Retrieve.ipynb
70.4 KB
搭建倒排表
倒排表的作用是让搜索更加快速，是搜索引擎中常用的技术。根据课程中所讲的方法，你需要完成这部分的代码。
In [2]:
import pandas as pd from tqdm import tqdm import numpy as np import pickle from gensim.models import KeyedVectors # 词向量用来比较俩俩之间相似度
In [3]:
# 读取数据： 导入在preprocessor.ipynb中生成的data/question_answer_pares.pkl文件，并将其保存在变量QApares中 with open('data/question_answer_pares.pkl','rb') as f: QApares = pickle.load(f)
In [4]:
QApares.head()
| question | answer | question_after_preprocessing | |
|---|---|---|---|
| 0 | 买二份有没有少点呀 | 亲亲真的不好意思我们已经是优惠价了呢小本生意请亲谅解 | [买, 二份, 有没有, 少点] |
| 1 | 那就等你们处理喽 | 好的亲退了 | [处理] |
| 2 | 那我不喜欢 | 颜色的话一般茶刀茶针和二合一的话都是红木檀和黑木檀哦 | [喜欢] |
| 3 | 不是免运费 | 本店茶具订单满99包邮除宁夏青海内蒙古海南新疆西藏满39包邮 | [免, 运费] |
| 4 | 好吃吗 | 好吃的 | [好吃] |
TODO1 构造一个倒排表，不需要考虑单词的相似度
In [6]:
# 构建一个倒排表，有关倒排表的详细内容参考实验手册 # 为了能够快速检索，倒排表应用哈希表来存储。python中字典内部便是用哈希表来存储的，所以这里我们直接将倒排表保存在字典中 # 注意：在这里不需要考虑单词之间的相似度。 inverted_list = {} for index,sentence in enumerate(QApares.question_after_preprocessing): ### 你需要完成的代码 for word in sentence: if word in inverted_list: inverted_list[word].add(index) else: inverted_list[word] = set() inverted_list[word].add(index) ### 你需要完成的代码结束
In [24]:
inverted_list["发货"]
Out [24]:
{5, 65541, 32776, 17, 18, 65554, 29, 65566, 32800, 32803, 98339, 32810, 98346, 32818, 55, 98366, 64, 65604, 65611, 32850, 98387, 98398, 65631, 102, 65639, 65640, 65646, 98415, 98416, 118, 122, 65659, 125, 32894, 133, 65669, 65670, 65671, 142, 65679, 32912, 98451, 150, 151, 32929, 65708, 98484, 98489, 187, 32957, 200, 32973, 65742, 98518, 65755, 220, 223, 65764, 65783, 33017, 65786, 254, 65790, 261, 65798, 65810, 275, 98586, 65833, 33068, 65838, 33073, 65843, 65844, 310, 65852, 318, 65862, 65863, 344, 65883, 65885, 350, 33120, 364, 98668, 65903, 33140, 98678, 65913, 33149, 33158, 33163, 33168, 401, 98709, 98715, 33189, 33191, 65960, 33193, 98740, 33210, 98751, 98754, 33220, 453, 98763, 461, 469, 33244, 98794, 495, 98803, 33273, 33276, 66046, 33290, 530, 33300, 66069, 66077, 542, 543, 66081, 66091, 556, 66093, 66094, 98860, 66097, 98877, 66111, 33346, 579, 33347, 588, 66152, 66153, 98932, 66166, 639, 640, 642, 98948, 66185, 651, 33422, 98962, 33427, 98970, 66207, 673, 33441, 33446, 66221, 687, 66224, 33459, 694, 33464, 33466, 33468, 99005, 66238, 99007, 710, 99017, 718, 99044, 743, 33513, 749, 33523, 759, 66299, 99068, 33535, 769, 33539, 66307, 66309, 99084, 66321, 99090, 66323, 99098, 66334, 800, 66337, 33576, 99120, 99122, 33592, 33593, 33600, 33604, 99140, 841, 845, 99149, 66389, 854, 33624, 33625, 858, 66395, 99163, 33629, 99173, 33638, 874, 66411, 878, 33647, 66415, 66417, 33650, 66418, 99182, 891, 33662, 66436, 901, 99208, 66441, 33674, 33675, 33682, 916, 66452, 33691, 33692, 66461, 66463, 33697, 99234, 931, 938, 944, 66480, 33716, 99254, 959, 33728, 99267, 33736, 33741, 978, 33750, 992, 33765, 99301, 1000, 1005, 99309, 33780, 99318, 1017, 33787, 66557, 1024, 1027, 1034, 1036, 66577, 66580, 99351, 99360, 1057, 33825, 99363, 33831, 66600, 33837, 66607, 99376, 33841, 99380, 66613, 33849, 66622, 1087, 1088, 99391, 1095, 1098, 66637, 1102, 99405, 99410, 99413, 66646, 99419, 66652, 1118, 99431, 1129, 66670, 33912, 33914, 66685, 33927, 99467, 99470, 99472, 66705, 66708, 66709, 99484, 99485, 99487, 66721, 33954, 1187, 99495, 66728, 66732, 66734, 99510, 99517, 66751, 1223, 33992, 99528, 99537, 66770, 99539, 1237, 1242, 66779, 66780, 1247, 99555, 34024, 1265, 34034, 1270, 34042, 1277, 66814, 99583, 34064, 1297, 66834, 66844, 1312, 66849, 1326, 66866, 34099, 99637, 1334, 66871, 66877, 1342, 1345, 99651, 66887, 1352, 1354, 66895, 99671, 66909, 99681, 1392, 66929, 34162, 1401, 1402, 34169, 99709, 1407, 66944, 66945, 1412, 1415, 34184, 1420, 99729, 1426, 34196, 66971, 1436, 66974, 1443, 34217, 66986, 1452, 66993, 1459, 34227, 66996, 34258, 99797, 99803, 1508, 99814, 1512, 1515, 67053, 99821, 99824, 99825, 34303, 67071, 34307, 67077, 67079, 99854, 67092, 67094, 99862, 1565, 34333, 1567, 34338, 1579, 1585, 67127, 99897, 1603, 1604, 67147, 67152, 34388, 67156, 99924, 1623, 34395, 99931, 1635, 99946, 67180, 99949, 99954, 34419, 99957, 67197, 34434, 99970, 34440, 1673, 1675, 1676, 34443, 67212, 67216, 34457, 1691, 1692, 1697, 34473, 67241, 34480, 67250, 67269, 1743, 67284, 34528, 1775, 34553, 67323, 1798, 34572, 1805, 34576, 67344, 67350, 1820, 1821, 34590, 1825, 34606, 1839, 67375, 67379, 1852, 1861, 1866, 67402, 1874, 67416, 67431, 1904, 67441, 34687, 1920, 34689, 34701, 34708, 34710, 67492, 67493, 1970, 34744, 34753, 67531, 2004, 2008, 2009, 2023, 67560, 34793, 67562, 34795, 2037, 34807, 2046, 34819, 2052, 34827, 2068, 67613, 34848, 67626, 67628, 34864, 2099, 67636, 34869, 67641, 2110, 67649, 2117, 67655, 67666, 67669, 67680, 34914, 67682, 2151, 34926, 2160, 2161, 34929, 67699, 2170, 2178, 34949, 67718, 2187, 2193, 67730, 67735, 2216, 2218, 2230, 34998, 2235, 67771, 2244, 2247, 67791, 2263, 67800, 2266, 35037, 67810, 67816, 35049, 35055, 67824, 67825, 35059, 35063, 2296, 35067, 2323, 2326, 2330, 67867, 2335, 67886, 2365, 2370, 2372, 2377, 67915, 2380, 2392, 67928, 67934, 35174, 35180, 35188, 67958, 35195, 67968, 35202, 2441, 35211, 67981, 35215, 67984, 35219, 35220, 35228, 2465, 2483, 2484, 35256, 68025, 2510, 35280, 35281, 68066, 2532, 35310, 68084, 2553, 68089, 68097, 2566, 35351, 35358, 2594, 2607, 35378, 2612, 68151, 35385, 2620, 35394, 35395, 68163, 68164, 68169, 35415, 2653, 68199, 68200, 68209, 2677, 68215, 35458, 35459, 2692, 35466, 35472, 2705, 35473, 35474, 68240, 2710, 2712, 35481, 2715, 2721, 68257, 68264, 68265, 35503, 2744, 68290, 2761, 68306, 35539, 35547, 35549, 2786, 35557, 35559, 68329, 68332, 68334, 68337, 35570, 2804, 68343, 35587, 2824, 35603, 2838, 68375, 35613, 2853, 35622, 35634, 2868, 35636, 68408, 68419, 35654, 2887, 68440, 68445, 2916, 35689, 68457, 35697, 35729, 2968, 68506, 35740, 68512, 2981, 35758, 2991, 35759, 2996, 3000, 68537, 3009, 68549, 68556, 35792, 35793, 68562, 68568, 35804, 35809, 35810, 68578, 3050, 3053, 68590, 3061, 3067, 35835, 35843, 3076, 68613, 35846, 3081, 35855, 68629, 3094, 35863, 35867, 35888, 35899, 35915, 68683, 68685, 3151, 68687, 68690, 68697, 35932, 68705, 3171, 68708, 68711, 68721, 3187, 35974, 68744, 3210, 3211, 35983, 68761, 36002, 3236, 68774, 3240, 68790, 3256, 68792, 3265, 3269, 3271, 68809, 3275, 36052, 3285, 3286, 68823, 3288, 3291, 3293, 3303, 68855, 68865, 3333, 36107, 3356, 3364, 3375, 68914, 68930, 68948, 3414, 36192, 36197, 36201, 3434, 36211, 36214, 3455, 36226, 69014, 69015, 3482, 3483, 69022, 3496, 3500, 36273, 36277, 3526, 3527, 36294, 36300, 3536, 3539, 36308, 3542, 3543, 69082, 36320, 36321, 36326, 3560, 3572, 36340, 36341, 36346, 69124, 3590, 36370, 3610, 3613, 69158, 3627, 3636, 36404, 36416, 36419, 3652, 3655, 3657, 3669, 3673, 36444, 36446, 3680, 69220, 69221, 69231, 36469, 69244, 3712, 3713, 36481, 69251, 36488, 69264, 36501, 36509, 36510, 3744, 69282, 69287, 69294, 36527, 36528, 3764, 3785, 3791, 69338, 36572, 3807, 69344, 3815, 69365, 36598, 3832, 3837, 36619, 69405, 3882, 3883, 36650, 69426, 36687, 3928, 3931, 69474, 36707, 69477, 36710, 3945, 36723, 3960, 69512, 3980, 69516, 69518, 69519, 69523, 36756, 36762, 36765, 36771, 36778, 4026, 69563, 69569, 69578, 69596, 36829, 4069, 36838, 69607, 36841, 4074, 36845, 69617, 4093, 36871, 36874, 36878, 36880, 36888, 36890, 4124, 69662, 4127, 36898, 4133, 36916, 36917, 36924, 69692, 69695, 4161, 4162, 4165, 69707, 4175, 36948, 4182, 36958, 69732, 36974, 36978, 69748, 4220, 36995, 69766, 36999, 69767, 69770, 4238, 37007, 69783, 4250, 69800, 69802, 4267, 37036, 37037, 37039, 4278, 37049, 4282, 4287, 4290, 4291, 37064, 4301, 37072, 37073, 69847, 4312, 4321, 4323, 4324, 4332, 4336, 69874, 69880, 37114, 37120, 4354, 4360, 69898, 37131, 69903, 37150, 4387, 69923, 69928, 37166, 69934, 69944, 37177, 4425, 37193, 4429, 37216, 4463, 37234, 37235, 70002, 4492, 4493, 4500, 70040, ...}
In [11]:
len(inverted_list)
Out [11]:
3832
In [7]:
#d ata/retrieve/sgns.zhihu.word是从https://github.com/Embedding/Chinese-Word-Vectors下载到的预训练好的中文词向量文件 #使 用KeyedVectors.load_word2vec_format()函数加载预训练好的词向量文件 model = KeyedVectors.load_word2vec_format('data/retrieve/sgns.zhihu.word')
In [8]:
def get_similar_by_word(word,topk): ''' 返回与一个单词word相似度最高的topk个单词所组成的单词列表 出参： word_list：与word相似度最高的topk个单词所组成的单词列表。格式为[单词1，单词2，单词3，单词4，单词5] ''' similar_words = model.similar_by_word(word,topk) word_list = [word[0] for word in similar_words] return word_list
In [9]:
get_similar_by_word("今天",5)
Out [9]:
['昨天', '现在', '今天下午', '明天', '今日']
TODO2 构造一个新的倒排表，考虑单词之间的语义相似度
In [25]:
# TODO： # 构造一个新的倒排表，并将结果保存在字典inverted_list_new中 # 新的倒排表键为word，值为老倒排表[word]、老倒排表[单词1]、老倒排表[单词2]、老倒排表[单词3]、老倒排表[单词4]的并集 # 即新倒排表保存了包含单词word或包含与单词word最相近的5个单词中的某一个的问题的index inverted_list_new = {} OOV_count = 0 for word in tqdm(inverted_list): ### 你需要完成的部分 try: top_4_words = get_similar_by_word(word,4) inverted_list_new[word] = set() inverted_list_new[word] = inverted_list_new[word].union(inverted_list[word]) for t_word in top_4_words: if t_word in inverted_list: inverted_list_new[word] = inverted_list_new[word].union(inverted_list[t_word]) except Exception as e: OOV_count += 1 print("OOV_count:",OOV_count) ### 你需要完成的代码结束
Out [25]:
100%|██████████| 3832/3832 [00:44<00:00, 85.74it/s]
In [26]:
inverted_list_new["发货"]
Out [26]:
{81920, 16386, 5, 65541, 81927, 32776, 81930, 81935, 17, 18, 65554, 16401, 81947, 98331, 29, 65566, 32800, 81953, 32803, 98339, 81959, 32810, 98346, 49194, 32818, 16435, 55, 49209, 98366, 64, 49219, 65604, 81988, 16458, 65611, 81995, 81998, 16463, 16464, 49233, 32850, 98387, 49234, 82004, 86, 81999, 98386, 16475, 32859, 49245, 98398, 65631, 82015, 65630, 102, 65639, 65640, 49259, 65646, 98415, 98416, 49263, 16495, 49267, 16500, 82035, 118, 16503, 65650, 65656, 122, 65659, 49275, 125, 32894, 65660, 133, 65669, 65670, 65671, 32902, 139, 49293, 142, 65679, 32912, 98451, 150, 151, 32929, 49318, 49320, 65708, 82092, 82093, 65711, 98484, 98489, 49337, 187, 49340, 32957, 200, 16588, 32973, 65742, 16589, 98518, 16598, 49366, 82137, 65755, 220, 223, 65764, 82149, 82155, 16621, 49396, 65783, 33017, 65786, 254, 65790, 82176, 261, 65798, 65806, 65810, 275, 279, 98586, 82208, 49442, 65833, 33068, 82220, 65838, 82221, 82224, 33073, 65843, 65844, 16691, 310, 16696, 82234, 65852, 49468, 318, 49471, 49472, 16706, 49475, 65862, 65863, 16712, 82248, 65866, 49483, 49493, 16726, 344, 16730, 65883, 82268, 65885, 350, 33120, 16745, 364, 98668, 65903, 33140, 98678, 65913, 33149, 49535, 33158, 49544, 82313, 33163, 16783, 33168, 401, 82322, 98709, 49558, 98715, 16796, 49565, 82333, 82334, 33189, 33191, 65960, 33193, 49579, 16812, 98740, 65973, 16822, 82359, 16825, 33210, 16827, 82365, 98751, 16832, 98754, 33220, 453, 49604, 98763, 461, 469, 49623, 16856, 33244, 49629, 16867, 16869, 98794, 82410, 82412, 495, 16882, 98803, 49651, 49656, 33273, 16889, 49658, 33276, 82426, 66046, 507, 16895, 49669, 82437, 33290, 49674, 16911, 530, 33300, 66069, 16918, 16922, 66077, 542, 543, 82463, 66081, 16932, 66091, 556, 66093, 66094, 98860, 82475, 66097, 49709, 16942, 49715, 49716, 66106, 98877, 66111, 49729, 33346, 579, 33347, 82499, 49735, 16967, 49739, 588, 16983, 82522, 16991, 82528, 49761, 49762, 66151, 66152, 66153, 17003, 49777, 98932, 17012, 66166, 17020, 17021, 639, 640, 49793, 642, 17026, 98948, 17027, 49796, 82563, 17030, 66185, 82566, 651, 33422, 82576, 98962, 33427, 17043, 49812, 82584, 98970, 17050, 17052, 66207, 82592, 673, 33441, 17057, 17061, 33446, 49831, 66221, 82605, 687, 66224, 49841, 33453, 33459, 17076, 66228, 694, 33464, 33466, 33468, 99005, 66238, 99007, 99006, 82628, 710, 82630, 82632, 99017, 711, 718, 82639, 49872, 82645, 82652, 49889, 82658, 99044, 743, 17127, 33513, 49897, 17132, 749, 49901, 49903, 82674, 33523, 17142, 759, 17144, 33527, 66299, 99068, 33535, 17152, 769, 33539, 66307, 66309, 82694, 17162, 99084, 49932, 17167, 49935, 66321, 99090, 66323, 783, 82713, 99098, 66333, 66334, 82719, 800, 66337, 17185, 49954, 17190, 33576, 49962, 17196, 99120, 99122, 33587, 49972, 49974, 17207, 33592, 33593, 33600, 33604, 99140, 82757, 841, 49994, 845, 99149, 17235, 82771, 66389, 854, 82775, 33624, 33625, 858, 66395, 99163, 33629, 17245, 82782, 17249, 17251, 82787, 99173, 33638, 17255, 874, 66411, 82794, 17259, 878, 33647, 66415, 66417, 33650, 66418, 99182, 82798, 33652, 891, 17275, 82812, 33662, 82813, 17283, 66436, 901, 99208, 66441, 33674, 33675, 17291, 82829, 33682, 916, 66452, 17304, 33688, 33691, 33692, 66461, 66463, 82847, 33697, 99234, 931, 82848, 17317, 938, 17324, 82863, 944, 66480, 33716, 99254, 82878, 959, 33728, 99267, 33736, 33741, 17359, 978, 17364, 33750, 82905, 17371, 992, 17378, 33765, 99301, 1000, 50152, 50155, 1005, 99309, 50158, 82928, 66542, 33780, 99318, 1017, 33787, 66557, 1024, 17408, 17409, 1027, 50179, 82949, 82950, 1034, 50187, 1036, 50191, 66577, 66580, 99351, 82969, 82970, 17437, 99360, 1057, 33825, 99363, 82977, 82979, 33831, 66600, 17447, 17448, 33837, 50221, 66607, 99376, 33841, 82993, 99380, 66613, 82996, 82998, 33849, 83003, 66622, 1087, 1088, 99391, 17473, 83011, 50245, 1095, 1098, 66637, 1102, 99405, 50256, 99410, 66644, 99413, 66646, 99419, 66652, 83037, 1118, 83040, 99431, 66663, 1129, 17516, 83052, 66670, 33900, 50290, 17524, 33912, 83065, 33914, 66685, 83071, 17537, 50309, 33927, 50314, 99467, 99470, 50318, 99472, 66705, 83088, 66708, 66709, 50327, 99484, 99485, 83101, 99487, 66721, 33954, 1187, 83105, 50337, 50341, 99495, 66728, 50345, 50346, 17578, 66732, 83116, 66734, 50351, 50356, 99510, 17591, 50363, 99517, 66751, 17599, 1223, 33992, 99528, 17607, 50376, 50379, 17616, 99537, 66770, 99539, 1237, 1242, 66779, 66780, 17628, 1247, 17631, 83168, 50402, 99555, 50405, 34024, 50409, 50411, 1265, 34034, 83186, 17651, 50419, 1270, 34042, 99580, 1277, 66814, 99583, 83200, 17671, 50441, 17677, 34064, 1297, 66834, 50448, 50449, 50456, 66844, 83229, 83231, 1312, 66849, 50464, 17698, 17699, 83236, 17702, 1326, 99632, 17713, 66866, 34099, 83252, 99637, 1334, 66871, 83258, 66877, 1342, 50493, 17728, 1345, 99651, 66887, 1352, 50503, 1354, 17737, 50508, 66895, 99671, 66909, 34143, 99681, 83300, 17770, 1392, 66929, 34162, 17777, 1401, 1402, 34169, 50554, 99709, 50557, 1407, 66944, 66945, 17788, 1412, 1415, 34184, 1420, 99729, 1426, 34196, 83351, 66971, 1436, 83357, 66974, 17821, 1443, 83363, 50598, 34217, 66986, 83369, 1452, 66993, 50609, 1459, 34227, 66996, 83385, 50618, 50626, 50627, 83406, 83407, 34258, 50643, 83412, 99797, 50647, 99803, 50652, 50654, 1508, 50661, 99814, 1512, 1515, 50667, 67053, 99821, 17901, 99824, 99825, 83438, 50672, 83441, 50677, 17910, 83446, 83447, 83448, 34292, 17916, 17918, 34303, 67071, 34307, 83460, 67077, 1540, 67079, 34310, 99854, 83473, 67092, 67094, 99862, 67100, 1565, 34333, 1567, 34338, 99877, 1579, 83502, 83504, 1585, 67127, 99897, 50753, 83521, 1603, 1604, 17991, 17992, 50760, 83527, 67147, 17996, 67152, 34388, 67156, 99924, 1623, 34395, 99931, 18016, 1635, 99946, 67180, 99949, 99954, 34419, 83571, 99957, 83573, 83572, 18043, 50811, 67197, 34434, 99970, 18053, 83590, 34440, 1673, 50826, 1675, 1676, 34443, 67212, 67216, 18065, 18066, 18068, 18069, 50838, 50839, 34457, 50841, 1691, 1692, 83615, 1697, 50855, 34473, 67241, 50861, 34480, 67250, 18104, 18108, 83645, 18112, 18116, 67269, 1743, 83667, 67284, 50904, 83674, 50910, 34528, 34529, 18146, 50917, 83688, 50923, 1775, 18160, 18168, 34553, 67322, 67323, 18171, 50939, 67327, 50944, 50947, 1798, 83718, 34572, 1805, 83724, 34576, 67344, 83730, 67350, 1820, 1821, 34590, 50972, 1825, 50986, 50988, 34606, 1839, 67375, 50990, 50993, 67379, 83767, 51000, 1852, 34620, 83774, 18239, 1861, 18245, 1866, 67402, 1874, 83795, 83799, 67416, 18270, 51039, 83807, 83810, 51043, 51046, 67431, 51047, 1904, 67441, 18289, 34687, 1920, 34689, 51071, 51074, 1919, 83846, 18314, 51084, 34701, 18319, 34708, 34710, 18326, 51095, 67492, 67493, 51108, 1959, 34725, 18348, 67500, 83887, 1970, 83895, 34744, 51128, 83898, 83902, 34753, 83906, 51140, 18373, 67531, 18384, 2004, 51159, 2008, 2009, 18397, 18401, 83940, 2023, 67560, 34793, 67562, 34795, 51175, 83951, 51184, 83953, 51185, ...}
In [27]:
# 将新的倒排表保存在文件data/retrieve/invertedList.pkl中 with open('data/retrieve/invertedList.pkl','wb') as f: pickle.dump(inverted_list_new,f)
以下为测试，完成上述过程之后，可以运行以下的代码来测试准确性。
In [28]:
#这一格的内容是从preprocessor.ipynb中粘贴而来，包含了数据预处理的几个关键函数 import emoji import re import jieba def clean(content): content = emoji.demojize(content) content = re.sub('<.*>','',content) return content #这一函数是用于对句子进行分词，在preprocessor.ipynb中由于数据是已经分好词的，所以我们并没有进行这一步骤，但是对于一个新的问句，这一步是必不可少的 def question_cut(content): return list(jieba.cut(content)) def strip(wordList): return [word.strip() for word in wordList if word.strip()!=''] with open("data/stopWord.json","r") as f: stopWords = f.read().split("\n") def rm_stop_word(wordList): return [word for word in wordList if word not in stopWords]
In [29]:
# 从data/retrieve/invertedList.pkl加载倒排表并将其保存在变量invertedList中 with open('data/retrieve/invertedList.pkl','rb') as f: invertedList = pickle.load(f)
In [30]:
def get_retrieve_result(sentence): ''' 输入一个句子sentence，根据倒排表进行快速检索，返回与该句子较相近的一些候选问题的index 候选问题由包含该句子中任一单词或包含与该句子中任一单词意思相近的单词的问题索引组成 ''' sentence = clean(sentence) sentence = question_cut(sentence) sentence = strip(sentence) sentence = rm_stop_word(sentence) candidate = set() for word in sentence: if word in invertedList: candidate = candidate | invertedList[word] return candidate
In [31]:
get_retrieve_result('什么时候发货') # 通过倒排表返回文档IDs
Out [31]:
{81920, 16386, 65541, 5, 81927, 32776, 81930, 81935, 17, 18, 65554, 16401, 98331, 81947, 29, 65566, 32800, 81953, 32803, 98339, 81959, 32810, 98346, 49194, 32818, 16435, 55, 49209, 98366, 64, 49219, 65604, 81988, 16458, 65611, 81995, 81998, 16463, 16464, 49233, 32850, 98387, 98386, 49234, 86, 81999, 82004, 32859, 16475, 49245, 98398, 65631, 65630, 82015, 102, 65639, 65640, 49259, 65646, 98415, 98416, 49263, 65650, 16495, 49267, 16500, 118, 82035, 65656, 16503, 122, 65659, 65660, 125, 32894, 49275, 133, 65669, 65670, 65671, 32902, 139, 49293, 142, 65679, 32912, 98451, 150, 151, 32929, 49318, 49320, 65708, 82092, 82093, 65711, 98484, 98489, 49337, 187, 49340, 32957, 200, 16588, 32973, 65742, 16589, 98518, 16598, 49366, 82137, 65755, 220, 223, 65764, 82149, 82155, 16621, 49396, 65783, 33017, 65786, 254, 65790, 82176, 261, 65798, 65806, 65810, 275, 279, 98586, 82208, 49442, 65833, 33068, 82220, 65838, 82221, 82224, 33073, 65843, 65844, 16691, 310, 16696, 82234, 65852, 49468, 318, 49471, 49472, 16706, 49475, 65862, 65863, 16712, 82248, 65866, 49483, 49493, 16726, 344, 16730, 65883, 82268, 65885, 350, 33120, 16745, 364, 98668, 65903, 33140, 98678, 65913, 33149, 49535, 33158, 49544, 82313, 33163, 16783, 33168, 401, 82322, 98709, 49558, 98715, 16796, 49565, 82333, 82334, 33189, 33191, 65960, 33193, 49579, 16812, 98740, 65973, 16822, 82359, 16825, 33210, 16827, 82365, 98751, 16832, 98754, 33220, 453, 49604, 98763, 461, 469, 49623, 16856, 33244, 49629, 16867, 16869, 98794, 82410, 82412, 495, 16882, 98803, 49651, 49656, 33273, 16889, 507, 33276, 82426, 66046, 49658, 16895, 49669, 82437, 33290, 49674, 16911, 530, 33300, 66069, 16918, 16922, 66077, 542, 543, 82463, 66081, 16932, 66091, 556, 66093, 66094, 98860, 33324, 66097, 82475, 49709, 16942, 49715, 49716, 66106, 98877, 66111, 49729, 33346, 579, 33347, 82499, 49735, 16967, 49739, 588, 16983, 82522, 16991, 82528, 49761, 49762, 66151, 66152, 66153, 17003, 49777, 98932, 17012, 66166, 17020, 17021, 639, 640, 49793, 642, 17026, 98948, 17027, 82563, 49796, 17030, 66185, 82566, 651, 33422, 82576, 98962, 33427, 17043, 49812, 82584, 98970, 17050, 17052, 66207, 82592, 673, 33441, 17057, 17061, 33446, 49831, 66221, 33453, 687, 66224, 82605, 49841, 33459, 66228, 17076, 694, 33464, 33466, 33468, 99005, 66238, 99007, 99006, 82628, 710, 711, 82630, 99017, 82632, 718, 82639, 49872, 82645, 82652, 49889, 82658, 99044, 743, 17127, 33513, 49897, 17132, 749, 49901, 49903, 82674, 33523, 17142, 759, 33527, 17144, 66299, 99068, 33535, 17152, 769, 33539, 66307, 66309, 82694, 17162, 99084, 49932, 783, 17167, 66321, 99090, 66323, 49935, 82713, 99098, 66333, 66334, 82719, 800, 66337, 17185, 49954, 17190, 33576, 49962, 17196, 99120, 99122, 33587, 49972, 49974, 17207, 33592, 33593, 33600, 33604, 99140, 82757, 841, 49994, 845, 99149, 17235, 82771, 66389, 854, 82775, 33624, 33625, 858, 66395, 99163, 33629, 17245, 82782, 17249, 17251, 82787, 99173, 33638, 17255, 874, 66411, 82794, 17259, 878, 33647, 66415, 66417, 33650, 66418, 99182, 33652, 82798, 891, 17275, 82812, 33662, 82813, 17283, 66436, 901, 99208, 66441, 33674, 33675, 17291, 82829, 33682, 916, 66452, 33688, 17304, 33691, 33692, 66461, 66463, 82847, 33697, 99234, 931, 82848, 17317, 938, 17324, 82863, 944, 66480, 33716, 99254, 82878, 959, 33728, 99267, 33736, 33741, 17359, 978, 17364, 33750, 82905, 17371, 992, 17378, 33765, 99301, 1000, 50152, 50155, 1005, 99309, 66542, 50158, 82928, 33780, 99318, 1017, 33787, 66557, 1024, 17408, 17409, 1027, 50179, 82949, 82950, 1034, 50187, 1036, 50191, 66577, 66580, 99351, 82969, 82970, 17437, 99360, 1057, 33825, 99363, 82977, 82979, 33831, 66600, 17447, 17448, 33837, 50221, 66607, 99376, 33841, 82993, 99380, 66613, 82996, 82998, 33849, 83003, 66622, 1087, 1088, 99391, 17473, 83011, 50245, 1095, 1098, 66637, 1102, 99405, 50256, 99410, 66644, 99413, 66646, 99419, 66652, 83037, 1118, 83040, 99431, 66663, 1129, 33900, 17516, 66670, 83052, 50290, 17524, 33912, 83065, 33914, 66685, 83071, 17537, 50309, 33927, 50314, 99467, 99470, 50318, 99472, 66705, 83088, 66708, 66709, 50327, 99484, 99485, 83101, 99487, 66721, 33954, 1187, 83105, 50337, 50341, 99495, 66728, 50345, 50346, 17578, 66732, 83116, 66734, 50351, 50356, 99510, 17591, 50363, 99517, 66751, 17599, 1223, 33992, 99528, 17607, 50376, 50379, 17616, 99537, 66770, 99539, 1237, 1242, 66779, 66780, 17628, 1247, 17631, 83168, 50402, 99555, 50405, 34024, 50409, 50411, 1265, 34034, 83186, 17651, 50419, 1270, 34042, 99580, 1277, 66814, 99583, 83200, 17671, 50441, 17677, 34064, 1297, 66834, 50448, 50449, 50456, 66844, 83229, 83231, 1312, 66849, 50464, 17698, 17699, 83236, 17702, 1326, 99632, 17713, 66866, 34099, 83252, 99637, 1334, 66871, 83258, 66877, 1342, 50493, 17728, 1345, 99651, 66887, 1352, 50503, 1354, 17737, 50508, 66895, 99671, 66909, 34143, 99681, 83300, 17770, 1392, 66929, 34162, 17777, 1401, 1402, 34169, 50554, 99709, 17788, 1407, 66944, 66945, 50557, 1412, 1415, 34184, 1420, 99729, 1426, 34196, 83351, 66971, 1436, 83357, 66974, 17821, 1443, 83363, 50598, 34217, 66986, 83369, 1452, 66993, 50609, 1459, 34227, 66996, 83385, 50618, 50626, 50627, 83406, 83407, 34258, 50643, 83412, 99797, 50647, 99803, 50652, 50654, 1508, 50661, 99814, 1512, 1515, 50667, 67053, 99821, 17901, 99824, 99825, 83438, 50672, 34292, 83441, 50677, 17910, 83446, 83447, 83448, 17916, 17918, 34303, 67071, 34307, 1540, 67077, 34310, 67079, 83460, 99854, 83473, 67092, 67094, 99862, 67100, 1565, 34333, 1567, 34338, 99877, 1579, 83502, 83504, 1585, 67127, 99897, 50753, 83521, 1603, 1604, 17991, 17992, 50760, 83527, 67147, 17996, 67152, 34388, 67156, 99924, 1623, 34395, 99931, 18016, 1635, 99946, 67180, 99949, 99954, 34419, 83571, 99957, 83572, 83573, 18043, 50811, 67197, 34434, 99970, 18053, 83590, 34440, 1673, 50826, 1675, 1676, 34443, 67212, 67216, 18065, 18066, 18068, 18069, 50838, 50839, 34457, 50841, 1691, 1692, 83615, 1697, 50855, 34473, 67241, 50861, 34480, 67250, 18104, 18108, 83645, 18112, 18116, 67269, 1743, 83667, 67284, 50904, 83674, 50910, 34528, 34529, 18146, 50917, 83688, 50923, 1775, 18160, 18168, 34553, 67322, 67323, 18171, 50939, 67327, 50944, 50947, 1798, 83718, 34572, 1805, 83724, 34576, 67344, 83730, 67350, 1820, 1821, 34590, 50972, 1825, 50986, 50988, 34606, 1839, 67375, 50990, 50993, 67379, 83767, 51000, 1852, 34620, 83774, 18239, 1861, 18245, 1866, 67402, 1874, 83795, 83799, 67416, 18270, 51039, 83807, 83810, 51043, 51046, 67431, 51047, 1904, 67441, 18289, 34687, 1920, 34689, 1919, 51071, 51074, 83846, 18314, 51084, 34701, 18319, 34708, 34710, 18326, 51095, 67492, 67493, 34725, 1959, 51108, 67500, 18348, 83887, 1970, 83895, 34744, 51128, 83898, 83902, 34753, 83906, 51140, 18373, 67531, 18384, 2004, 51159, 2008, 2009, 18397, 18401, 83940, 2023, 67560, 34793, 67562, 34795, 51175, 83951, 51184, 83953, ...}