\n" |
| 246 | + ] |
| 247 | + } |
| 248 | + ], |
| 249 | + "source": [ |
| 250 | + "pointing_db = \"DES_TNO.db\"\n", |
| 251 | + "process_pointings_to_sqlite(True,db_path=pointing_db)\n", |
| 252 | + "process_ccd_visits_to_sqlite(True,pointings_db_path=pointing_db)" |
| 253 | + ] |
| 254 | + }, |
| 255 | + { |
| 256 | + "cell_type": "markdown", |
| 257 | + "id": "ad8e5071", |
| 258 | + "metadata": {}, |
| 259 | + "source": [ |
| 260 | + "For running DES, obtain the databases using this notebook, run the command line arg `sorcha init` and select the **DES config file** option. Then run the code with:\n", |
| 261 | + "\n", |
| 262 | + "sorcha run -c DES_config_file.ini --pd DES_TNO.db --ob orbits_filename.csv -p colours_filename.csv -o ./ -s des --vd DES_visits.db\n", |
| 263 | + "\n", |
| 264 | + "`orbits_filename.csv` and `colours_filename.csv` should be your input files for object's orbits and colours, respectively." |
| 265 | + ] |
| 266 | + }, |
| 267 | + { |
| 268 | + "cell_type": "code", |
| 269 | + "execution_count": null, |
| 270 | + "id": "115468ba", |
| 271 | + "metadata": {}, |
| 272 | + "outputs": [], |
| 273 | + "source": [] |
| 274 | + } |
| 275 | + ], |
| 276 | + "metadata": { |
| 277 | + "kernelspec": { |
| 278 | + "display_name": "sorcha", |
| 279 | + "language": "python", |
| 280 | + "name": "python3" |
| 281 | + }, |
| 282 | + "language_info": { |
| 283 | + "codemirror_mode": { |
| 284 | + "name": "ipython", |
| 285 | + "version": 3 |
| 286 | + }, |
| 287 | + "file_extension": ".py", |
| 288 | + "mimetype": "text/x-python", |
| 289 | + "name": "python", |
| 290 | + "nbconvert_exporter": "python", |
| 291 | + "pygments_lexer": "ipython3", |
| 292 | + "version": "3.11.11" |
| 293 | + } |
| 294 | + }, |
| 295 | + "nbformat": 4, |
| 296 | + "nbformat_minor": 5 |
| 297 | +} |
0 commit comments