.SRCINFO


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

pkgbase = python-trl
	pkgdesc = Train transformer language models with reinforcement learning
	pkgver = 0.28.0
	pkgrel = 1
	url = https://github.com/huggingface/trl
	arch = any
	groups = huggingface
	license = Apache-2.0
	makedepends = python-build
	makedepends = python-installer
	makedepends = python-setuptools
	makedepends = python-wheel
	depends = python-accelerate
	depends = python-datasets
	depends = python-packaging
	depends = python-transformers>=4.56.1
	optdepends = python-bitsandbytes: Quantization support
	optdepends = python-deepspeed: Low-level routines for distributed training
	optdepends = python-diffusers: Diffusion models
	optdepends = python-peft: Parameter efficient fine-tuning
	optdepends = python-rich: Fancy text, tables, progress bars, and so on
	source = python-trl-0.28.0.tar.gz::https://github.com/lvwerra/trl/archive/refs/tags/v0.28.0.tar.gz
	sha256sums = 5d197c929c7a1e1b2edc694fe7bd11b03e15ae818f51472a3203b1e056d65d7b

pkgname = python-trl