{tools}[foss/2023a] SentencePiece v0.2.0, gperftools v2.15, huggingface-hub v0.20.3, ... w/ PyTorch 2.1.2 CUDA 12.1.1

easybuild/easyconfigs/g/gperftools/gperftools-2.15-GCCcore-12.3.0.eb

@@ -0,0 +1,38 @@
+easyblock = 'ConfigureMake'
+
+name = 'gperftools'
+version = '2.15'
+
+homepage = 'https://github.com/gperftools/gperftools'
+description = """
+gperftools is a collection of a high-performance multi-threaded malloc()
+implementation, plus some pretty nifty performance analysis tools.
+Includes TCMalloc, heap-checker, heap-profiler and cpu-profiler.
+"""
+
+toolchain = {'name': 'GCCcore', 'version': '12.3.0'}
+
+github_account = 'gperftools'
+source_urls = [GITHUB_SOURCE]
+sources = [SOURCE_TAR_GZ]
+checksums = ['3918ff2e21bb3dbb5a801e1daf55fb20421906f7c42fbb482bede7bdc15dfd2e']
+
+builddependencies = [
+    ('Autotools', '20220317'),
+    ('binutils', '2.40'),
+]
+
+dependencies = [
+    ('libunwind', '1.6.2'),
+]
+
+preconfigopts = "autoreconf -f -i && "
+configopts = '--enable-libunwind'
+
+sanity_check_paths = {
+    'files': ['bin/pprof', 'lib/libprofiler.a', 'lib/libprofiler.%s' % SHLIB_EXT,
+              'lib/libtcmalloc.a', 'lib/libtcmalloc.%s' % SHLIB_EXT],
+    'dirs': ['include'],
+}
+
+moduleclass = 'tools'

easybuild/easyconfigs/h/HyperPyYAML/HyperPyYAML-1.2.2-foss-2023a.eb

@@ -0,0 +1,30 @@
+easyblock = 'PythonPackage'
+
+name = 'HyperPyYAML'
+version = '1.2.2'
+
+homepage = 'https://github.com/speechbrain/HyperPyYAML'
+description = """A crucial element of systems for data-analysis is laying out all the
+ hyperparameters of that system so they can be easily examined and modified. We add a
+ few useful extensions to a popular human-readable data-serialization language known as
+ YAML (YAML Ain't Markup Language). This provides support for a rather expansive idea
+ of what constitutes a hyperparameter, and cleans up python files for data analysis to
+ just the bare algorithm."""
+
+toolchain = {'name': 'foss', 'version': '2023a'}
+
+sources = [name + '-%(version)s.tar.gz']
+checksums = ['bdb734210d18770a262f500fe5755c7a44a5d3b91521b06e24f7a00a36ee0f87']
+
+dependencies = [
+    ('Python', '3.11.3'),
+    ('PyYAML', '6.0'),
+    ('ruamel.yaml', '0.17.32'),
+]
+
+
+download_dep_fail = True
+use_pip = True
+sanity_pip_check = True
+
+moduleclass = 'tools'

easybuild/easyconfigs/h/huggingface-hub/huggingface-hub-0.20.3-foss-2023a.eb

@@ -0,0 +1,27 @@
+easyblock = 'PythonPackage'
+
+name = 'huggingface-hub'
+version = '0.20.3'
+
+homepage = 'https://github.com/huggingface/datasets'
+description = """The largest hub of ready-to-use datasets for ML models with
+fast, easy-to-use and efficient data manipulation tools"""
+
+toolchain = {'name': 'foss', 'version': '2023a'}
+
+sources = ['huggingface_hub-%(version)s.tar.gz']
+checksums = ['94e7f8e074475fbc67d6a71957b678e1b4a74ff1b64a644fd6cbb83da962d05d']
+
+dependencies = [
+    ('Python', '3.11.3'),
+    ('Python-bundle-PyPI', '2023.06'),
+    ('tqdm', '4.66.1'),
+    ('PyYAML', '6.0'),
+    ('typing-extensions', '4.9.0'),
+]
+
+download_dep_fail = True
+use_pip = True
+sanity_pip_check = True
+
+moduleclass = 'lib'

easybuild/easyconfigs/l/libmad/libmad-0.15.1b-GCCcore-12.3.0.eb

@@ -0,0 +1,27 @@
+easyblock = 'ConfigureMake'
+
+name = 'libmad'
+version = '0.15.1b'
+
+homepage = 'https://www.underbit.com/products/mad/'
+description = """MAD is a high-quality MPEG audio decoder."""
+
+toolchain = {'name': 'GCCcore', 'version': '12.3.0'}
+
+source_urls = ['https://sourceforge.net/projects/mad/files/%(name)s/%(version)s/']
+sources = [SOURCELOWER_TAR_GZ]
+patches = ['libmad-0.15.1b-remove-depreciated-gcc-option.patch']
+checksums = [
+    'bbfac3ed6bfbc2823d3775ebb931087371e142bb0e9bb1bee51a76a6e0078690',  # libmad-0.15.1b.tar.gz
+    # libmad-0.15.1b-remove-depreciated-gcc-option.patch
+    '8f96a23a22ba66e62f32e20064d01f4c7f6a18ba0aab85d3be9ce63794b2c678',
+]
+
+builddependencies = [('binutils', '2.40')]
+
+sanity_check_paths = {
+    'files': ['include/mad.h', 'lib/libmad.a', 'lib/libmad.la', 'lib/libmad.%s' % SHLIB_EXT],
+    'dirs': ['include', 'lib', 'lib64']
+}
+
+moduleclass = 'lib'

easybuild/easyconfigs/p/python-soundfile/python-soundfile-0.12.1-foss-2023a.eb

@@ -0,0 +1,27 @@
+easyblock = 'PythonPackage'
+
+name = 'python-soundfile'
+version = '0.12.1'
+
+homepage = 'https://github.com/bastibe/python-soundfile'
+description = """ The soundfile module can read and write sound files based on
+libsndfile, CFFI, and NumPy """
+
+toolchain = {'name': 'foss', 'version': '2023a'}
+
+sources = ['soundfile-%(version)s.tar.gz']
+checksums = ['e8e1017b2cf1dda767aef19d2fd9ee5ebe07e050d430f77a0a7c66ba08b8cdae']
+
+dependencies = [
+    ('Python', '3.11.3'),
+    ('SciPy-bundle', '2023.07'),
+    ('libsndfile', '1.2.2'),
+]
+
+download_dep_fail = True
+use_pip = True
+sanity_pip_check = True
+
+options = {'modulename': 'soundfile'}
+
+moduleclass = 'lib'

easybuild/easyconfigs/s/SentencePiece/SentencePiece-0.2.0-GCC-12.3.0.eb

@@ -0,0 +1,63 @@
+easyblock = 'Bundle'
+
+name = 'SentencePiece'
+version = '0.2.0'
+
+homepage = 'https://github.com/google/sentencepiece'
+description = "Unsupervised text tokenizer for Neural Network-based text generation."
+github_account = 'google'
+
+toolchain = {'name': 'GCC', 'version': '12.3.0'}
+
+builddependencies = [
+    ('CMake', '3.26.3'),
+]
+
+dependencies = [
+    ('Python', '3.11.3'),
+    ('gperftools', '2.15'),
+    ('protobuf', '24.0'),
+]
+
+default_component_specs = {
+    'source_urls': [GITHUB_LOWER_SOURCE],
+    'sources': ['v%(version)s.tar.gz'],
+    'checksums': ['9970f0a0afee1648890293321665e5b2efa04eaec9f1671fcf8048f456f5bb86'],
+}
+
+components = [
+    (name, version, {
+        'easyblock': 'CMakeMake',
+        'separate_build_dir': True,
+        'start_dir': '%(namelower)s-%(version)s',
+    }),
+    ('sentencepiece', version, {
+        'easyblock': 'PythonPackage',
+        'start_dir': '%(namelower)s-%(version)s/python',
+        'prebuildopts': 'export PKG_CONFIG_PATH=%(installdir)s/lib/pkgconfig:$PKG_CONFIG_PATH && ',
+        'preinstallopts': 'export PKG_CONFIG_PATH=%(installdir)s/lib/pkgconfig:$PKG_CONFIG_PATH && ',
+        'use_pip': True,
+        'download_dep_fail': True,
+        'sanity_pip_check': True,
+    }),
+]
+
+postinstallcmds = ['cp -a %(builddir)s/%(namelower)s-%(version)s/{data,doc} %(installdir)s/']
+
+sanity_check_paths = {
+    'files': ['bin/spm_%s' % x for x in ['decode', 'encode', 'export_vocab', 'normalize', 'train']] +
+             ['lib/libsentencepiece.%s' % SHLIB_EXT, 'lib/libsentencepiece_train.%s' % SHLIB_EXT] +
+             ['include/sentencepiece_processor.h', 'include/sentencepiece_trainer.h'],
+    'dirs': ['lib/python%(pyshortver)s/site-packages', 'data', 'doc'],
+}
+
+sanity_check_commands = [
+    'spm_train --help',
+    "python -c 'import sentencepiece'",
+]
+
+modextrapaths = {
+    'PYTHONPATH': ['lib/python%(pyshortver)s/site-packages']
+}
+
+moduleclass = 'lib'

easybuild/easyconfigs/s/SoX/SoX-14.4.2-GCCcore-12.3.0.eb

@@ -0,0 +1,34 @@
+easyblock = 'ConfigureMake'
+
+name = 'SoX'
+version = '14.4.2'
+
+homepage = 'http://sox.sourceforge.net/'
+docurls = 'http://sox.sourceforge.net/Docs/Documentation'
+description = """Sound eXchange, the Swiss Army knife of audio manipulation"""
+
+toolchain = {'name': 'GCCcore', 'version': '12.3.0'}
+
+source_urls = [SOURCEFORGE_SOURCE]
+sources = [SOURCELOWER_TAR_GZ]
+checksums = ['b45f598643ffbd8e363ff24d61166ccec4836fea6d3888881b8df53e3bb55f6c']
+
+builddependencies = [('binutils', '2.40')]
+
+dependencies = [
+    ('FLAC', '1.4.2'),
+    ('LAME', '3.100'),
+    ('libmad', '0.15.1b'),
+    ('libvorbis', '1.3.7'),
+    ('FFmpeg', '6.0'),
+]
+
+sanity_check_paths = {
+    'files': ['bin/play', 'bin/rec', 'bin/sox', 'bin/soxi', 'include/sox.h',
+              'lib/libsox.la', 'lib/libsox.a', 'lib/pkgconfig/sox.pc', 'lib/libsox.%s' % SHLIB_EXT],
+    'dirs': ['bin', 'include', 'lib', 'lib/pkgconfig', 'share/man'],
+}
+
+sanity_check_commands = ['sox --help']
+
+moduleclass = 'data'

easybuild/easyconfigs/s/speechbrain/speechbrain-0.5.16-foss-2023a-PyTorch-2.1.2-CUDA-12.1.1.eb

@@ -0,0 +1,33 @@
+easyblock = 'PythonPackage'
+
+name = 'speechbrain'
+version = '0.5.16'
+local_pytorch_version = '2.1.2'
+versionsuffix = '-PyTorch-' + local_pytorch_version + '-CUDA-%(cudaver)s'
+
+homepage = 'https://speechbrain.github.io/'
+description = "An Open-Source Conversational AI Toolkit"
+
+toolchain = {'name': 'foss', 'version': '2023a'}
+
+sources = [SOURCELOWER_TAR_GZ]
+checksums = ['cfbf6176da8879b541e1a2d1342d43806e82d0ef55604c2f38e21a5027a2eecb']
+
+dependencies = [
+    ('CUDA', '12.1.1', '', SYSTEM),
+    ('Python', '3.11.3'),
+    ('SciPy-bundle', '2023.07'),
+    ('PyTorch', local_pytorch_version, '-CUDA-%(cudaver)s'),
+    ('python-soundfile', '0.12.1'),
+    ('torchaudio', '2.2.0', versionsuffix),
+    ('huggingface-hub', '0.20.3'),
+    ('SentencePiece', '0.2.0'),
+    ('HyperPyYAML', '1.2.2'),
+]
+
+
+download_dep_fail = True
+use_pip = True
+sanity_pip_check = True
+
+moduleclass = 'tools'

easybuild/easyconfigs/t/torchaudio/torchaudio-2.2.0-foss-2023a-PyTorch-2.1.2-CUDA-12.1.1.eb

@@ -0,0 +1,49 @@
+easyblock = 'PythonPackage'
+
+name = 'torchaudio'
+version = '2.2.0'
+local_pytorch_version = '2.1.2'
+versionsuffix = '-PyTorch-' + local_pytorch_version + '-CUDA-%(cudaver)s'
+
+homepage = 'https://github.com/pytorch/audio'
+description = """ Data manipulation and transformation for audio signal
+processing, powered by PyTorch """
+
+toolchain = {'name': 'foss', 'version': '2023a'}
+
+sources = [{
+    'filename': '%(name)s-%(version)s.tar.gz',
+    'git_config': {
+        'url': 'https://github.com/pytorch',
+        'repo_name': 'audio',
+        'tag': 'v%(version)s',
+        'recursive': True,
+        'keep_git_dir': True,
+    }
+}]
+
+checksums = [None]
+
+builddependencies = [
+    ('binutils', '2.40'),
+    ('CMake', '3.26.3'),
+    ('Ninja', '1.11.1'),
+]
+
+dependencies = [
+    ('CUDA', '12.1.1', '', SYSTEM),
+    ('Python', '3.11.3'),
+    ('SciPy-bundle', '2023.07'),
+    ('PyTorch', local_pytorch_version, '-CUDA-%(cudaver)s'),
+    ('SoX', '14.4.2'),
+    ('FFmpeg', '6.0'),
+]
+
+preinstallopts = 'USE_FFMPEG=1'
+installopts = '--no-use-pep517'
+
+download_dep_fail = True
+use_pip = True
+sanity_pip_check = True
+
+moduleclass = 'tools'