12 2024 档案
摘要：LB.4 推理框架 主求解函数(Solver Function) # Solver Function # 主求解函数，循环生成多轮样本并提取最终答案 from tqdm import tqdm num_generations = 2 # 每轮生成 128 个样本，设置总轮次数 def solve(q
阅读全文
摘要：Q-learning(Notebook) environment !apt-get update !apt install -y python3.9 !pip install virtualenv %cd /kaggle/working !virtualenv venv -p $(which pyt
阅读全文
摘要：export KAGGLE_CONFIG_DIR=/home/user/kaggle_config
阅读全文
摘要：各版本python的Notebook !apt-get update !apt install -y python3.9 !pip install virtualenv %cd /kaggle/working !virtualenv venv -p $(which python3.9) # !vir
阅读全文
摘要：LoRA (Low-Rank Adaptation) LoRA官方文档 Qwen2.5-0.5B微调Notebook Data preprocess pip&&import !pip config !pip install modelscope==1.18.0 !pip install transf
阅读全文
摘要：https://zhuanlan.zhihu.com/p/681353195?utm_campaign=shareopn&utm_medium=social&utm_psn=1854692762752008192&utm_source=wechat_session
阅读全文
摘要：https://www.latexlive.com/ https://zhuanlan.zhihu.com/p/702423411
阅读全文
摘要：VMware虚拟机突然连接不上网络 https://blog.csdn.net/dong__ge/article/details/123581117 VMware 虚拟机克隆详细教程 https://blog.csdn.net/weixin_36665875/article/details/1063
阅读全文
摘要：https://github.com/Linjunjie99/RL-LLM-DT
阅读全文
摘要：https://github.com/GAIR-NLP/O1-Journey#about-the-team
阅读全文
摘要：exploring the GPT-2 (124M) OpenAI checkpoint pipeline # https://www.bilibili.com/video/BV12s421u7sZ?spm_id_from=333.788.videopod.sections&vd_source=06
阅读全文
