Tina

textvqa_grounding_task_qwen2.5-vl-ft. Jupyter Notebook

deepseek-llm-7B-chat-lora-ft. 使用多轮对话数据集对deepseek进行lora微调教程

sudoku_trl_grpo. 基于trl框架对Qwen模型做grpo训练，从而完成4*4数独游戏的训练任务

cosyvoice-paimon-sft. Fine-tune the Paimon speech using the CosyVoice2 model

qwen3-ft-swift. Python

evalscope_qwen3_eval. 使用evalscope工具对qwen3微调后模型进行评估

llama3-fine-tuning-aplace-zh. llama3微调训练+测试结果

llmcourse-docs. This is a downloadable large model tutorial document moved from Zhihu to github. The author on Zhihu, Tina, is myself, and the rest are reprints

transformer_test. Jupyter Notebook

llama2-7B-instrcut-finetune-swanlab. Python

BERT-Chinese-QA. 基于BERT的中文问答微调训练