pytorch-nlp-setup
Set up Python environments for NLP/ML projects using PyTorch, transformers, and TRL. Use this skill when installing dependencies for preference optimization, RLHF, or transformer-based training pipelines.
- License 许可
- MIT
- 最近更新
- 2026-10-02
- Stars
- ★ 93
- Forks
- 8
01 前言 Front Matter
- name
- pytorch-nlp-setup
- description
- Set up Python environments for NLP/ML projects using PyTorch, transformers, and TRL. Use this skill when installing dependencies for preference optimization, RLHF, or transformer-based training pipelines.
03 SKILL.md 正文
PyTorch NLP Environment Setup
Standard Stack for Preference Optimization Projects
pip install torch transformers datasets accelerate trl peft wandb numpy
Version Compatibility Notes
- TRL (Transformer Reinforcement Learning) provides base trainers like CPOTrainer, DPOTrainer
- Older TRL versions (< 0.8) have different import paths for utilities
- Check
from trl.trainer.utils import DPODataCollatorWithPaddingavailability - For
trl_sanitze_kwargs_for_tagging(note: typo is intentional in some versions)
Common Issues
- If
from trl.import_utils import is_peft_availablefails, check TRL version - Some projects pin specific transformers/TRL versions — check requirements.txt or setup.py
- CUDA availability: use
torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
Running Unit Tests
cd /path/to/project && python -m pytest unit_test/ -v
# or
cd /path/to/project && python -m unittest unit_test.unit_test_1
04 同仓库更多 Skills
object_counter
Count occurrences of an object in the image using computer vision algorithm.
travel-planning-workflow
Invoke this skill when performing travel planning. It defines a structured workflow for generating itineraries. Following this workflow is essential to ensure the final plan satisfies user requirements.
bundled-dataset-search-tool
invoke this skill when you need to perform database search for travel planning. This skill provides some useful pre-packaged tools to look up accommodations, attractions, cities, driving distance, flights, and restaurants from the bundled dataset.
glm-output
Output-processing guidance for GLM tasks. Especially useful after glm-basics and glm-calibration when you need verifier-matching metrics from output.nc and a final /root/metrics.json.
glm-calibration
Calibration guidance for GLM tasks. Often most effective after glm-basics has clarified the setup; glm-output is the companion skill for exact final metric computation.
glm-basics
A good starting point for GLM calibration tasks. Use to inspect glm3.nml, confirm how GLM runs, and identify the relevant files before moving on to calibration and output evaluation.