test_compress.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "dfab84bb-b622-4641-85ec-57cae732c600",
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "import numpy as np\n",
    "\n",
    "from matplotlib import pyplot as plt\n",
    "from astropy    import units\n",
    "\n",
    "from common import hallmark as hm\n",
    "from common import io_ipole as io\n",
    "from common import viz\n",
    "from common import dalt\n",
    "from common import mockservation as ms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "0e1cc6f5-7965-4268-bb9a-2266c9a7311f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Rhigh [10 40]\n",
      "snapshot [5000 5001 5002 5003 5004 5005 5006 5007 5008 5009 5010 5011 5012 5013\n",
      " 5014 5015]\n",
      "window [5]\n",
      "aspin [0.94]\n",
      "mag ['S']\n",
      "inc [50 70]\n"
     ]
    }
   ],
   "source": [
    "pf = hm.ParaFrame(\n",
    "    'model/Illinois_thermal/230GHz/{mag}a{aspin:g}_w{window:d}/img_s{snapshot:d}_Rh{Rhigh:d}_i{inc:d}.h5'\n",
    ")\n",
    "\n",
    "for k in set(pf.keys()) - {'path'}:\n",
    "    globals()[k] = np.unique(pf[k])\n",
    "    print(k, globals()[k][:16])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "616a52df-ac34-42d5-9e85-23957df93f8a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>path</th>\n",
       "      <th>mag</th>\n",
       "      <th>aspin</th>\n",
       "      <th>window</th>\n",
       "      <th>snapshot</th>\n",
       "      <th>Rhigh</th>\n",
       "      <th>inc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>model/Illinois_thermal/230GHz/Sa+0.94_w5/img_s...</td>\n",
       "      <td>S</td>\n",
       "      <td>0.94</td>\n",
       "      <td>5</td>\n",
       "      <td>5000</td>\n",
       "      <td>40</td>\n",
       "      <td>70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>model/Illinois_thermal/230GHz/Sa+0.94_w5/img_s...</td>\n",
       "      <td>S</td>\n",
       "      <td>0.94</td>\n",
       "      <td>5</td>\n",
       "      <td>5001</td>\n",
       "      <td>40</td>\n",
       "      <td>70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>model/Illinois_thermal/230GHz/Sa+0.94_w5/img_s...</td>\n",
       "      <td>S</td>\n",
       "      <td>0.94</td>\n",
       "      <td>5</td>\n",
       "      <td>5002</td>\n",
       "      <td>40</td>\n",
       "      <td>70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>model/Illinois_thermal/230GHz/Sa+0.94_w5/img_s...</td>\n",
       "      <td>S</td>\n",
       "      <td>0.94</td>\n",
       "      <td>5</td>\n",
       "      <td>5003</td>\n",
       "      <td>40</td>\n",
       "      <td>70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>model/Illinois_thermal/230GHz/Sa+0.94_w5/img_s...</td>\n",
       "      <td>S</td>\n",
       "      <td>0.94</td>\n",
       "      <td>5</td>\n",
       "      <td>5004</td>\n",
       "      <td>40</td>\n",
       "      <td>70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1991</th>\n",
       "      <td>model/Illinois_thermal/230GHz/Sa+0.94_w5/img_s...</td>\n",
       "      <td>S</td>\n",
       "      <td>0.94</td>\n",
       "      <td>5</td>\n",
       "      <td>5995</td>\n",
       "      <td>40</td>\n",
       "      <td>70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1993</th>\n",
       "      <td>model/Illinois_thermal/230GHz/Sa+0.94_w5/img_s...</td>\n",
       "      <td>S</td>\n",
       "      <td>0.94</td>\n",
       "      <td>5</td>\n",
       "      <td>5996</td>\n",
       "      <td>40</td>\n",
       "      <td>70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1995</th>\n",
       "      <td>model/Illinois_thermal/230GHz/Sa+0.94_w5/img_s...</td>\n",
       "      <td>S</td>\n",
       "      <td>0.94</td>\n",
       "      <td>5</td>\n",
       "      <td>5997</td>\n",
       "      <td>40</td>\n",
       "      <td>70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1997</th>\n",
       "      <td>model/Illinois_thermal/230GHz/Sa+0.94_w5/img_s...</td>\n",
       "      <td>S</td>\n",
       "      <td>0.94</td>\n",
       "      <td>5</td>\n",
       "      <td>5998</td>\n",
       "      <td>40</td>\n",
       "      <td>70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1999</th>\n",
       "      <td>model/Illinois_thermal/230GHz/Sa+0.94_w5/img_s...</td>\n",
       "      <td>S</td>\n",
       "      <td>0.94</td>\n",
       "      <td>5</td>\n",
       "      <td>5999</td>\n",
       "      <td>40</td>\n",
       "      <td>70</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1000 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                   path mag  aspin  window  \\\n",
       "1     model/Illinois_thermal/230GHz/Sa+0.94_w5/img_s...   S   0.94       5   \n",
       "3     model/Illinois_thermal/230GHz/Sa+0.94_w5/img_s...   S   0.94       5   \n",
       "5     model/Illinois_thermal/230GHz/Sa+0.94_w5/img_s...   S   0.94       5   \n",
       "7     model/Illinois_thermal/230GHz/Sa+0.94_w5/img_s...   S   0.94       5   \n",
       "9     model/Illinois_thermal/230GHz/Sa+0.94_w5/img_s...   S   0.94       5   \n",
       "...                                                 ...  ..    ...     ...   \n",
       "1991  model/Illinois_thermal/230GHz/Sa+0.94_w5/img_s...   S   0.94       5   \n",
       "1993  model/Illinois_thermal/230GHz/Sa+0.94_w5/img_s...   S   0.94       5   \n",
       "1995  model/Illinois_thermal/230GHz/Sa+0.94_w5/img_s...   S   0.94       5   \n",
       "1997  model/Illinois_thermal/230GHz/Sa+0.94_w5/img_s...   S   0.94       5   \n",
       "1999  model/Illinois_thermal/230GHz/Sa+0.94_w5/img_s...   S   0.94       5   \n",
       "\n",
       "      snapshot  Rhigh  inc  \n",
       "1         5000     40   70  \n",
       "3         5001     40   70  \n",
       "5         5002     40   70  \n",
       "7         5003     40   70  \n",
       "9         5004     40   70  \n",
       "...        ...    ...  ...  \n",
       "1991      5995     40   70  \n",
       "1993      5996     40   70  \n",
       "1995      5997     40   70  \n",
       "1997      5998     40   70  \n",
       "1999      5999     40   70  \n",
       "\n",
       "[1000 rows x 7 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sel = pf(inc=70)\n",
    "sel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "8ba0eeb0-a699-4bff-b2cb-63e4e3063185",
   "metadata": {},
   "outputs": [],
   "source": [
    "imgs = io.load_mov(sel[:100].path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "38e93f14-0ba5-428f-8985-de8e1a9a7a89",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXwAAAD8CAYAAAB0IB+mAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAAi+ElEQVR4nO3deXgV9aH/8ff3ZDnZSEIWAiEkISEIkVUjVEEBRUW94FoXqAtytbZa29reilV/Lm21t3WjvagXFRFtXbC4ouKCXMSdRTbZAklIQlay78k58/tD2ioGSEhO5iyf1/PwPOY7c2Y+yciHb+bMmTGWZSEiIv7PYXcAERHpGyp8EZEAocIXEQkQKnwRkQChwhcRCRAqfBGRABFsd4AjSUhIsNLT0+2OISLiM9avX19pWVZiZ8u8uvDT09NZt26d3TFERHyGMabgcMt0SkdEJECo8EVEAoQKX0QkQKjwRUQChFcWvjFmpjFmUW1trd1RRET8hlcWvmVZb1iWdX1MTIzdUURE/IZXX5ZZ0VjLG19+yNDGStyWm60teQwKiScxOJbq+FHsaNlHCk5SGytotzrY0bKP5JAE4oOjqUoYw66WfFItJ8lNlbS629nVVkRKSCJxwdGUx49ib3M+Qy0nA1qqaXa3sae1iFTnQPoFRVEWN5J9zflkWE7iW+todDVT0FZKmnMQkUGRlPYfTlFTHsMIJ7atgXp3I4WtpaQ7U4gIjqAkdij7G/MZTgT9OpppcDWwv72CjLAhOEPCqYweQkVLCVlWKOGuZuo6Gqlqr2FIeDKhIRFU9RtITWsl6ZaDUFcLDa5m6tvrGRQxkKCQCOoi4ml01TPI7SDYctHsbqXZ1Up8RCKOYCctIU7aXC3EBYWDgQ6rA0dQEJHOSIKDgnA4jN2HV0T6mPHm++GHDw23pt04i7ca3qHJGCamD+GWqmrm1tZzUfut7B7+PMPLsvlH0ztUBjmYlprCnZVVXFrfwFmu2ygZ9jfGlY7k2eaV7AsO5rwhydxXXsnMxiYmWbdTl/Esk/cP57HW99keGsKlgwexoKyC05uaGc8ddAxdyjnFmfyp7UPWhTmZOyiJJ0vKmNjSygjHnYSkPcOlhenc2bGGNeFh3DhwAM8XlzKitYMRofMJT3mO6woGc7P7U96KjODWAQm8VrSfhLZQTgj7CeHJy/iv/ESustbzcr9I7kmI5719xbg7YpgSMZuwga9z394YZpotPBPdjwfi+/NJfiGlrmTO63cuzgEreXxvGJPMLh6LjebR/rFsytvHZncml8ZMJjT+/3hpr0W2o4AH+8fyYnQUXxQU8bHreK6NG0tI9Fcs31dHChU8Et+PT8NDea64jk8cOdyZkIE7JJ+lpSVEWs0sjYWCEMMtVVFsDjuRp/vH4HbUcN+B/QQZ+CC8nuYgOM+VSmnMOD6IjsBhXFxWX4EJjeBrKgkKDmN0zPGYuAyqI0KJDXOSFhxERFQ/nOEx4PDKXzhFfIoxZr1lWTmdLvPmws84/jhr0fPPMri1AbflZldzPgNC4ogLiaU+Zii7mwsYaMIZ1NpIh9VBbnMBSSEJxAbHUBedzp6WApJNOEmtjbS529jTUkhySAIxwf2oiskkvzmfVCuMhNY6Wtyt5LUUMdiZRHRQJBWxWRQ0Fxyc4dfS0NFEQVsJqSEDiQoKpyQ+m6KmPLLcIcS2VFPvamRfWxkZzoFEOMIoTBzD/sY8RnYYoluqqe5ooKC1lCxnCqGOcPbEjaSoMZdxHSFEt1ZT1l5DQXsZY0LTCAqOYltsBsWNuZzSEUZkazX72w+Q7yrjpOBhuEOi+So6hZLm3ZzZEkJEWxX5HWUUuCo5zZFJq7M/n0SnUdq6h4sb2glrPcAOdzn7qOEsdxo1oYN4u98AKtrymFfTSFh7NV84yigyjVzYMphiZwYvRUVS7y7i/1UUE95Rz9vhNZQFt3FDdTibneNYGB1Kq6nkhf0bCbVaeCAhiKog+EtJLS+5pvJQcjMO08q2so8BuGbgABzA4tJyFnbMYmHKAYJcIWypWA3A7EFJJHbArZUWL4ddwosJB4g3sdxRV4jljGZNeD2JzgTGxY7DGpxDXVQwqTEJpMYmEBSkfyhE/slnCz8nJ8fSJ219j2VZNHe00dTahqOpnvaWBkpqi2htqae/5aQ2OJ5PW8pob2njlIpc3K2NvN++lciODqY0hfJ52Km84NxEVJuTZ0rfIcpqZFZqHJObW7insopb26/jzYyPiW5IYXXVh1SZWOYPjGRcezSTHMPZl34JO6NqGR+XyYSkFOITkjD67UECxJEK36vP4YtvMsYQEeIkIsQJUf0ASGL4d9Y56ZDXTDpk2U2HLH+5voaq6lJy6+q40IohtmoC4Y2tfFU4AJrKaHPsJaS9hpG1K/jL/nQ2DX+JteuO5+36t6m2wpibMoAZLYmMDRtL5YjLsRIMU4aOIiYsore/fRGvpcIXn9C/Xyz9+8UCMAz4AccdXHIlAM8f/MqyLP6nuZX1JefTWlbK5wWjqK3KJdLaSlxzKRMqN3B64SCqMp9l3Osj+e+WTeyOyuTdhEimJJ/DhBNmEhvf6X2nRHyeTulIQGlvbWZnZQ0r9qxhwP4yRhe+R7l7F/OTQni6pJzxLW3M6P9nHAkbuTpxKidnZJOWma1TQuIzdEpH5KAQZzijBoczavBlB0duBmB8eR6VOzewdt9OIttd5LZtJPzTLaR/upnlEUm8FzuAOUmXMGrS5cQmDLTvGxDpAc3wRTrhdrspyt1M+ab3WFW2gtXOct4sLmajO4vfDL6CrEGGO8bMIiV9mN1RRb5DV+mI9FBHRwe5mz9ma34pD1StpKG9lPWln1MYnMaq1AlMPvU6sjNOtDumiApfpLflFhdSvvoZYvJe5brkZqY0NTOnIY3aU27nxJOnEhais6ViDxW+iAd9seVd9n++nAlFHzDbdQMH0l/l4tjZ3DzpXOIGptkdTwKM3rQV8aAJo8+C0Wfhcrn48ZYNLPxqPcO2LiNq8+2sSJhO3JTrOHnMmXbHFNEMX8QTCnZvofSdB1hu1rImIowHa7JInXk3KVnj7I4mfk4zfJE+lpY1mrSsZ4gr+IrMd//I2NpVLF36IKtHjuOu6ZeRPSDF7ogSgFT4Ih6UlTaOrOteoLKsiOKPt7Ct7jZ+9dxX3N1/IGMvv5uw8Ei7I0oA0ccHRfpAQlIKv7/oHJ48/UVucIQxsWARHz9yIkvf+qPd0SSAqPBF+tAPUo/j/F88wbbpS1kWE8zSkmf47MFZVJXvtzuaBACvLHw901b83fGTz+fPV6/lBjOVE+rW8uhjD/HoZ+/aHUv8nK7SEbFZ3s5NXLvqXSojljLTcSV3XHQDEZHRdscSH3Wkq3S8coYvEkiGHjeWN6+9mUmR8/h17gLKHpzE3u2a6EjvU+GLeIFIp5PHL/kF+6cvxE0tP197JUteu9PuWOJnVPgiXmTUqRfQPOcVItxOTt72GJ8++StcLrfdscRPqPBFvMyozBNZcuXH1EacSXHBLn64dBFVTfV2xxI/oMIX8ULhEZGcdPPf2H/Gb9hlHmP2s/OpKttndyzxcSp8ES9lHA5unnoqPxl+P38qWUPz42dSvHeH3bHEh6nwRbzcjaecS+h5jxBMPfe/fQEffLnc7kjio1T4Ij5gxElnkH/+k2wPC6Luw1vYs+FDuyOJD1Lhi/iIiePO5Ympz5PTHEbl63fwZV6p3ZHEx6jwRXxIRuYYgq5dwe39fsS1qy7hlW1f2B1JfIgKX8THJA/J5MHZ5xDqSqJp+R3kbf3M7kjiI1T4Ij5o7KB0Vsz4I9NducS8/EPydnxidyTxASp8ER81MDWLhite4bH+4fxqzX9SUrDJ7kji5VT4Ij4s87ixDBl3Kyc2t+FaejlN1XojVw5PhS/i466aMo/p4/4CHW3cu+wdWtrb7Y4kXkqFL+IHJk45lw/Oeo3XWcrFL96G5XbZHUm8kApfxE9cPWkUJyWczvj8PWx57Grw4ocbiT1U+CJ+ZMmFtzMtPo0xFW+waunP7I4jXkaFL+JHjDFMvf4hnog7jV9bq1m67Ba7I4kXUeGL+JmgIAc/nPccZzRGccG2p9m+6m92RxIvocIX8UOxkTHcceVK8oNG8uTqr/lg99d2RxIvoMIX8VMx0TEk3fw+Hw6s5BcfzWNbsR6gEuhU+CJ+LCkmgodn3Ex29QhinjyTA/vz7I4kNlLhi/i5KRnHc/fp84h11/L1klkUlu21O5LYRIUvEgBGjp/MjmkLuTvJxV2vXExTQ63dkcQGKnyRAJEz9WJ+FHcRd1YUkvfX82lrabY7kvQxFb5IAJl74e8pH3U3HzUO5ooXn6GhtcXuSNKHVPgiAWbihTdRM+1qdrGQuc/dTUtzo92RpI+o8EUC0PwzzmLOoNtZsO958h4+m7qaSrsjSR/os8I3xkQaY54xxjxhjJnTV/sVkc7NP+tyCk/8LUlt27n179PYsOMjuyOJh/Wo8I0xi40x5caYrYeMzzDG7DTG5Bpj5h8cvgh42bKs64BZPdmviPSOibN+zNrJv2NrmJuqV3/Epg+X2R1JPKinM/wlwIxvDxhjgoCFwDlANnCFMSYbSAEKD66mm3WLeIlZ03/KEycvJqMtlvJVj/LTl1/gQGOD3bHEA3pU+JZlrQGqDhmeAORalrXXsqw24AXgfKCIb0q/x/sVkd41YuREkn+9lrVjb2VN/R+58Omb+fy1x3F1dNgdTXqRJ4p3MP+eycM3RT8YWA5cbIx5DHjjcC82xlxvjFlnjFlXUVHhgXgi0pmwiH7cc8kM5o9/kJ821TFx4618/KdR3LV0Dnn7dtodT3pBsAe2aToZsyzLagTmHu3FlmUtAhYB5OTk6JE9In3sR+POwD16Khvf/xuf7XyYN12b+PXiH7A69FT+nnUZibGtzOmXTP/4JOIGDcUZFWt3ZOkiTxR+ETDkW1+nAPs9sB8R8RBHUBDjz76K8WdfxfQt77OTd9hXDZ+Ur8Sq2cetRTuINs3clhhPYXAovy8PZX3MWawYNIzYyCDm9UtkyPGnEB03wO5vRb7FE4X/JZBljBkKFAOXA7M9sB8R6QMnjJ4Oo6eTA8x2udi0v5A9BdtpqSwgpuZjQlorqAqF0ibDlwdWQkU7D5V/BB/Ab+NTcUYdx/VT7mFQxvF2fysBz1g9eNCxMeZ5YCqQAJQBd1mW9ZQx5lzgESAIWGxZ1h+OZfs5OTnWunXrjjmfiPQty7Ioqa6iYtsnNO75mIWt7zChuZqfVdfy1MA7GXPuhUxMy7A7pl8zxqy3LCun02U9KXxPMcbMBGYOGzbsut27d9sdR0R6oLRoD3vf/iu3VKbRMGgJM8Kv5PczbyAsIsruaH7J5wr/nzTDF/Efew6U8l8r/8rdO18lIthJ+I+eZ/DQ4+yO5XeOVPi6Hl5E+kRm/ECWz/4DrlPvJMZdwh/fnsXKz563O1ZAUeGLSJ8aP/1yCi56jp2hwTR89Fv2bP7Y7kgBQ4UvIn0uZ/QZLJr2d05pdFDxym2U1OphLH1BhS8itkjPHEvTFa9wY+gFzFp2DbUtTXZH8nteWfjGmJnGmEW1tXrupog/yzxuDNdMHUazq4KXXnjI7jh+zysL37KsNyzLuj4mJsbuKCLiYT87eRYPm3Fcl/9nNn/4st1x/JpXFr6IBJbTrvoduxyprP3qdmrq9PQtT1Hhi4jtnGGRfDXpBh6LC+XZl35mdxy/pcIXEa9w6Rk3cm9NGj8ufJfyfTvsjuOXVPgi4jUmXriA1dY4Fq/V/fc9QYUvIl4jOTWTl8dczXMt9/GZHrrS67yy8HVZpkjg+uVppxPelMqmd5baHcXveGXh67JMkcA1ckAKfzExzCt+lMr9eXbH8SteWfgiEthSZtxCeZCD196+2+4ofkWFLyJeZ0hmNr9NyuAZ1tHQWGd3HL+hwhcRrzRn2PW8UFLC1+//ze4ofkOFLyJeafq0a6l3p7EnV9fk9xYVvoh4JeMI4h8TF3BPcC2f79tldxy/4JWFr8syRQTg3DGpBPfbxsvrV9sdxS94ZeHrskwRARg3KJ2HDmTw880P4na57Y7j87yy8EVE/ikuazIpVim7NnxgdxSfp8IXEa+WOfVy5g5MYvEmPSClp1T4IuLV+scm0t/dn2F1eVhundbpCRW+iHi92UmXcX1dCbu2fG53FJ+mwhcRr5c15XJu6fgJL+bX2x3Fp6nwRcTrxSQk83nmfl6pfMDuKD5NhS8iPuHMxGmcWBFNyb7ddkfxWV5Z+PrglYgcau7wE3iy5T0KP3/V7ig+yysLXx+8EpFDpQ4bzXZHIrlF79gdxWd5ZeGLiBzKOBwsHJjCgthSGpr05u2xUOGLiM/4j5TzWVxSTu7Gj+yO4pNU+CLiM06bdCVZrW5Kdm+yO4pPCrY7gIhIV0XEJHDRoIepbi3mHLvD+CDN8EXEp0Qk5lHhfJHyBj36sLtU+CLiU+alTeORvDCKN6y2O4rPUeGLiE/5wXFjmMYOGnautjuKz1Hhi4hPCe8Xyz/6pfF2y1q7o/gcFb6I+JytsQP5PKKBqgZdj98dKnwR8TmXplzA+4XF7Nm60e4oPsUrC1/30hGRIxk2/izWusewrfCA3VF8ilcWvu6lIyJHEpaUyZ0pZ/JUnZ5z2x1eWfgiIkeTENNGbet+2jpcdkfxGSp8EfFJd4QNYGPZKnLzCuyO4jNU+CLik5IzRxNkLEq3rbE7is9Q4YuIT0ocPpHbEhJYVv53u6P4DBW+iPgkExpBkIkmoVlX6nSVCl9EfNZlYT/g9qoCKmsb7I7iE1T4IuKzgkdfwn0ds9mQV2p3FJ+gwhcRn5U69lT+kb6FxbuX2h3FJ6jwRcRnRTpDSXYfR3hlq91RfIIKX0R82sMt5fyhdBntLrfdUbyeCl9EfJo1ZCL9HVVs27nZ7iheT4UvIj7NZI7ilLQU3tyy2O4oXs8rC193yxSRrsrOns61NQ2kV1TYHcXreWXh626ZItJVjhAnZ7UM4cSq3XZH8XpeWfgiIt2xIXs+czuuoKyuzu4oXk2FLyI+rzk5hMbMJSzfoefcHokKX0R83nnDJ3BC6Uhi8vbYHcWrqfBFxOclRcfyh9YtjMh/y+4oXk2FLyJ+YU9MNkVmBy5Xh91RvJYKX0T8wvoBCdyeFMUX21bZHcVrqfBFxC9Mzf4hz+4vxezZbncUr6XCFxG/MHbUNNJanDSV7rU7itcKtjuAiEhvCAoK4qbUB8lr2ckndofxUprhi4jfiEosoj5qGbsr9UCUzqjwRcRvXJMxhf/e24+a9R/YHcUrqfBFxG/kDB/FVHJx7X7f7iheSYUvIn4jJDSM5TEjecf9OViW3XG8jgpfRPzKvqQMPoyCrdt1X51DqfBFxK/MOeVXfLivmLoN+gDWoVT4IuJX0oaOYk3IFNaWhx52nY0fvMhbf57EaYuPZ8GCk9i59pU+TGgfFb6I+J2PTrqJJZEf8H7upu+M17Y0cNfSBYz/6HrGNlWQ057A6MZ6st6by5tvv2FT2r6jD16JiN+ZkzOWl/IcfLnmdaYPGwtARV0zFyy7hVqzkeFZ93LpD2/goVAnTU0NPPDssyzOfZ7da+GXk2fanN5zvHKGr2faikhPZMQl8ajJ5taCB9m86iV2bf6M2ocnklCYwdVZtzFnzs8JCXUCEBERxXVX/YioqBre/GwFBXt32pzec4zlxZcu5eTkWOvWrbM7hoj4oIa6asofmcozcXU0GcNtFR1UzXqGYSdM63T9kvJSIh8dz97QkYy59X0cQV45Hz4qY8x6y7JyOlvmm9+RiMhRREX3J/Zn7+GKOZmgqBxcP/7osGUPMGjAQHJH/RLMFhYs+3kfJu07muGLiBxkuV3MfSKHA44WHjv7dVJSs+yO1G2a4YuIdIFxBPGbKf/D08UHqPz7jVhut92RepUKX0TkW7JHTCJ3xC95tz6Vpz7zrzMMKnwRkUNMuHQ+r2cEs2D7L8irqrQ7Tq9R4YuIHMLhMNw15ToyD4yk6Omb/OZGbCp8EZFOnDnsRH6TnMqp9W+zYcX/2h2nV6jwRUQOY8Jlv2VZ5HDuKvkL+/bvsDtOj6nwRUQOIzgkhLhptxPt7mDLst8c83Z2lWxi5uIT+PiBcZTv+rIXE3aPCl9E5AjOyLmAnzrO5bzq/2PHho+OaRtvvvJrhrTU0V4Xyc9f3UtLu6uXU3aNCl9E5CjGXX43vwz6Obds7P7sfPfWdfyi7EtmB80g+MrX+KKljJ+89XsPpDw6Fb6IyFFE9oulbXwchcFP8/rX67v12j2rFuLCwZhLbmdyVgLpyasoLPsQO+5yoMIXEemCe0+fx/lFY4h6//Euv6axtY07I7Zzy5BJRMcPAuDB4GRWFG9he/5+T0U9LBW+iEgXxEf0Y1ZMCJMOvELNgfIuvWbVzhJqqydz/Jgf/2ssdcKlOE0Hxev6/oErKnwRkS6KPuVKFsRH8uwH93Vp/aJNa0ltOIH/PPHcf43FjTiV++MG8FHZC56KeVgqfBGRLho+dir/Fx5FZfkXR13X5XYxtPg+/jv8SYK/fW99RxC7wxPocJfS1sdX66jwRUS6yBEUxB3tJ/Pbil001h/5iXzrdq/lvwY5WJOc9r1lv0y8lNsO1LA3f4+nonZKhS8i0g0RYy/mTfdENu8pPOJ6rbu2saCsgskjL/resuiTr2Jc6yI2VDk9FbNTKnwRkW4YljOdOwdGsDD3tSOuF77vSyY0wvhxZ39vWeqAOCIGr+Dp3Ls8FbNTwX26NxERHxfpDKG/M5aaAzVHXK+geSPV4cM4KyT0e8uMMUwKriW6ep+HUnZOM3wRkW663xHPq/ufpKGuqtPl7S4X98QHsTgl+7DbuCw4lT9VbqO9sdpTMb9HhS8i0k0xGScRbNzkb1rb6fK9FY005t3I9Mz/POw2QgaPB6B0Z9/dTE2FLyLSTcFDMzhrSDKr8l7tdHnppveZ61rH1LSMw24jNH0kU1IHs2zX3z2U8vtU+CIi3TR8yChGtEBcTXGny/PzniMn+k2GJsUddhsjhmYzucHF4Nq+O6WjN21FRLopxBHCNY0ZJDbv7XT5e2YPtXH9OSfo8HPq6PBQTq4ZSXj/ZE/F/B7N8EVEjkFexhz+0H4BdS1t3xm33C7+UlrCdW2jj7qN55Pns8Bc5qmI36PCFxE5BgWDQlk7bAWr9277znhpwQ7irFYSk0486jaaopazN/y2PrtVsgpfROQYTBs6jrgD46jO2/md8c92rOLJ6GhC0kYcdRunOWOZW1dJ5e6+uVJHhS8icgzGJw/j5fqPGLXr5e+Mv93qZkF8LIOHH32GPy3lNH5VW8aBwu2eivkdffamrTEmA7gdiLEs65K+2q+IiKfsC8/C1bzju4N1k4g/kMWgmAFHff2gjJE0GUP5/q0c/feBnuvSDN8Ys9gYU26M2XrI+AxjzE5jTK4xZv6RtmFZ1l7Lsub1JKyIiDd5LjGEmwZZNNT/+9LKiwruZW7E5i69vl90OKelpvBO0wZPRfyOrp7SWQLM+PaAMSYIWAicA2QDVxhjso0xo40xbx7y5+j/1ImI+JjJSdO4q7KK/O2fAVBVXcGXsZtxhHV+ueahIkIimF0bysSGFk/G/JcuFb5lWWuAQ28aMQHIPThzbwNeAM63LGuLZVn/ccifrj0PDDDGXG+MWWeMWVdRUdHlb0REpK9NGnsx/9HYRPPeb2boW7etZmVkBLVx8V3eRnrwGVS1Hv0Szt7QkzdtBwPfviF00cGxThlj4o0xjwPjjTG3HW49y7IWWZaVY1lWTmJiYg/iiYh41sDB6cwJmc+zVhYAQUX5fLSvmEtPuKbL29iZeQ33t59EWYPnJ7g9KXzTydhhLya1LOuAZVk3WJaVaVnW/T3Yr4iIVzDGsGfwOtY2Lf1moGwrDUQwYPDwLm8jProVZ/ojLN+xwkMp/60nV+kUAUO+9XUKsL9ncUREfMt5UWfSsWMVDfU1LHOUsDIhg3tNZ/Phzv2AElLLDhCc6vmLJnuyhy+BLGPMUKAYuByY3SupRER8xAUDkhi9dSXrv3iXlVYyQxO7V6tJyemMaWrkk6rO763fm7p6WebzwKfAccaYImPMPMuyOoCbgJXAduAly7K2HWk7IiL+JivnTNaHRvDG5iU077+EG7Pv6dbrEwdnsjs4hLzqLR5K+G9d+qfIsqwrDjP+FvBWryYCjDEzgZnDhg3r7U2LiPSqsIgoXuyfxt6gPbzrnM+QzM+69XpHaBj3JgyggU3M8VDGf+3Lw9s/JpZlvWFZ1vUxMTF2RxEROaqfnXY/P61pJSprMuGRUd1+/WUNMcyrtFixdwVF9UUeSPgN3Q9fRKSHhhw/jSHZedCNN2u/rT3hYrbsLWbPzpdYvns5T539VC8n/IYKX0SkNxxj2QMUpE/hmcI1vHbCTUQ4j307R+OVp3RERAJJpWsdEalLaKyuJyP28M/B7SkVvoiIzc4aMJ6FpeV8uXkhFU2e+8StVxa+MWamMWZRbW2t3VFERDxuRNpY2txO/rf5M/Lr8j22H68sfF2lIyKBJDLMYrcjkd/X9mNUwiiP7ccrC19EJJB0WO08OrCNA64ywoPCPLYfFb6IiM1inDFcUJfDNbWV0Bhg5/BFRAKNM/aH3BvySwh2emwfKnwRES8QmpjBc40TcIVGe2wfXln4ukpHRAJNckwoJ1hfU1243WP78MrC11U6IhJokmPD+VvofXSsf9Zj+9CtFUREvMCg2EgqiMFd67nnSHnlDF9EJNAMjA6j3OoP9WUe24cKX0TEC8RGhFBJf0Kbyz22DxW+iIgXMMbQEJpARJvnrsPXOXwRES+xJvZC1rvP4Hce2r5m+CIiXsKdcByrWzz3aFevLHxdhy8igSg9oo2culW4a4o9sn2vLHxdhy8igSgjtJqHg/9Kw55PPbJ9ryx8EZFAFJ2UBkBd+T6PbF+FLyLiJVKSB9NihdBcWeCR7avwRUS8xJD4SEqseNw1RR7ZvgpfRMRLOIODqHQOprH2AO0ud69vX4UvIuJFWnJ+yr1ht1Dd1Nbr29YHr0REvMipZ1/CSae7CAsJ6vVta4YvIuJlPFH24KWFrw9eiYj0Pq8sfH3wSkSk93ll4YuISO9T4YuIBAgVvohIgFDhi4gECBW+iEiAMJZl2Z3hsIwxFUBP7iIUAxzLtZ1dfV1X1jvaOodb3p3xBKDyKDn6wrH+vHt7ezp+x0bH79jGve34pVmWldjpGpZl+e0fYJEnX9eV9Y62zuGWd2ccWGf3z7onP28dPx0/Hb+++Xn7+ymdNzz8uq6sd7R1Dre8u+PeoLez6fj1LR2/Yx/3BkfN5tWndKRrjDHrLMvKsTuHHBsdP9/mS8fP32f4gWKR3QGkR3T8fJvPHD/N8EVEAoRm+CIiAUKFLyISIFT4IiIBQoXvh4wxqcaY140xi40x8+3OI91jjMk2xrxkjHnMGHOJ3Xnk6IwxGcaYp4wxL39rLNIY84wx5gljzBw78/2TCt9HHCzvcmPM1kPGZxhjdhpjcr9V7sOBFZZlXQtk93lY+Z5uHr9zgL9alvUT4Ko+DytA946ZZVl7Lcuad8gmLgJetizrOmBWH8U+IhW+71gCzPj2gDEmCFjINwWRDVxhjMkGNgKXG2NWAR/2cU7p3BK6fvye5Zvj92cgvo9zyr8toevHrDMpQOHB/3Z5KGO3qPB9hGVZa4CqQ4YnALkHZxdtwAvA+cBc4C7Lsk4HzuvbpNKZ7hw/y7LKLcu6EZiPd9yjJSB18+9cZ4r4pvTBS7rWK0LIMRvMv2cQ8M3/YIOBd4CbjTGPA/k25JKu6fT4GWPSjTGLgKXAn21JJodzuGMWf/Dv23hjzG0Hly0HLjbGPIaX3JIh2O4A0iOmkzHLsqytgN7s836HO375wPV9nEW65nDH7ABwwyGDjXzz27bX0AzftxUBQ771dQqw36Ys0n06fr7Hp4+ZCt+3fQlkGWOGGmNCgcuB123OJF2n4+d7fPqYqfB9hDHmeeBT4DhjTJExZp5lWR3ATcBKYDvwkmVZ2+zMKZ3T8fM9/njMdPM0EZEAoRm+iEiAUOGLiAQIFb6ISIBQ4YuIBAgVvohIgFDhi4gECBW+iEiAUOGLiAQIFb6ISID4/y9eEOakVLS5AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def mkplot(N=400, **kwargs):\n",
    "    tmp = ms.compress(imgs, N=N)\n",
    "    vis = ms.mockserve(tmp, [1024,1024])\n",
    "\n",
    "    U = vis.uvd[0]\n",
    "    U = np.linspace(-U/2,U/2,vis.shape[-2],endpoint=False)\n",
    "    plt.loglog(U, abs(vis[0,:,0]), **kwargs)\n",
    "    \n",
    "mkplot(400)\n",
    "mkplot(640,  linestyle='--')\n",
    "mkplot(1280, linestyle=':')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "5a66ce8f-3c76-4659-802c-826a3e1a7175",
   "metadata": {},
   "outputs": [],
   "source": [
    "from copy import copy\n",
    "from math import ceil\n",
    "\n",
    "def truncation_scale(tolerant=0.01, n=2):\n",
    "    return (1/(tolerant*tolerant) - 1)**(0.5/n)\n",
    "\n",
    "def compress(imgs, N=None, eht_f=15e9):\n",
    "    meta = copy(imgs.meta)\n",
    "    fov  = imgs.fov.to(units.rad).value\n",
    "    \n",
    "    spec, U, V = ms.upfft(imgs, *fov, N=N)\n",
    "    \n",
    "    u = np.linspace(-U/2,U/2,spec.shape[-2],endpoint=False)\n",
    "    v = np.linspace( 0,  V/2,spec.shape[-1])\n",
    "    u, v = np.meshgrid(u, v, indexing='ij')\n",
    "    \n",
    "    n  = 2\n",
    "    rr = (u*u + v*v) / (eht_f*eht_f)\n",
    "    Bw = 1 / np.sqrt(1 + rr**n)\n",
    "    \n",
    "    Nu =    spec.shape[-2]\n",
    "    Nv = 2*(spec.shape[-1]-1)\n",
    "    \n",
    "    scale = truncation_scale(n=n)\n",
    "    Ndown = [\n",
    "        2 * ceil(abs(Nu * scale * eht_f / U / 2)),\n",
    "        2 * ceil(abs(Nv * scale * eht_f / V / 2)),\n",
    "    ]\n",
    "    \n",
    "    print(Bw.shape, spec.shape)\n",
    "    \n",
    "    print(Ndown)\n",
    "    \n",
    "    I, W, H = ms.downifft(spec * Bw[np.newaxis,:,:], U, V, N=Ndown)\n",
    "    \n",
    "    meta.width  = abs(W * meta.dist).to(meta.rg)\n",
    "    meta.height = abs(H * meta.dist).to(meta.rg)\n",
    "    return dalt.Image(I, meta=meta)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "910f3871-bc4b-472d-a591-050160025c61",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(400, 201) (100, 400, 201)\n",
      "[146, 146]\n",
      "(700, 351) (100, 700, 351)\n",
      "[256, 256]\n",
      "(1404, 703) (100, 1404, 703)\n",
      "[512, 512]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def mkplot(N=400, **kwargs):\n",
    "    tmp = compress(imgs, N)\n",
    "    vis = ms.mockserve(tmp, [1024,1024])\n",
    "\n",
    "    U = vis.uvd[0]\n",
    "    U = np.linspace(-U/2,U/2,vis.shape[-2],endpoint=False)\n",
    "    plt.loglog(U, abs(vis[0,:,0]), **kwargs)\n",
    "    \n",
    "mkplot(400)\n",
    "mkplot(700,  linestyle='--')\n",
    "mkplot(1404, linestyle=':')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ae9c8e36-9bbd-49eb-a711-42805b799c85",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(700, 351) (100, 700, 351)\n",
      "[256, 256]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='Relative R.A. [$\\\\mathrm{\\\\mu as}$]', ylabel='Relative Declination [$\\\\mathrm{\\\\mu as}$]'>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "tmp = compress(imgs, 700)\n",
    "viz.show(tmp, s=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0aa90dff-097c-4992-8d3a-ca3257265386",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}