File size: 2,638 Bytes
097ab1f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import selfies as sf\n",
    "from tokenizers import Tokenizer\n",
    "from tokenizers.models import WordLevel\n",
    "from tokenizers.pre_tokenizers import Split\n",
    "from tokenizers.processors import TemplateProcessing\n",
    "from tokenizers.trainers import WordLevelTrainer\n",
    "from tqdm import tqdm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(\"./train.txt\") as f:\n",
    "    smiles = [line.strip() for line in f]\n",
    "\n",
    "selfies = []\n",
    "for smile in tqdm(smiles):\n",
    "    try:\n",
    "        selfies.append(sf.encoder(smile))\n",
    "    except:\n",
    "        pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tokenizer = Tokenizer(WordLevel(unk_token=\"<UNK>\"))\n",
    "\n",
    "tokenizer.pre_tokenizer = Split(\n",
    "    pattern=\"]\", \n",
    "    behavior=\"merged_with_previous\"\n",
    ")\n",
    "\n",
    "trainer = WordLevelTrainer(\n",
    "    special_tokens=[\"<CLS>\", \"<EOS>\", \"<PAD>\", \"<UNK>\"]\n",
    ")\n",
    "\n",
    "tokenizer.train_from_iterator(selfies, trainer=trainer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tokenizer.post_processor = TemplateProcessing(\n",
    "    single=\"<CLS> $A <EOS>\",\n",
    "    special_tokens=[\n",
    "        (\"<CLS>\", tokenizer.token_to_id(\"<CLS>\")),\n",
    "        (\"<EOS>\", tokenizer.token_to_id(\"<EOS>\")),\n",
    "    ],\n",
    ")\n",
    "\n",
    "tokenizer.enable_padding(\n",
    "    direction=\"right\",\n",
    "    pad_id=tokenizer.token_to_id(\"<PAD>\"),\n",
    "    pad_token=\"<PAD>\",\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tokenizer.save(\"./tokenizer.json\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "ddpm",
   "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": "3.11.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}