-
Notifications
You must be signed in to change notification settings - Fork 91
/
pyproject.toml
186 lines (176 loc) · 7.46 KB
/
pyproject.toml
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
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
# Copyright (c) 2024, NVIDIA CORPORATION. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
[build-system]
requires = ["setuptools"]
build-backend = "setuptools.build_meta"
[project]
name = "nemo_curator"
description = "Scalable Data Preprocessing Tool for Training Large Language Models"
readme = { file = "README.md", content-type = "text/markdown" }
authors = [
{ name = "Joseph Jennings", email = "[email protected]" },
{ name = "Mostofa Patwary", email = "[email protected]" },
{ name = "Sandeep Subramanian", email = "[email protected]" },
{ name = "Shrimai Prabhumoye", email = "[email protected]" },
{ name = "Ayush Dattagupta", email = "[email protected]" },
{ name = "Vibhu Jawa", email = "[email protected]" },
{ name = "Jiwei Liu", email = "[email protected]" },
{ name = "Ryan Wolf", email = "[email protected]" },
{ name = "Sarah Yurick", email = "[email protected]" },
]
classifiers = [
"Development Status :: 3 - Alpha",
"Programming Language :: Python :: 3",
"Programming Language :: Python :: 3.10",
]
requires-python = ">=3.10, <3.11"
dependencies = [
"awscli>=1.22.55",
"beautifulsoup4",
"charset_normalizer>=3.1.0",
"comment_parser",
# TODO: Pin CrossFit 0.0.8 when it is released
"crossfit @ git+https://github.com/rapidsai/crossfit.git@main",
"dask-mpi>=2021.11.0",
"dask[complete]>=2021.7.1",
"datasets",
"distributed>=2021.7.1",
"fasttext==0.9.3",
"ftfy==6.1.1",
"in-place==0.5.0",
"jieba==0.42.1",
"justext==3.0.1",
"lxml_html_clean",
"mecab-python3",
"mwparserfromhell==0.6.5",
"numpy<2",
"openai",
"peft",
"platformdirs",
"presidio-analyzer==2.2.351",
"presidio-anonymizer==2.2.351",
"pycld2",
"resiliparse",
"sentencepiece",
"spacy>=3.6.0, <3.8.0",
# TODO: Remove this pin once newer version is released
"transformers==4.46.3",
"unidic-lite==1.0.8",
"usaddress==0.5.10",
"warcio==1.7.4",
"zstandard==0.18.0",
]
dynamic = ["version"]
[project.optional-dependencies]
# Installs CPU + GPU text curation modules
cuda12x = [
"cudf-cu12>=24.12",
"cugraph-cu12>=24.12",
"cuml-cu12>=24.12",
"dask-cuda>=24.12",
"dask-cudf-cu12>=24.12",
"spacy[cuda12x]>=3.6.0, <3.8.0",
]
# Installs CPU + GPU text curation modules with RAPIDS Nightlies
cuda12x_nightly = [
"cudf-cu12>=25.02.0a0,<=25.02",
"cugraph-cu12>=25.02.0a0,<=25.02",
"cuml-cu12>=25.02.0a0,<=25.02",
"dask-cuda>=25.02.0a0,<=25.02",
"dask-cudf-cu12>=25.02.0a0,<=25.02",
"spacy[cuda12x]>=3.6.0, <3.8.0",
]
# Installs CPU + GPU text and image curation modules
image = [
"nvidia-dali-cuda120",
"nvidia-nvjpeg2k-cu12",
"timm>=1.0.8",
"nemo_curator[cuda12x]",
]
# Installs CPU + GPU text and image curation modules with RAPIDS Nightlies
image_nightly = [
"nvidia-dali-cuda120",
"nvidia-nvjpeg2k-cu12",
"timm>=1.0.8",
"nemo_curator[cuda12x_nightly]",
]
# Installs bitext curation modules
bitext = [
"huggingface-hub",
"tqdm",
"transformers",
"nemo_curator[cuda12x]",
]
# Installs all of the above with Stable RAPIDS
all = [
"nemo_curator[image]",
"nemo_curator[bitext]",
]
# Installs all of the above with RAPIDS Nightlies
all_nightly = [
"nemo_curator[image_nightly]",
]
[project.scripts]
get_common_crawl_urls = "nemo_curator.scripts.get_common_crawl_urls:console_script"
get_wikipedia_urls = "nemo_curator.scripts.get_wikipedia_urls:console_script"
download_and_extract = "nemo_curator.scripts.download_and_extract:console_script"
text_cleaning = "nemo_curator.scripts.text_cleaning:console_script"
add_id = "nemo_curator.scripts.add_id:console_script"
make_data_shards = "nemo_curator.scripts.make_data_shards:console_script"
prepare_fasttext_training_data = "nemo_curator.scripts.prepare_fasttext_training_data:console_script"
train_fasttext = "nemo_curator.scripts.train_fasttext:console_script"
filter_documents = "nemo_curator.scripts.filter_documents:console_script"
separate_by_metadata = "nemo_curator.scripts.separate_by_metadata:console_script"
prepare_task_data = "nemo_curator.scripts.prepare_task_data:console_script"
find_matching_ngrams = "nemo_curator.scripts.find_matching_ngrams:console_script"
remove_matching_ngrams = "nemo_curator.scripts.remove_matching_ngrams:console_script"
gpu_compute_minhashes = "nemo_curator.scripts.fuzzy_deduplication.compute_minhashes:console_script"
minhash_buckets = "nemo_curator.scripts.fuzzy_deduplication.minhash_lsh:console_script"
jaccard_map_buckets = "nemo_curator.scripts.fuzzy_deduplication.map_buckets:console_script"
jaccard_shuffle = "nemo_curator.scripts.fuzzy_deduplication.jaccard_shuffle:console_script"
jaccard_compute = "nemo_curator.scripts.fuzzy_deduplication.jaccard_compute:console_script"
gpu_connected_component = "nemo_curator.scripts.fuzzy_deduplication.connected_components:console_script"
buckets_to_edges = "nemo_curator.scripts.fuzzy_deduplication.buckets_to_edges:console_script"
gpu_exact_dups = "nemo_curator.scripts.find_exact_duplicates:console_script"
deidentify = "nemo_curator.scripts.find_pii_and_deidentify:console_script"
domain_classifier_inference = "nemo_curator.scripts.classifiers.domain_classifier_inference:console_script"
quality_classifier_inference = "nemo_curator.scripts.classifiers.quality_classifier_inference:console_script"
aegis_classifier_inference = "nemo_curator.scripts.classifiers.aegis_classifier_inference:console_script"
fineweb_edu_classifier_inference = "nemo_curator.scripts.classifiers.fineweb_edu_classifier_inference:console_script"
instruction_data_guard_classifier_inference = "nemo_curator.scripts.classifiers.instruction_data_guard_classifier_inference:console_script"
multilingual_domain_classifier_inference = "nemo_curator.scripts.classifiers.multilingual_domain_classifier_inference:console_script"
content_type_classifier_inference = "nemo_curator.scripts.classifiers.content_type_classifier_inference:console_script"
prompt_task_complexity_classifier_inference = "nemo_curator.scripts.classifiers.prompt_task_complexity_classifier_inference:console_script"
verify_classification_results = "nemo_curator.scripts.verify_classification_results:console_script"
blend_datasets = "nemo_curator.scripts.blend_datasets:console_script"
semdedup_extract_embeddings = "nemo_curator.scripts.semdedup.compute_embeddings:console_script"
semdedup_clustering = "nemo_curator.scripts.semdedup.clustering:console_script"
semdedup_extract_unique_ids = "nemo_curator.scripts.semdedup.extract_dedup_data:console_script"
[project.urls]
Homepage = "https://github.com/NVIDIA/NeMo-Curator"
[tool.black]
line-length = 88
[tool.isort]
profile = "black" # black-compatible
line_length = 88 # should match black parameters
py_version = 310
[tool.pytest.ini_options]
markers = [
"gpu: marks tests as GPU tests (deselect with '-m \"not gpu\"')"
]
[tool.setuptools.dynamic]
version = { attr = "nemo_curator.package_info.__version__" }
[tool.setuptools.packages.find]
include = ["*"]
exclude = ["tests", "tests.*"]