commit solved problems to repository and code for dumping in analsysi.ipynb

Author: igorpejic

analsysis.ipynb

@@ -122,6 +122,50 @@
     "best_their_ids = get_best_problem_ids(guillotine=False)"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 37,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "'''\n",
+    "Output just problems to csvs\n",
+    "'''\n",
+    "g_problems = Result.objects.filter(problem_id__in=best_problem_ids, n_simulations=5000, \n",
+    "                                 improved_sel=True, strategy='max_depth', score__isnull=False\n",
+    "                                 ).values('rows', 'cols', 'tiles', 'n_tiles')\n",
+    "\n",
+    "b_problems = Result.objects.filter(their_id__in=best_their_ids, n_simulations=5000, \n",
+    "                                 improved_sel=True, strategy='max_depth', score__isnull=False\n",
+    "                                 ).values('rows', 'cols', 'tiles', 'n_tiles', 'their_id')\n",
+    "\n",
+    "def eliminate_duplicate_tiles(tiles):\n",
+    "    i = 0\n",
+    "    new_tiles = []\n",
+    "    while i < len(tiles):\n",
+    "        tile = tiles[i]\n",
+    "        tiles_without_current_tile = tiles[i + 1:]\n",
+    "        tile_pair_index = tiles_without_current_tile.index([tile[1], tile[0]]) + i + 1\n",
+    "        tiles = tiles[0: tile_pair_index] + tiles[tile_pair_index + 1:]\n",
+    "        i += 1 \n",
+    "    return tiles\n",
+    "        \n",
+    "import csv, os\n",
+    "for file_name, problems in [('problems_g.csv', g_problems), ('problems_b.csv', b_problems)]:\n",
+    "    with open(os.path.join('problems/', file_name), 'w') as csv_file:\n",
+    "        fieldnames = ['rows', 'cols', 'tiles', 'n_tiles']\n",
+    "        if file_name == 'problems_b.csv':\n",
+    "            fieldnames.append('external_id')\n",
+    "        csv_writer = csv.writer(csv_file, delimiter=';')\n",
+    "        csv_writer.writerow(fieldnames)\n",
+    "        for problem in problems:\n",
+    "            row = [problem['rows'], problem['cols'],\n",
+    "                   str(eliminate_duplicate_tiles(problem['tiles'])), problem['n_tiles']]\n",
+    "            if file_name == 'problems_b.csv':\n",
+    "                row.append(problem['their_id'])\n",
+    "            csv_writer.writerow(row)"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 65,
@@ -181,6 +225,39 @@
     "plt.xlabel('columns')"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 76,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Count of squares in problems: defaultdict(<class 'int'>, {(1, 1): 5534, (4, 4): 242, (2, 2): 498, (8, 8): 42, (6, 6): 86, (3, 3): 132, (5, 5): 70, (7, 7): 34, (14, 14): 22, (20, 20): 2, (9, 9): 38, (10, 10): 16, (13, 13): 8, (16, 16): 16, (12, 12): 4, (24, 24): 2, (11, 11): 6, (15, 15): 8, (21, 21): 2, (22, 22): 2, (18, 18): 2, (17, 17): 2})\n"
+     ]
+    }
+   ],
+   "source": [
+    "tiles = Result.objects.filter(problem_id__in=best_problem_ids, n_simulations=5000, \n",
+    "                                 improved_sel=True, strategy='max_depth', score__isnull=False\n",
+    "                                 ).values('tiles')\n",
+    "\n",
+    "from collections import defaultdict\n",
+    "tiles_count = defaultdict(int)\n",
+    "for tile_dict in tiles:\n",
+    "    _tiles = tile_dict['tiles']\n",
+    "    for one_tile in _tiles:\n",
+    "        if one_tile[0] == one_tile[1]:\n",
+    "            tiles_count[(one_tile[0], one_tile[1])] += 1\n",
+    "            \n",
+    "print('Count of squares in problems:', tiles_count)\n",
+    "\n",
+    "            \n",
+    "    \n",
+    "    "
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 7,