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
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
|
{ lib
, stdenv
, buildPythonPackage
, cargo
, datasets
, fetchFromGitHub
, fetchurl
, libiconv
, numpy
, openssl
, pkg-config
, pytestCheckHook
, pythonOlder
, requests
, rustPlatform
, rustc
, Security
, setuptools-rust
}:
let
# See https://github.com/huggingface/tokenizers/blob/main/bindings/python/tests/utils.py for details
# about URLs and file names
robertaVocab = fetchurl {
url = "https://s3.amazonaws.com/models.huggingface.co/bert/roberta-base-vocab.json";
sha256 = "0m86wpkfb2gdh9x9i9ng2fvwk1rva4p0s98xw996nrjxs7166zwy";
};
robertaMerges = fetchurl {
url = "https://s3.amazonaws.com/models.huggingface.co/bert/roberta-base-merges.txt";
sha256 = "1idd4rvkpqqbks51i2vjbd928inw7slij9l4r063w3y5fd3ndq8w";
};
albertVocab = fetchurl {
url = "https://s3.amazonaws.com/models.huggingface.co/bert/albert-base-v1-tokenizer.json";
sha256 = "1hra9pn8rczx7378z88zjclw2qsdrdwq20m56sy42s2crbas6akf";
};
bertVocab = fetchurl {
url = "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-uncased-vocab.txt";
sha256 = "18rq42cmqa8zanydsbzrb34xwy4l6cz1y900r4kls57cbhvyvv07";
};
norvigBig = fetchurl {
url = "https://norvig.com/big.txt";
sha256 = "0yz80icdly7na03cfpl0nfk5h3j3cam55rj486n03wph81ynq1ps";
};
docPipelineTokenizer = fetchurl {
url = "https://s3.amazonaws.com/models.huggingface.co/bert/anthony/doc-pipeline/tokenizer.json";
hash = "sha256-i533xC8J5CDMNxBjo+p6avIM8UOcui8RmGAmK0GmfBc=";
};
docQuicktourTokenizer = fetchurl {
url = "https://s3.amazonaws.com/models.huggingface.co/bert/anthony/doc-quicktour/tokenizer.json";
hash = "sha256-ipY9d5DR5nxoO6kj7rItueZ9AO5wq9+Nzr6GuEIfIBI=";
};
openaiVocab = fetchurl {
url = "https://s3.amazonaws.com/models.huggingface.co/bert/openai-gpt-vocab.json";
sha256 = "0y40gc9bixj5rxv674br1rxmxkd3ly29p80x1596h8yywwcrpx7x";
};
openaiMerges = fetchurl {
url = "https://s3.amazonaws.com/models.huggingface.co/bert/openai-gpt-merges.txt";
sha256 = "09a754pm4djjglv3x5pkgwd6f79i2rq8ydg0f7c3q1wmwqdbba8f";
};
in
buildPythonPackage rec {
pname = "tokenizers";
version = "0.13.3";
disabled = pythonOlder "3.7";
src = fetchFromGitHub {
owner = "huggingface";
repo = pname;
rev = "python-v${version}";
hash = "sha256-QZG5jmr3vbyQs4mVBjwVDR31O66dUM+p39R0htJ1umk=";
};
postPatch = ''
ln -s ${./Cargo.lock} Cargo.lock
'';
cargoDeps = rustPlatform.importCargoLock {
lockFile = ./Cargo.lock;
};
sourceRoot = "source/bindings/python";
nativeBuildInputs = [
pkg-config
setuptools-rust
rustPlatform.cargoSetupHook
cargo
rustc
];
buildInputs = [
openssl
] ++ lib.optionals stdenv.isDarwin [
libiconv
Security
];
propagatedBuildInputs = [
numpy
];
nativeCheckInputs = [
datasets
pytestCheckHook
requests
];
postUnpack = ''
# Add data files for tests, otherwise tests attempt network access
mkdir $sourceRoot/tests/data
( cd $sourceRoot/tests/data
ln -s ${robertaVocab} roberta-base-vocab.json
ln -s ${robertaMerges} roberta-base-merges.txt
ln -s ${albertVocab} albert-base-v1-tokenizer.json
ln -s ${bertVocab} bert-base-uncased-vocab.txt
ln -s ${docPipelineTokenizer} bert-wiki.json
ln -s ${docQuicktourTokenizer} tokenizer-wiki.json
ln -s ${norvigBig} big.txt
ln -s ${openaiVocab} openai-gpt-vocab.json
ln -s ${openaiMerges} openai-gpt-merges.txt )
'';
preCheck = ''
export HOME=$(mktemp -d);
'';
pythonImportsCheck = [
"tokenizers"
];
disabledTests = [
# Downloads data using the datasets module
"TestTrainFromIterators"
# Those tests require more data
"test_from_pretrained"
"test_from_pretrained_revision"
"test_continuing_prefix_trainer_mistmatch"
];
meta = with lib; {
description = "Fast State-of-the-Art Tokenizers optimized for Research and Production";
homepage = "https://github.com/huggingface/tokenizers";
license = licenses.asl20;
maintainers = with maintainers; [ ];
platforms = platforms.unix;
};
}
|