mirror of
https://github.com/ivabus/pantry
synced 2024-11-23 08:55:07 +03:00
d8a3e7c646
llama.cpp -p "Getting paid to write open source can be accomplished in 3 simple steps:"
43 lines
1.1 KiB
YAML
43 lines
1.1 KiB
YAML
distributable:
|
|
url: https://github.com/google/sentencepiece/archive/v{{version}}.tar.gz
|
|
strip-components: 1
|
|
|
|
versions:
|
|
github: google/sentencepiece/releases/tags
|
|
ignore: '1.0.0' # whoops, who tagged this then?
|
|
|
|
build:
|
|
dependencies:
|
|
tea.xyz/gx/cc: c99
|
|
tea.xyz/gx/make: '*'
|
|
cmake.org: ^3
|
|
python.org: ^3
|
|
pip.pypa.io: '*'
|
|
freedesktop.org/pkg-config: ~0.29
|
|
working-directory: build
|
|
script: |
|
|
cmake .. -DCMAKE_INSTALL_PREFIX="{{prefix}}" -DCMAKE_BUILD_TYPE=Release
|
|
make --jobs {{ hw.concurrency }} install
|
|
|
|
pip install ../python --verbose --prefix="{{prefix}}"
|
|
|
|
provides:
|
|
- bin/spm_decode
|
|
- bin/spm_encode
|
|
- bin/spm_export_vocab
|
|
- bin/spm_normalize
|
|
- bin/spm_train
|
|
|
|
runtime:
|
|
env:
|
|
PYTHONPATH: '{{prefix}}/lib/python3.11/site-packages'
|
|
|
|
test:
|
|
dependencies:
|
|
gnu.org/wget: '*'
|
|
python.org: ^3
|
|
script: |
|
|
wget https://raw.githubusercontent.com/google/sentencepiece/master/data/botchan.txt
|
|
spm_train --input=botchan.txt --model_prefix=m --vocab_size=1000
|
|
|
|
python -c 'import sentencepiece as spm; spm.SentencePieceProcessor()'
|