{
"cells": [
{
"cell_type": "markdown",
"id": "4ebeeae6",
"metadata": {
"toc": true
},
"source": [
"Содержание
\n",
""
]
},
{
"cell_type": "markdown",
"id": "12444a0f",
"metadata": {},
"source": [
"# collect texts"
]
},
{
"cell_type": "code",
"execution_count": 12,
"id": "f3ddc049",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-15T22:41:10.093684Z",
"start_time": "2024-06-15T22:41:10.089042Z"
}
},
"outputs": [],
"source": [
"import html\n",
"import os\n",
"import re\n",
"import shutil\n",
"from bs4 import BeautifulSoup"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "9e0e1c77",
"metadata": {},
"outputs": [],
"source": [
"!pip install razdel"
]
},
{
"cell_type": "code",
"execution_count": 10,
"id": "9470fce5",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-15T22:40:47.962801Z",
"start_time": "2024-06-15T22:40:47.854033Z"
}
},
"outputs": [],
"source": [
"from razdel import sentenize\n",
"from tqdm import tqdm"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "68b54deb",
"metadata": {},
"outputs": [],
"source": [
"!git clone https://github.com/tolstoydigital/TEI.git"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "0d9b57a0",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-15T22:38:10.450986Z",
"start_time": "2024-06-15T22:38:10.446063Z"
}
},
"outputs": [],
"source": [
"relevant_dirs = ['diaries', 'letters', 'notes', 'works']"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "375755d6",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-15T22:38:26.335811Z",
"start_time": "2024-06-15T22:38:25.268500Z"
}
},
"outputs": [],
"source": [
"path = 'TEI/reference/bibllist_works.xml'\n",
"xml = open(path).read()\n",
"soup = BeautifulSoup(xml, features=\"xml\")"
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "f69acb38",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-15T22:38:48.857463Z",
"start_time": "2024-06-15T22:38:48.668146Z"
}
},
"outputs": [],
"source": [
"group_texts = {}\n",
"for it in soup.find_all(\"item\"):\n",
" ref = it.find(\"ref\")\n",
" for related in it.find_all(\"relatedItem\"):\n",
" for ref_ana in related.find_all(\"ref\"):\n",
" group_texts[ref_ana.text] = ref.text"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "182964df",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-15T22:39:31.844575Z",
"start_time": "2024-06-15T22:39:31.796678Z"
}
},
"outputs": [],
"source": [
"prefix_texts = 'extracted_texts'\n",
"os.mkdir(prefix_texts)"
]
},
{
"cell_type": "code",
"execution_count": 13,
"id": "f75563a7",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-15T22:48:24.143625Z",
"start_time": "2024-06-15T22:41:13.138087Z"
}
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|██████████| 4584/4584 [01:26<00:00, 53.24it/s] \n",
"100%|██████████| 9087/9087 [02:26<00:00, 61.89it/s] \n",
"100%|██████████| 100/100 [00:28<00:00, 3.49it/s]\n",
"100%|██████████| 767/767 [02:49<00:00, 4.53it/s]\n"
]
}
],
"source": [
"if os.path.exists(prefix_texts):\n",
" shutil.rmtree(prefix_texts)\n",
"os.mkdir(prefix_texts)\n",
"\n",
"complex_texts = {}\n",
"for rel_dir in relevant_dirs:\n",
" path = os.path.join('TEI/texts', rel_dir)\n",
" for file in tqdm(sorted(os.listdir(path))):\n",
" fiction = 0\n",
" if not file.endswith('.xml'):\n",
" continue\n",
" xml = open(os.path.join(path, file)).read()\n",
" if 'Печатные варианты' in xml:\n",
" continue\n",
" nameID = file.replace('.xml', '')\n",
" soup = BeautifulSoup(xml, features=\"xml\")\n",
" if soup.find(\"catRef\", {\"ana\":\"#fiction\"}):\n",
" fiction = 1\n",
" s = soup.find(\"body\")\n",
" paragraphs = []\n",
" for erase in s.find_all([\"orig\", \"comments\", \"sic\", \"note\"]):\n",
" erase.decompose()\n",
" for p in s.find_all([\"p\", \"l\"]):\n",
" paragraphs.append(html.unescape(p.text.replace('\\n', ' ').strip()))\n",
" if not fiction:\n",
" with open(os.path.join(prefix_texts, rel_dir + '.txt'), 'a') as f:\n",
" for par in paragraphs:\n",
" par = re.sub(' ([.,;:!?)\"»])', '\\\\1', par)\n",
" par = par.replace('\\n', ' ')\n",
" par = par.strip()\n",
" par = re.sub('\\s+', ' ', par)\n",
" par = re.sub('\\[.+?\\]', '', par)\n",
" for sent in sentenize(par):\n",
" f.write(list(sent)[2].strip() + '\\n')\n",
" else:\n",
" if nameID in group_texts:\n",
" hyper_name = group_texts[nameID]\n",
" if hyper_name not in complex_texts:\n",
" complex_texts[hyper_name] = paragraphs\n",
" else:\n",
" complex_texts[hyper_name].extend(paragraphs)\n",
" else:\n",
" with open(os.path.join(prefix_texts, nameID + '.txt'), 'w') as f:\n",
" f.write('\\n'.join(paragraphs))\n",
"for hyper_name in complex_texts:\n",
" with open(os.path.join(prefix_texts, hyper_name + '.txt'), 'w') as f:\n",
" f.write('\\n'.join(complex_texts[hyper_name]))"
]
},
{
"cell_type": "markdown",
"id": "1eddfe2e",
"metadata": {},
"source": [
"# tagging"
]
},
{
"cell_type": "code",
"execution_count": 14,
"id": "115d0c54",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-15T23:52:40.491525Z",
"start_time": "2024-06-15T23:52:40.416283Z"
}
},
"outputs": [],
"source": [
"from pymystem3 import Mystem"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "c9441d01",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-15T23:53:11.904746Z",
"start_time": "2024-06-15T23:53:11.901127Z"
}
},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": 19,
"id": "6b1d6b0d",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-16T00:03:59.149011Z",
"start_time": "2024-06-16T00:03:59.132762Z"
}
},
"outputs": [],
"source": [
"def tagging():\n",
" m = Mystem()\n",
" for fl in os.listdir(prefix_texts):\n",
" #print(fl)\n",
" if 'mystem' in fl:\n",
" continue\n",
" with open(os.path.join(prefix_texts, fl)) as f:\n",
" text = f.read()\n",
" lines = text.split('\\n')\n",
" ana_lines = []\n",
" for line in lines:\n",
" line = ' '.join(line.split()[1:])\n",
" line = line.replace('ò', 'о')\n",
" line = line.replace('è', 'е')\n",
" line = line.replace('à', 'а')\n",
" line = line.replace('ѝ', 'и')\n",
" line = line.replace('ỳ', 'у')\n",
" line = line.replace('о̀', 'о')\n",
" #line = line.replace('Изд.̀', 'издательство')\n",
" ana = []\n",
" info = m.analyze(line)\n",
" for token in info:\n",
" if \"analysis\" in token:\n",
" try:\n",
" analysis = token[\"analysis\"][0]\n",
" except:\n",
" #print(token)\n",
" continue\n",
" # if \"lex\" in analysis:\n",
" lex = analysis[\"lex\"]\n",
" #if 'gr' in analysis:\n",
" gr = analysis['gr']\n",
" #print(gr)\n",
" const = gr.split('=')[0]\n",
" if ',' in const:\n",
" pos = const.split(',')[0]\n",
" else:\n",
" pos = const\n",
" \n",
" ana.append('{}_{}'.format(lex, pos))\n",
" ln = ' '.join(ana)\n",
" if re.search('[А-Яа-я]', ln):\n",
" ana_lines.append(ln)\n",
" with open('{}/mystem-{}'.format(prefix_texts, fl), 'w') as fw:\n",
" fw.write('\\n'.join(ana_lines))"
]
},
{
"cell_type": "code",
"execution_count": 20,
"id": "d02fd91a",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-16T00:12:05.148374Z",
"start_time": "2024-06-16T00:04:01.782191Z"
}
},
"outputs": [],
"source": [
"tagging()"
]
},
{
"cell_type": "code",
"execution_count": 18,
"id": "f9384f57",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-16T00:03:50.492957Z",
"start_time": "2024-06-16T00:03:50.485417Z"
}
},
"outputs": [],
"source": [
"pos = ['S', 'V', 'A', 'ADV']"
]
},
{
"cell_type": "code",
"execution_count": 22,
"id": "1bc596d8",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-16T00:13:58.529072Z",
"start_time": "2024-06-16T00:13:50.475301Z"
}
},
"outputs": [],
"source": [
"def mk_input():\n",
" inp = []\n",
" for fl in os.listdir(prefix_texts):\n",
" if not 'mystem' in fl:\n",
" continue\n",
" #print(fl)\n",
" with open(os.path.join(prefix_texts, fl)) as f:\n",
" text = f.read()\n",
" lines = text.split('\\n')\n",
" for line in lines:\n",
" words = []\n",
" for w in line.split():\n",
" word = w.split('_')\n",
" if word[1] in pos:\n",
" words.append(w)\n",
" if len(words) > 1:\n",
" inp.append(' '.join(words))\n",
" \n",
" with open('input.txt', 'w') as fw:\n",
" fw.write('\\n'.join(inp))\n",
" \n",
"mk_input()"
]
},
{
"cell_type": "markdown",
"id": "82faa45f",
"metadata": {},
"source": [
"# build models"
]
},
{
"cell_type": "code",
"execution_count": 26,
"id": "2d402b64",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-16T00:15:17.796668Z",
"start_time": "2024-06-16T00:15:13.952859Z"
}
},
"outputs": [],
"source": [
"import sys\n",
"import logging\n",
"import gensim"
]
},
{
"cell_type": "code",
"execution_count": 24,
"id": "5c994f11",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-16T00:14:45.969563Z",
"start_time": "2024-06-16T00:14:45.965851Z"
}
},
"outputs": [],
"source": [
"logging.basicConfig(format='%(asctime)s : %(levelname)s : %(message)s', level=logging.INFO)"
]
},
{
"cell_type": "code",
"execution_count": 27,
"id": "3ae5d312",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-16T00:15:19.248796Z",
"start_time": "2024-06-16T00:15:19.244970Z"
}
},
"outputs": [],
"source": [
"pth = './input.txt'\n",
"data = gensim.models.word2vec.LineSentence(pth)"
]
},
{
"cell_type": "code",
"execution_count": 28,
"id": "2af9f6a3",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-16T00:16:45.088436Z",
"start_time": "2024-06-16T00:15:24.270931Z"
}
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"2024-06-16 03:15:24,273 : INFO : collecting all words and their counts\n",
"2024-06-16 03:15:24,278 : INFO : PROGRESS: at sentence #0, processed 0 words, keeping 0 word types\n",
"2024-06-16 03:15:24,440 : INFO : PROGRESS: at sentence #10000, processed 163169 words, keeping 17313 word types\n",
"2024-06-16 03:15:24,556 : INFO : PROGRESS: at sentence #20000, processed 296024 words, keeping 23165 word types\n",
"2024-06-16 03:15:24,672 : INFO : PROGRESS: at sentence #30000, processed 405629 words, keeping 26267 word types\n",
"2024-06-16 03:15:24,800 : INFO : PROGRESS: at sentence #40000, processed 494131 words, keeping 27548 word types\n",
"2024-06-16 03:15:24,929 : INFO : PROGRESS: at sentence #50000, processed 582330 words, keeping 28502 word types\n",
"2024-06-16 03:15:25,090 : INFO : PROGRESS: at sentence #60000, processed 706637 words, keeping 30127 word types\n",
"2024-06-16 03:15:25,233 : INFO : PROGRESS: at sentence #70000, processed 847070 words, keeping 35040 word types\n",
"2024-06-16 03:15:25,373 : INFO : PROGRESS: at sentence #80000, processed 993651 words, keeping 36885 word types\n",
"2024-06-16 03:15:25,543 : INFO : PROGRESS: at sentence #90000, processed 1136268 words, keeping 38211 word types\n",
"2024-06-16 03:15:25,664 : INFO : PROGRESS: at sentence #100000, processed 1224248 words, keeping 38914 word types\n",
"2024-06-16 03:15:25,767 : INFO : PROGRESS: at sentence #110000, processed 1315622 words, keeping 39146 word types\n",
"2024-06-16 03:15:25,920 : INFO : PROGRESS: at sentence #120000, processed 1402410 words, keeping 41123 word types\n",
"2024-06-16 03:15:25,996 : INFO : PROGRESS: at sentence #130000, processed 1445449 words, keeping 43388 word types\n",
"2024-06-16 03:15:26,136 : INFO : PROGRESS: at sentence #140000, processed 1568666 words, keeping 45060 word types\n",
"2024-06-16 03:15:26,215 : INFO : PROGRESS: at sentence #150000, processed 1625965 words, keeping 46676 word types\n",
"2024-06-16 03:15:26,317 : INFO : PROGRESS: at sentence #160000, processed 1676634 words, keeping 47593 word types\n",
"2024-06-16 03:15:26,404 : INFO : PROGRESS: at sentence #170000, processed 1740516 words, keeping 48188 word types\n",
"2024-06-16 03:15:26,493 : INFO : PROGRESS: at sentence #180000, processed 1813933 words, keeping 48617 word types\n",
"2024-06-16 03:15:26,572 : INFO : PROGRESS: at sentence #190000, processed 1873429 words, keeping 49050 word types\n",
"2024-06-16 03:15:26,729 : INFO : PROGRESS: at sentence #200000, processed 2053959 words, keeping 51906 word types\n",
"2024-06-16 03:15:26,891 : INFO : PROGRESS: at sentence #210000, processed 2244539 words, keeping 54019 word types\n",
"2024-06-16 03:15:27,068 : INFO : PROGRESS: at sentence #220000, processed 2470575 words, keeping 56680 word types\n",
"2024-06-16 03:15:27,261 : INFO : PROGRESS: at sentence #230000, processed 2707756 words, keeping 58963 word types\n",
"2024-06-16 03:15:27,461 : INFO : PROGRESS: at sentence #240000, processed 2944125 words, keeping 60490 word types\n",
"2024-06-16 03:15:27,606 : INFO : PROGRESS: at sentence #250000, processed 3100643 words, keeping 61888 word types\n",
"2024-06-16 03:15:27,697 : INFO : PROGRESS: at sentence #260000, processed 3175131 words, keeping 62638 word types\n",
"2024-06-16 03:15:27,789 : INFO : PROGRESS: at sentence #270000, processed 3246248 words, keeping 63144 word types\n",
"2024-06-16 03:15:27,887 : INFO : PROGRESS: at sentence #280000, processed 3324748 words, keeping 63561 word types\n",
"2024-06-16 03:15:27,991 : INFO : PROGRESS: at sentence #290000, processed 3406817 words, keeping 64030 word types\n",
"2024-06-16 03:15:28,091 : INFO : PROGRESS: at sentence #300000, processed 3491208 words, keeping 64525 word types\n",
"2024-06-16 03:15:28,190 : INFO : PROGRESS: at sentence #310000, processed 3567554 words, keeping 64999 word types\n",
"2024-06-16 03:15:28,300 : INFO : PROGRESS: at sentence #320000, processed 3653814 words, keeping 65369 word types\n",
"2024-06-16 03:15:28,397 : INFO : PROGRESS: at sentence #330000, processed 3727990 words, keeping 65832 word types\n",
"2024-06-16 03:15:28,482 : INFO : PROGRESS: at sentence #340000, processed 3785783 words, keeping 66225 word types\n",
"2024-06-16 03:15:28,561 : INFO : PROGRESS: at sentence #350000, processed 3851922 words, keeping 66596 word types\n",
"2024-06-16 03:15:28,682 : INFO : PROGRESS: at sentence #360000, processed 3955668 words, keeping 67421 word types\n",
"2024-06-16 03:15:28,732 : INFO : collected 67995 word types from a corpus of 4013071 raw words and 362579 sentences\n",
"2024-06-16 03:15:28,733 : INFO : Creating a fresh vocabulary\n",
"2024-06-16 03:15:29,065 : INFO : Word2Vec lifecycle event {'msg': 'effective_min_count=2 retains 40255 unique words (59.20% of original 67995, drops 27740)', 'datetime': '2024-06-16T03:15:29.048822', 'gensim': '4.3.2', 'python': '3.8.10 (default, Nov 22 2023, 10:22:35) \\n[GCC 9.4.0]', 'platform': 'Linux-5.4.0-182-generic-x86_64-with-glibc2.29', 'event': 'prepare_vocab'}\n",
"2024-06-16 03:15:29,066 : INFO : Word2Vec lifecycle event {'msg': 'effective_min_count=2 leaves 3985331 word corpus (99.31% of original 4013071, drops 27740)', 'datetime': '2024-06-16T03:15:29.066649', 'gensim': '4.3.2', 'python': '3.8.10 (default, Nov 22 2023, 10:22:35) \\n[GCC 9.4.0]', 'platform': 'Linux-5.4.0-182-generic-x86_64-with-glibc2.29', 'event': 'prepare_vocab'}\n",
"2024-06-16 03:15:29,499 : INFO : deleting the raw counts dictionary of 67995 items\n",
"2024-06-16 03:15:29,502 : INFO : sample=0.001 downsamples 30 most-common words\n",
"2024-06-16 03:15:29,503 : INFO : Word2Vec lifecycle event {'msg': 'downsampling leaves estimated 3707673.978375597 word corpus (93.0%% of prior 3985331)', 'datetime': '2024-06-16T03:15:29.503269', 'gensim': '4.3.2', 'python': '3.8.10 (default, Nov 22 2023, 10:22:35) \\n[GCC 9.4.0]', 'platform': 'Linux-5.4.0-182-generic-x86_64-with-glibc2.29', 'event': 'prepare_vocab'}\n",
"2024-06-16 03:15:30,266 : INFO : estimated required memory for 40255 words and 500 dimensions: 181147500 bytes\n",
"2024-06-16 03:15:30,267 : INFO : resetting layer weights\n",
"2024-06-16 03:15:30,558 : INFO : Word2Vec lifecycle event {'update': False, 'trim_rule': 'None', 'datetime': '2024-06-16T03:15:30.558380', 'gensim': '4.3.2', 'python': '3.8.10 (default, Nov 22 2023, 10:22:35) \\n[GCC 9.4.0]', 'platform': 'Linux-5.4.0-182-generic-x86_64-with-glibc2.29', 'event': 'build_vocab'}\n",
"2024-06-16 03:15:30,559 : INFO : Word2Vec lifecycle event {'msg': 'training model with 3 workers on 40255 vocabulary and 500 features, using sg=1 hs=0 sample=0.001 negative=5 window=2 shrink_windows=True', 'datetime': '2024-06-16T03:15:30.559294', 'gensim': '4.3.2', 'python': '3.8.10 (default, Nov 22 2023, 10:22:35) \\n[GCC 9.4.0]', 'platform': 'Linux-5.4.0-182-generic-x86_64-with-glibc2.29', 'event': 'train'}\n",
"2024-06-16 03:15:31,645 : INFO : EPOCH 0 - PROGRESS: at 3.31% examples, 175241 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:15:32,667 : INFO : EPOCH 0 - PROGRESS: at 11.85% examples, 231581 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:15:33,729 : INFO : EPOCH 0 - PROGRESS: at 18.62% examples, 236754 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:15:34,753 : INFO : EPOCH 0 - PROGRESS: at 24.47% examples, 246985 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:15:35,796 : INFO : EPOCH 0 - PROGRESS: at 36.59% examples, 256599 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:15:36,796 : INFO : EPOCH 0 - PROGRESS: at 46.75% examples, 256986 words/s, in_qsize 6, out_qsize 0\n",
"2024-06-16 03:15:37,830 : INFO : EPOCH 0 - PROGRESS: at 55.05% examples, 259692 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:15:38,868 : INFO : EPOCH 0 - PROGRESS: at 59.24% examples, 261775 words/s, in_qsize 6, out_qsize 0\n",
"2024-06-16 03:15:39,868 : INFO : EPOCH 0 - PROGRESS: at 62.57% examples, 261708 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:15:40,877 : INFO : EPOCH 0 - PROGRESS: at 66.04% examples, 263237 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:15:41,922 : INFO : EPOCH 0 - PROGRESS: at 74.34% examples, 264375 words/s, in_qsize 6, out_qsize 0\n",
"2024-06-16 03:15:42,924 : INFO : EPOCH 0 - PROGRESS: at 84.50% examples, 265285 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:15:43,935 : INFO : EPOCH 0 - PROGRESS: at 95.74% examples, 265237 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:15:44,508 : INFO : EPOCH 0: training on 4013071 raw words (3706579 effective words) took 13.9s, 266440 effective words/s\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"2024-06-16 03:15:45,538 : INFO : EPOCH 1 - PROGRESS: at 4.82% examples, 245175 words/s, in_qsize 5, out_qsize 1\n",
"2024-06-16 03:15:46,538 : INFO : EPOCH 1 - PROGRESS: at 13.74% examples, 264455 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:15:47,560 : INFO : EPOCH 1 - PROGRESS: at 20.30% examples, 271388 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:15:48,571 : INFO : EPOCH 1 - PROGRESS: at 27.11% examples, 273492 words/s, in_qsize 6, out_qsize 0\n",
"2024-06-16 03:15:49,589 : INFO : EPOCH 1 - PROGRESS: at 37.36% examples, 266737 words/s, in_qsize 6, out_qsize 0\n",
"2024-06-16 03:15:50,610 : INFO : EPOCH 1 - PROGRESS: at 49.03% examples, 270570 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:15:51,651 : INFO : EPOCH 1 - PROGRESS: at 55.81% examples, 269964 words/s, in_qsize 3, out_qsize 2\n",
"2024-06-16 03:15:52,652 : INFO : EPOCH 1 - PROGRESS: at 59.80% examples, 270869 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:15:53,694 : INFO : EPOCH 1 - PROGRESS: at 63.30% examples, 270642 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:15:54,753 : INFO : EPOCH 1 - PROGRESS: at 67.06% examples, 270853 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:15:55,794 : INFO : EPOCH 1 - PROGRESS: at 75.77% examples, 268905 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:15:56,813 : INFO : EPOCH 1 - PROGRESS: at 87.02% examples, 271322 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:15:57,832 : INFO : EPOCH 1 - PROGRESS: at 98.72% examples, 271369 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:15:58,134 : INFO : EPOCH 1: training on 4013071 raw words (3707481 effective words) took 13.6s, 272298 effective words/s\n",
"2024-06-16 03:15:59,190 : INFO : EPOCH 2 - PROGRESS: at 4.90% examples, 238625 words/s, in_qsize 4, out_qsize 1\n",
"2024-06-16 03:16:00,191 : INFO : EPOCH 2 - PROGRESS: at 13.74% examples, 260706 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:01,207 : INFO : EPOCH 2 - PROGRESS: at 19.69% examples, 260333 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:02,207 : INFO : EPOCH 2 - PROGRESS: at 25.56% examples, 261725 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:03,234 : INFO : EPOCH 2 - PROGRESS: at 37.33% examples, 265675 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:04,250 : INFO : EPOCH 2 - PROGRESS: at 48.30% examples, 266960 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:05,339 : INFO : EPOCH 2 - PROGRESS: at 55.19% examples, 262437 words/s, in_qsize 3, out_qsize 2\n",
"2024-06-16 03:16:06,359 : INFO : EPOCH 2 - PROGRESS: at 58.68% examples, 258004 words/s, in_qsize 6, out_qsize 0\n",
"2024-06-16 03:16:07,365 : INFO : EPOCH 2 - PROGRESS: at 61.57% examples, 255105 words/s, in_qsize 6, out_qsize 0\n",
"2024-06-16 03:16:08,426 : INFO : EPOCH 2 - PROGRESS: at 65.03% examples, 255134 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:09,446 : INFO : EPOCH 2 - PROGRESS: at 69.40% examples, 254291 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:10,471 : INFO : EPOCH 2 - PROGRESS: at 81.05% examples, 257874 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:11,491 : INFO : EPOCH 2 - PROGRESS: at 91.69% examples, 258851 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:12,338 : INFO : EPOCH 2: training on 4013071 raw words (3707528 effective words) took 14.2s, 261187 effective words/s\n",
"2024-06-16 03:16:13,375 : INFO : EPOCH 3 - PROGRESS: at 4.15% examples, 214882 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:14,381 : INFO : EPOCH 3 - PROGRESS: at 10.86% examples, 222177 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:15,407 : INFO : EPOCH 3 - PROGRESS: at 17.63% examples, 227504 words/s, in_qsize 6, out_qsize 0\n",
"2024-06-16 03:16:16,434 : INFO : EPOCH 3 - PROGRESS: at 22.14% examples, 224431 words/s, in_qsize 6, out_qsize 0\n",
"2024-06-16 03:16:17,437 : INFO : EPOCH 3 - PROGRESS: at 27.69% examples, 221242 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:18,443 : INFO : EPOCH 3 - PROGRESS: at 37.76% examples, 224767 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:19,444 : INFO : EPOCH 3 - PROGRESS: at 46.77% examples, 224511 words/s, in_qsize 4, out_qsize 1\n",
"2024-06-16 03:16:20,507 : INFO : EPOCH 3 - PROGRESS: at 54.22% examples, 226818 words/s, in_qsize 6, out_qsize 1\n",
"2024-06-16 03:16:21,534 : INFO : EPOCH 3 - PROGRESS: at 58.79% examples, 231683 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:22,541 : INFO : EPOCH 3 - PROGRESS: at 61.92% examples, 233456 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:23,544 : INFO : EPOCH 3 - PROGRESS: at 64.52% examples, 230900 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:24,567 : INFO : EPOCH 3 - PROGRESS: at 68.05% examples, 232123 words/s, in_qsize 4, out_qsize 1\n",
"2024-06-16 03:16:25,606 : INFO : EPOCH 3 - PROGRESS: at 76.77% examples, 230723 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:26,610 : INFO : EPOCH 3 - PROGRESS: at 86.37% examples, 232550 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:27,619 : INFO : EPOCH 3 - PROGRESS: at 95.34% examples, 231091 words/s, in_qsize 4, out_qsize 1\n",
"2024-06-16 03:16:28,232 : INFO : EPOCH 3: training on 4013071 raw words (3707859 effective words) took 15.9s, 233372 effective words/s\n",
"2024-06-16 03:16:29,237 : INFO : EPOCH 4 - PROGRESS: at 3.66% examples, 202153 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:30,249 : INFO : EPOCH 4 - PROGRESS: at 9.25% examples, 201858 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:31,265 : INFO : EPOCH 4 - PROGRESS: at 16.41% examples, 211957 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:32,302 : INFO : EPOCH 4 - PROGRESS: at 20.89% examples, 209733 words/s, in_qsize 4, out_qsize 1\n",
"2024-06-16 03:16:33,338 : INFO : EPOCH 4 - PROGRESS: at 24.80% examples, 202976 words/s, in_qsize 6, out_qsize 0\n",
"2024-06-16 03:16:34,352 : INFO : EPOCH 4 - PROGRESS: at 31.25% examples, 201556 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:35,404 : INFO : EPOCH 4 - PROGRESS: at 38.49% examples, 198993 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:36,419 : INFO : EPOCH 4 - PROGRESS: at 49.03% examples, 201321 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:37,456 : INFO : EPOCH 4 - PROGRESS: at 54.64% examples, 202724 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:38,505 : INFO : EPOCH 4 - PROGRESS: at 58.45% examples, 204538 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:39,580 : INFO : EPOCH 4 - PROGRESS: at 61.92% examples, 209768 words/s, in_qsize 6, out_qsize 0\n",
"2024-06-16 03:16:40,591 : INFO : EPOCH 4 - PROGRESS: at 64.80% examples, 210755 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:41,604 : INFO : EPOCH 4 - PROGRESS: at 68.06% examples, 212147 words/s, in_qsize 4, out_qsize 1\n",
"2024-06-16 03:16:42,605 : INFO : EPOCH 4 - PROGRESS: at 78.95% examples, 216708 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:43,606 : INFO : EPOCH 4 - PROGRESS: at 87.31% examples, 217568 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:44,611 : INFO : EPOCH 4 - PROGRESS: at 98.97% examples, 221720 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:16:44,891 : INFO : EPOCH 4: training on 4013071 raw words (3706946 effective words) took 16.7s, 222554 effective words/s\n",
"2024-06-16 03:16:44,892 : INFO : Word2Vec lifecycle event {'msg': 'training on 20065355 raw words (18536393 effective words) took 74.3s, 249375 effective words/s', 'datetime': '2024-06-16T03:16:44.892076', 'gensim': '4.3.2', 'python': '3.8.10 (default, Nov 22 2023, 10:22:35) \\n[GCC 9.4.0]', 'platform': 'Linux-5.4.0-182-generic-x86_64-with-glibc2.29', 'event': 'train'}\n",
"2024-06-16 03:16:44,895 : INFO : Word2Vec lifecycle event {'params': 'Word2Vec', 'datetime': '2024-06-16T03:16:44.895721', 'gensim': '4.3.2', 'python': '3.8.10 (default, Nov 22 2023, 10:22:35) \\n[GCC 9.4.0]', 'platform': 'Linux-5.4.0-182-generic-x86_64-with-glibc2.29', 'event': 'created'}\n",
"2024-06-16 03:16:44,896 : INFO : Word2Vec lifecycle event {'fname_or_handle': 'skipgram_500_2.model', 'separately': 'None', 'sep_limit': 10485760, 'ignore': frozenset(), 'datetime': '2024-06-16T03:16:44.896765', 'gensim': '4.3.2', 'python': '3.8.10 (default, Nov 22 2023, 10:22:35) \\n[GCC 9.4.0]', 'platform': 'Linux-5.4.0-182-generic-x86_64-with-glibc2.29', 'event': 'saving'}\n",
"2024-06-16 03:16:44,897 : INFO : storing np array 'vectors' to skipgram_500_2.model.wv.vectors.npy\n",
"2024-06-16 03:16:45,009 : INFO : storing np array 'syn1neg' to skipgram_500_2.model.syn1neg.npy\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"2024-06-16 03:16:45,056 : INFO : not storing attribute cum_table\n",
"2024-06-16 03:16:45,084 : INFO : saved skipgram_500_2.model\n"
]
}
],
"source": [
"modelLNT1 = gensim.models.Word2Vec(data, vector_size=500, window=2, min_count=2, sg=1) # comparable with web_mystem_skipgram_500_2_2015.bin\n",
"modelLNT1.save('skipgram_500_2.model') # modelLNT1 = Word2Vec.load(\"skipgram_500_2.model\")"
]
},
{
"cell_type": "code",
"execution_count": 29,
"id": "0a7ec57a",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-16T00:17:48.857219Z",
"start_time": "2024-06-16T00:16:56.047446Z"
}
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"2024-06-16 03:16:56,049 : INFO : collecting all words and their counts\n",
"2024-06-16 03:16:56,052 : INFO : PROGRESS: at sentence #0, processed 0 words, keeping 0 word types\n",
"2024-06-16 03:16:56,262 : INFO : PROGRESS: at sentence #10000, processed 163169 words, keeping 17313 word types\n",
"2024-06-16 03:16:56,392 : INFO : PROGRESS: at sentence #20000, processed 296024 words, keeping 23165 word types\n",
"2024-06-16 03:16:56,513 : INFO : PROGRESS: at sentence #30000, processed 405629 words, keeping 26267 word types\n",
"2024-06-16 03:16:56,602 : INFO : PROGRESS: at sentence #40000, processed 494131 words, keeping 27548 word types\n",
"2024-06-16 03:16:56,705 : INFO : PROGRESS: at sentence #50000, processed 582330 words, keeping 28502 word types\n",
"2024-06-16 03:16:56,833 : INFO : PROGRESS: at sentence #60000, processed 706637 words, keeping 30127 word types\n",
"2024-06-16 03:16:56,977 : INFO : PROGRESS: at sentence #70000, processed 847070 words, keeping 35040 word types\n",
"2024-06-16 03:16:57,114 : INFO : PROGRESS: at sentence #80000, processed 993651 words, keeping 36885 word types\n",
"2024-06-16 03:16:57,270 : INFO : PROGRESS: at sentence #90000, processed 1136268 words, keeping 38211 word types\n",
"2024-06-16 03:16:57,381 : INFO : PROGRESS: at sentence #100000, processed 1224248 words, keeping 38914 word types\n",
"2024-06-16 03:16:57,483 : INFO : PROGRESS: at sentence #110000, processed 1315622 words, keeping 39146 word types\n",
"2024-06-16 03:16:57,597 : INFO : PROGRESS: at sentence #120000, processed 1402410 words, keeping 41123 word types\n",
"2024-06-16 03:16:57,668 : INFO : PROGRESS: at sentence #130000, processed 1445449 words, keeping 43388 word types\n",
"2024-06-16 03:16:57,791 : INFO : PROGRESS: at sentence #140000, processed 1568666 words, keeping 45060 word types\n",
"2024-06-16 03:16:57,871 : INFO : PROGRESS: at sentence #150000, processed 1625965 words, keeping 46676 word types\n",
"2024-06-16 03:16:57,967 : INFO : PROGRESS: at sentence #160000, processed 1676634 words, keeping 47593 word types\n",
"2024-06-16 03:16:58,070 : INFO : PROGRESS: at sentence #170000, processed 1740516 words, keeping 48188 word types\n",
"2024-06-16 03:16:58,181 : INFO : PROGRESS: at sentence #180000, processed 1813933 words, keeping 48617 word types\n",
"2024-06-16 03:16:58,277 : INFO : PROGRESS: at sentence #190000, processed 1873429 words, keeping 49050 word types\n",
"2024-06-16 03:16:58,476 : INFO : PROGRESS: at sentence #200000, processed 2053959 words, keeping 51906 word types\n",
"2024-06-16 03:16:58,643 : INFO : PROGRESS: at sentence #210000, processed 2244539 words, keeping 54019 word types\n",
"2024-06-16 03:16:58,832 : INFO : PROGRESS: at sentence #220000, processed 2470575 words, keeping 56680 word types\n",
"2024-06-16 03:16:59,029 : INFO : PROGRESS: at sentence #230000, processed 2707756 words, keeping 58963 word types\n",
"2024-06-16 03:16:59,219 : INFO : PROGRESS: at sentence #240000, processed 2944125 words, keeping 60490 word types\n",
"2024-06-16 03:16:59,368 : INFO : PROGRESS: at sentence #250000, processed 3100643 words, keeping 61888 word types\n",
"2024-06-16 03:16:59,462 : INFO : PROGRESS: at sentence #260000, processed 3175131 words, keeping 62638 word types\n",
"2024-06-16 03:16:59,551 : INFO : PROGRESS: at sentence #270000, processed 3246248 words, keeping 63144 word types\n",
"2024-06-16 03:16:59,650 : INFO : PROGRESS: at sentence #280000, processed 3324748 words, keeping 63561 word types\n",
"2024-06-16 03:16:59,750 : INFO : PROGRESS: at sentence #290000, processed 3406817 words, keeping 64030 word types\n",
"2024-06-16 03:16:59,852 : INFO : PROGRESS: at sentence #300000, processed 3491208 words, keeping 64525 word types\n",
"2024-06-16 03:16:59,946 : INFO : PROGRESS: at sentence #310000, processed 3567554 words, keeping 64999 word types\n",
"2024-06-16 03:17:00,056 : INFO : PROGRESS: at sentence #320000, processed 3653814 words, keeping 65369 word types\n",
"2024-06-16 03:17:00,141 : INFO : PROGRESS: at sentence #330000, processed 3727990 words, keeping 65832 word types\n",
"2024-06-16 03:17:00,219 : INFO : PROGRESS: at sentence #340000, processed 3785783 words, keeping 66225 word types\n",
"2024-06-16 03:17:00,314 : INFO : PROGRESS: at sentence #350000, processed 3851922 words, keeping 66596 word types\n",
"2024-06-16 03:17:00,423 : INFO : PROGRESS: at sentence #360000, processed 3955668 words, keeping 67421 word types\n",
"2024-06-16 03:17:00,477 : INFO : collected 67995 word types from a corpus of 4013071 raw words and 362579 sentences\n",
"2024-06-16 03:17:00,478 : INFO : Creating a fresh vocabulary\n",
"2024-06-16 03:17:00,844 : INFO : Word2Vec lifecycle event {'msg': 'effective_min_count=2 retains 40255 unique words (59.20% of original 67995, drops 27740)', 'datetime': '2024-06-16T03:17:00.844097', 'gensim': '4.3.2', 'python': '3.8.10 (default, Nov 22 2023, 10:22:35) \\n[GCC 9.4.0]', 'platform': 'Linux-5.4.0-182-generic-x86_64-with-glibc2.29', 'event': 'prepare_vocab'}\n",
"2024-06-16 03:17:00,845 : INFO : Word2Vec lifecycle event {'msg': 'effective_min_count=2 leaves 3985331 word corpus (99.31% of original 4013071, drops 27740)', 'datetime': '2024-06-16T03:17:00.845547', 'gensim': '4.3.2', 'python': '3.8.10 (default, Nov 22 2023, 10:22:35) \\n[GCC 9.4.0]', 'platform': 'Linux-5.4.0-182-generic-x86_64-with-glibc2.29', 'event': 'prepare_vocab'}\n",
"2024-06-16 03:17:01,316 : INFO : deleting the raw counts dictionary of 67995 items\n",
"2024-06-16 03:17:01,318 : INFO : sample=0.001 downsamples 30 most-common words\n",
"2024-06-16 03:17:01,320 : INFO : Word2Vec lifecycle event {'msg': 'downsampling leaves estimated 3707673.978375597 word corpus (93.0%% of prior 3985331)', 'datetime': '2024-06-16T03:17:01.320105', 'gensim': '4.3.2', 'python': '3.8.10 (default, Nov 22 2023, 10:22:35) \\n[GCC 9.4.0]', 'platform': 'Linux-5.4.0-182-generic-x86_64-with-glibc2.29', 'event': 'prepare_vocab'}\n",
"2024-06-16 03:17:02,054 : INFO : estimated required memory for 40255 words and 300 dimensions: 116739500 bytes\n",
"2024-06-16 03:17:02,055 : INFO : resetting layer weights\n",
"2024-06-16 03:17:02,261 : INFO : Word2Vec lifecycle event {'update': False, 'trim_rule': 'None', 'datetime': '2024-06-16T03:17:02.261886', 'gensim': '4.3.2', 'python': '3.8.10 (default, Nov 22 2023, 10:22:35) \\n[GCC 9.4.0]', 'platform': 'Linux-5.4.0-182-generic-x86_64-with-glibc2.29', 'event': 'build_vocab'}\n",
"2024-06-16 03:17:02,262 : INFO : Word2Vec lifecycle event {'msg': 'training model with 3 workers on 40255 vocabulary and 300 features, using sg=0 hs=0 sample=0.001 negative=5 window=10 shrink_windows=True', 'datetime': '2024-06-16T03:17:02.262796', 'gensim': '4.3.2', 'python': '3.8.10 (default, Nov 22 2023, 10:22:35) \\n[GCC 9.4.0]', 'platform': 'Linux-5.4.0-182-generic-x86_64-with-glibc2.29', 'event': 'train'}\n",
"2024-06-16 03:17:03,289 : INFO : EPOCH 0 - PROGRESS: at 7.88% examples, 359855 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:04,312 : INFO : EPOCH 0 - PROGRESS: at 20.90% examples, 420774 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:05,346 : INFO : EPOCH 0 - PROGRESS: at 32.77% examples, 417947 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:06,372 : INFO : EPOCH 0 - PROGRESS: at 52.69% examples, 425485 words/s, in_qsize 4, out_qsize 2\n",
"2024-06-16 03:17:07,393 : INFO : EPOCH 0 - PROGRESS: at 60.15% examples, 436648 words/s, in_qsize 4, out_qsize 1\n",
"2024-06-16 03:17:08,400 : INFO : EPOCH 0 - PROGRESS: at 66.70% examples, 450158 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:09,415 : INFO : EPOCH 0 - PROGRESS: at 81.39% examples, 447094 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:10,423 : INFO : EPOCH 0 - PROGRESS: at 97.78% examples, 440450 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:10,718 : INFO : EPOCH 0: training on 4013071 raw words (3707249 effective words) took 8.4s, 439586 effective words/s\n",
"2024-06-16 03:17:11,751 : INFO : EPOCH 1 - PROGRESS: at 4.82% examples, 245164 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:12,762 : INFO : EPOCH 1 - PROGRESS: at 11.53% examples, 231748 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:13,770 : INFO : EPOCH 1 - PROGRESS: at 19.14% examples, 253363 words/s, in_qsize 6, out_qsize 0\n",
"2024-06-16 03:17:14,817 : INFO : EPOCH 1 - PROGRESS: at 27.11% examples, 271322 words/s, in_qsize 6, out_qsize 0\n",
"2024-06-16 03:17:15,819 : INFO : EPOCH 1 - PROGRESS: at 39.77% examples, 287807 words/s, in_qsize 5, out_qsize 0\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"2024-06-16 03:17:16,847 : INFO : EPOCH 1 - PROGRESS: at 55.19% examples, 308798 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:17,859 : INFO : EPOCH 1 - PROGRESS: at 62.24% examples, 336579 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:18,881 : INFO : EPOCH 1 - PROGRESS: at 66.86% examples, 337849 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:19,889 : INFO : EPOCH 1 - PROGRESS: at 79.56% examples, 342120 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:20,947 : INFO : EPOCH 1 - PROGRESS: at 87.02% examples, 326518 words/s, in_qsize 6, out_qsize 0\n",
"2024-06-16 03:17:21,959 : INFO : EPOCH 1 - PROGRESS: at 100.00% examples, 330219 words/s, in_qsize 0, out_qsize 1\n",
"2024-06-16 03:17:21,961 : INFO : EPOCH 1: training on 4013071 raw words (3707615 effective words) took 11.2s, 330162 effective words/s\n",
"2024-06-16 03:17:22,981 : INFO : EPOCH 2 - PROGRESS: at 11.53% examples, 467727 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:23,992 : INFO : EPOCH 2 - PROGRESS: at 22.81% examples, 468651 words/s, in_qsize 4, out_qsize 1\n",
"2024-06-16 03:17:25,012 : INFO : EPOCH 2 - PROGRESS: at 36.59% examples, 439519 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:26,046 : INFO : EPOCH 2 - PROGRESS: at 53.83% examples, 447931 words/s, in_qsize 4, out_qsize 1\n",
"2024-06-16 03:17:27,054 : INFO : EPOCH 2 - PROGRESS: at 60.26% examples, 441219 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:28,061 : INFO : EPOCH 2 - PROGRESS: at 67.06% examples, 455493 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:29,087 : INFO : EPOCH 2 - PROGRESS: at 81.78% examples, 449594 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:30,098 : INFO : EPOCH 2 - PROGRESS: at 99.61% examples, 452851 words/s, in_qsize 3, out_qsize 1\n",
"2024-06-16 03:17:30,133 : INFO : EPOCH 2: training on 4013071 raw words (3707669 effective words) took 8.2s, 454472 effective words/s\n",
"2024-06-16 03:17:31,161 : INFO : EPOCH 3 - PROGRESS: at 12.18% examples, 477366 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:32,209 : INFO : EPOCH 3 - PROGRESS: at 21.96% examples, 438616 words/s, in_qsize 4, out_qsize 1\n",
"2024-06-16 03:17:33,241 : INFO : EPOCH 3 - PROGRESS: at 31.91% examples, 406354 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:34,258 : INFO : EPOCH 3 - PROGRESS: at 52.60% examples, 419940 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:35,271 : INFO : EPOCH 3 - PROGRESS: at 59.11% examples, 420036 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:36,279 : INFO : EPOCH 3 - PROGRESS: at 64.67% examples, 422528 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:37,294 : INFO : EPOCH 3 - PROGRESS: at 76.41% examples, 426170 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:38,349 : INFO : EPOCH 3 - PROGRESS: at 87.92% examples, 409510 words/s, in_qsize 4, out_qsize 1\n",
"2024-06-16 03:17:39,361 : INFO : EPOCH 3 - PROGRESS: at 98.97% examples, 393706 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:39,471 : INFO : EPOCH 3: training on 4013071 raw words (3707281 effective words) took 9.3s, 397223 effective words/s\n",
"2024-06-16 03:17:40,477 : INFO : EPOCH 4 - PROGRESS: at 7.88% examples, 360534 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:41,484 : INFO : EPOCH 4 - PROGRESS: at 18.81% examples, 373825 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:42,505 : INFO : EPOCH 4 - PROGRESS: at 30.42% examples, 398180 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:43,509 : INFO : EPOCH 4 - PROGRESS: at 45.52% examples, 388348 words/s, in_qsize 6, out_qsize 0\n",
"2024-06-16 03:17:44,574 : INFO : EPOCH 4 - PROGRESS: at 53.82% examples, 357575 words/s, in_qsize 3, out_qsize 2\n",
"2024-06-16 03:17:45,583 : INFO : EPOCH 4 - PROGRESS: at 60.26% examples, 366828 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:46,604 : INFO : EPOCH 4 - PROGRESS: at 66.70% examples, 386252 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:47,625 : INFO : EPOCH 4 - PROGRESS: at 81.78% examples, 392361 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:48,629 : INFO : EPOCH 4 - PROGRESS: at 99.09% examples, 397753 words/s, in_qsize 5, out_qsize 0\n",
"2024-06-16 03:17:48,744 : INFO : EPOCH 4: training on 4013071 raw words (3707868 effective words) took 9.3s, 399986 effective words/s\n",
"2024-06-16 03:17:48,745 : INFO : Word2Vec lifecycle event {'msg': 'training on 20065355 raw words (18537682 effective words) took 46.5s, 398815 effective words/s', 'datetime': '2024-06-16T03:17:48.745392', 'gensim': '4.3.2', 'python': '3.8.10 (default, Nov 22 2023, 10:22:35) \\n[GCC 9.4.0]', 'platform': 'Linux-5.4.0-182-generic-x86_64-with-glibc2.29', 'event': 'train'}\n",
"2024-06-16 03:17:48,746 : INFO : Word2Vec lifecycle event {'params': 'Word2Vec', 'datetime': '2024-06-16T03:17:48.746089', 'gensim': '4.3.2', 'python': '3.8.10 (default, Nov 22 2023, 10:22:35) \\n[GCC 9.4.0]', 'platform': 'Linux-5.4.0-182-generic-x86_64-with-glibc2.29', 'event': 'created'}\n",
"2024-06-16 03:17:48,749 : INFO : Word2Vec lifecycle event {'fname_or_handle': 'cbow_300_10.model', 'separately': 'None', 'sep_limit': 10485760, 'ignore': frozenset(), 'datetime': '2024-06-16T03:17:48.749250', 'gensim': '4.3.2', 'python': '3.8.10 (default, Nov 22 2023, 10:22:35) \\n[GCC 9.4.0]', 'platform': 'Linux-5.4.0-182-generic-x86_64-with-glibc2.29', 'event': 'saving'}\n",
"2024-06-16 03:17:48,750 : INFO : storing np array 'vectors' to cbow_300_10.model.wv.vectors.npy\n",
"2024-06-16 03:17:48,792 : INFO : storing np array 'syn1neg' to cbow_300_10.model.syn1neg.npy\n",
"2024-06-16 03:17:48,821 : INFO : not storing attribute cum_table\n",
"2024-06-16 03:17:48,852 : INFO : saved cbow_300_10.model\n"
]
}
],
"source": [
"modelLNT2 = gensim.models.Word2Vec(data, vector_size=300, window=10, min_count=2, sg=0) # comparable with ruwikiruscorpora_upos_cbow_300_10_2021\n",
"modelLNT2.save('cbow_300_10.model')"
]
},
{
"cell_type": "markdown",
"id": "e3a7d2c3",
"metadata": {},
"source": [
"# most similar words viz"
]
},
{
"cell_type": "code",
"execution_count": 33,
"id": "d443adf9",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-16T00:25:19.723274Z",
"start_time": "2024-06-16T00:25:19.698591Z"
}
},
"outputs": [],
"source": [
"import numpy as np\n",
"import pandas as pd\n",
"import matplotlib.pyplot as plt\n",
"%matplotlib inline\n",
"\n",
"import seaborn as sns\n",
"sns.set_style(\"darkgrid\")\n",
"\n",
"from sklearn.decomposition import PCA\n",
"from sklearn.manifold import TSNE"
]
},
{
"cell_type": "code",
"execution_count": 34,
"id": "1c88ab62",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-16T00:25:24.532880Z",
"start_time": "2024-06-16T00:25:24.513702Z"
}
},
"outputs": [],
"source": [
"def tsnescatterplot(model, word, list_names):\n",
" \"\"\" Plot in seaborn the results from the t-SNE dimensionality reduction algorithm of the vectors of a query word,\n",
" its list of most similar words, and a list of words.\n",
" \"\"\"\n",
" arrays = np.empty((0, 300), dtype='f')\n",
" word_labels = [word]\n",
" color_list = ['red']\n",
"\n",
" # adds the vector of the query word\n",
" arrays = np.append(arrays, model.wv.__getitem__([word]), axis=0)\n",
" \n",
" # gets list of most similar words\n",
" close_words = model.wv.most_similar([word])\n",
" \n",
" # adds the vector for each of the closest words to the array\n",
" for wrd_score in close_words:\n",
" wrd_vector = model.wv.__getitem__([wrd_score[0]])\n",
" word_labels.append(wrd_score[0])\n",
" color_list.append('blue')\n",
" arrays = np.append(arrays, wrd_vector, axis=0)\n",
" \n",
" # adds the vector for each of the words from list_names to the array\n",
" for wrd in list_names:\n",
" wrd_vector = model.wv.__getitem__([wrd])\n",
" word_labels.append(wrd)\n",
" color_list.append('green')\n",
" arrays = np.append(arrays, wrd_vector, axis=0)\n",
" \n",
" # Reduces the dimensionality from 300 to 50 dimensions with PCA\n",
" reduc = PCA(n_components=20).fit_transform(arrays)\n",
" \n",
" # Finds t-SNE coordinates for 2 dimensions\n",
" np.set_printoptions(suppress=True)\n",
" \n",
" Y = TSNE(n_components=2, random_state=0, perplexity=15).fit_transform(reduc)\n",
" \n",
" # Sets everything up to plot\n",
" df = pd.DataFrame({'x': [x for x in Y[:, 0]],\n",
" 'y': [y for y in Y[:, 1]],\n",
" 'words': word_labels,\n",
" 'color': color_list})\n",
" \n",
" fig, _ = plt.subplots()\n",
" fig.set_size_inches(9, 9)\n",
" \n",
" # Basic plot\n",
" p1 = sns.regplot(data=df,\n",
" x=\"x\",\n",
" y=\"y\",\n",
" fit_reg=False,\n",
" marker=\"o\",\n",
" scatter_kws={'s': 40,\n",
" 'facecolors': df['color']\n",
" }\n",
" )\n",
" \n",
" # Adds annotations one by one with a loop\n",
" for line in range(0, df.shape[0]):\n",
" p1.text(df[\"x\"][line],\n",
" df['y'][line],\n",
" ' ' + df[\"words\"][line].title(),\n",
" horizontalalignment='left',\n",
" verticalalignment='bottom', size='medium',\n",
" color=df['color'][line],\n",
" weight='normal'\n",
" ).set_size(15)\n",
"\n",
" \n",
" plt.xlim(Y[:, 0].min()-50, Y[:, 0].max()+50)\n",
" plt.ylim(Y[:, 1].min()-50, Y[:, 1].max()+50)\n",
" \n",
" plt.title('t-SNE visualization for {}'.format(word.title()))"
]
},
{
"cell_type": "code",
"execution_count": 35,
"id": "97bdf62c",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-16T00:25:26.611407Z",
"start_time": "2024-06-16T00:25:25.364703Z"
}
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Exception ignored on calling ctypes callback function: .match_module_callback at 0x7fa0f8423f70>\n",
"Traceback (most recent call last):\n",
" File \"/usr/local/lib/python3.8/dist-packages/threadpoolctl.py\", line 400, in match_module_callback\n",
" self._make_module_from_path(filepath)\n",
" File \"/usr/local/lib/python3.8/dist-packages/threadpoolctl.py\", line 515, in _make_module_from_path\n",
" module = module_class(filepath, prefix, user_api, internal_api)\n",
" File \"/usr/local/lib/python3.8/dist-packages/threadpoolctl.py\", line 606, in __init__\n",
" self.version = self.get_version()\n",
" File \"/usr/local/lib/python3.8/dist-packages/threadpoolctl.py\", line 646, in get_version\n",
" config = get_config().split()\n",
"AttributeError: 'NoneType' object has no attribute 'split'\n"
]
},
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"tsnescatterplot(modelLNT2, 'бог_S', [i[0] for i in modelLNT2.wv.most_similar(negative=[\"бог_S\"])])"
]
},
{
"cell_type": "code",
"execution_count": 36,
"id": "45b23917",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-16T00:25:50.147338Z",
"start_time": "2024-06-16T00:25:49.258384Z"
}
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Exception ignored on calling ctypes callback function: .match_module_callback at 0x7fa0fc640dc0>\n",
"Traceback (most recent call last):\n",
" File \"/usr/local/lib/python3.8/dist-packages/threadpoolctl.py\", line 400, in match_module_callback\n",
" self._make_module_from_path(filepath)\n",
" File \"/usr/local/lib/python3.8/dist-packages/threadpoolctl.py\", line 515, in _make_module_from_path\n",
" module = module_class(filepath, prefix, user_api, internal_api)\n",
" File \"/usr/local/lib/python3.8/dist-packages/threadpoolctl.py\", line 606, in __init__\n",
" self.version = self.get_version()\n",
" File \"/usr/local/lib/python3.8/dist-packages/threadpoolctl.py\", line 646, in get_version\n",
" config = get_config().split()\n",
"AttributeError: 'NoneType' object has no attribute 'split'\n"
]
},
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"tsnescatterplot(modelLNT2, 'жизнь_S', [i[0] for i in modelLNT2.wv.most_similar(negative=[\"жизнь_S\"])])"
]
},
{
"cell_type": "code",
"execution_count": 40,
"id": "d74cd56f",
"metadata": {
"ExecuteTime": {
"end_time": "2024-06-16T00:27:15.367573Z",
"start_time": "2024-06-16T00:27:14.269526Z"
}
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Exception ignored on calling ctypes callback function: .match_module_callback at 0x7fa0fa6c4280>\n",
"Traceback (most recent call last):\n",
" File \"/usr/local/lib/python3.8/dist-packages/threadpoolctl.py\", line 400, in match_module_callback\n",
" self._make_module_from_path(filepath)\n",
" File \"/usr/local/lib/python3.8/dist-packages/threadpoolctl.py\", line 515, in _make_module_from_path\n",
" module = module_class(filepath, prefix, user_api, internal_api)\n",
" File \"/usr/local/lib/python3.8/dist-packages/threadpoolctl.py\", line 606, in __init__\n",
" self.version = self.get_version()\n",
" File \"/usr/local/lib/python3.8/dist-packages/threadpoolctl.py\", line 646, in get_version\n",
" config = get_config().split()\n",
"AttributeError: 'NoneType' object has no attribute 'split'\n"
]
},
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"tsnescatterplot(modelLNT2, 'друг_S', [i[0] for i in modelLNT2.wv.most_similar(negative=[\"друг_S\"])])"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "3be42745",
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"hide_input": false,
"kernelspec": {
"display_name": "Python 3",
"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": "3.8.10"
},
"toc": {
"base_numbering": 1,
"nav_menu": {},
"number_sections": true,
"sideBar": true,
"skip_h1_title": true,
"title_cell": "Содержание",
"title_sidebar": "Contents",
"toc_cell": true,
"toc_position": {},
"toc_section_display": true,
"toc_window_display": true
},
"varInspector": {
"cols": {
"lenName": 16,
"lenType": 16,
"lenVar": 40
},
"kernels_config": {
"python": {
"delete_cmd_postfix": "",
"delete_cmd_prefix": "del ",
"library": "var_list.py",
"varRefreshCmd": "print(var_dic_list())"
},
"r": {
"delete_cmd_postfix": ") ",
"delete_cmd_prefix": "rm(",
"library": "var_list.r",
"varRefreshCmd": "cat(var_dic_list()) "
}
},
"types_to_exclude": [
"module",
"function",
"builtin_function_or_method",
"instance",
"_Feature"
],
"window_display": false
}
},
"nbformat": 4,
"nbformat_minor": 5
}