datawhalechina/self-llm issues and pull requests

#275 - LLaMA3_1-8B-Instruct Lora 微调数据格式化问题

Issue - State: open - Opened by Evilxya 12 days ago

#274 - 使用langchain接入模型，如何适配with_structured_output

Issue - State: open - Opened by Und3r1ine 18 days ago

#273 - 增加天机案例

Pull Request - State: closed - Opened by sanbuphy 19 days ago

#272 - docs: add Qwen2.5 Lora 微调 SwanLab可视化训练版

Pull Request - State: closed - Opened by Zeyi-Lin 19 days ago

#271 - qwen2.5 lora微调报错

Issue - State: open - Opened by qihouji 19 days ago - 1 comment

#270 - uvicorn running之后，点击网页，之后显示”目前无法处理此请求“，有没有赛博神医救救

Issue - State: open - Opened by crow-1412 23 days ago - 7 comments

#269 - 请问可以出一个mistral-instruct-v0.1版本的sft模型吗

Issue - State: open - Opened by drewqueaz 29 days ago

#268 - AutoDL的llama3环境镜像transformers版本需要更新

Issue - State: closed - Opened by HackerGary about 1 month ago - 1 comment

#267 - The attention mask is not set and cannot be inferred from input because pad token is same as eos token. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.

Issue - State: open - Opened by yangyang1-0 about 1 month ago

#266 - 关于GPU利用率低的问题

Issue - State: open - Opened by Walking-Z about 1 month ago

#265 - Qwen1.5微调报错

Issue - State: open - Opened by Archerer about 1 month ago - 3 comments

#264 - 使用AutoDL报错

Issue - State: open - Opened by cyk2018 about 1 month ago - 1 comment

#263 - [解决方案]Qwen-7B-Chat FastApi 部署调用

Issue - State: closed - Opened by YuGuilliman about 1 month ago

#262 - vllm能够直接调用gguf格式的权重吗

Issue - State: open - Opened by liguoyu666 about 1 month ago - 1 comment

#261 - 要是能出个多模态的就好了

Issue - State: open - Opened by robin-programmer about 2 months ago - 1 comment

#260 - xtuner 微调报错，求指点

Issue - State: open - Opened by zhanbaohang about 2 months ago - 1 comment

#259 - xtuner 微调internlm2.5出错

Issue - State: closed - Opened by zhanbaohang about 2 months ago

#258 - LLaMA3.1 o1-like reasoning chain

Pull Request - State: closed - Opened by Tsumugii24 about 2 months ago

#257 - Qwen2.5 o1-like reasoning

Pull Request - State: closed - Opened by Tsumugii24 about 2 months ago

#256 - update Qwen2.5

Pull Request - State: closed - Opened by LinChentang about 2 months ago

#255 - update Qwen2.5 webDemo 部署调用.md

Pull Request - State: closed - Opened by 0-yy-0 about 2 months ago

#254 - Qwen2.5 vLLM update

Pull Request - State: closed - Opened by Tsumugii24 about 2 months ago

#253 - update Qwen

Pull Request - State: closed - Opened by lta155 about 2 months ago - 1 comment

#252 - self-llm project structure optimization and environment configuration update

Pull Request - State: closed - Opened by Tsumugii24 about 2 months ago

#251 - 运行使用huanhuan.json数据集lora微调LLaMA-3.1-8b-instruct，计算loss时的报错：

Issue - State: closed - Opened by cuishuting 2 months ago - 1 comment

#250 - GLM4私有部署如果要用langchain搭建agent，还需要做哪些改进？

Issue - State: open - Opened by 2500035435 3 months ago - 1 comment

#249 - GLM-4-9B-chat FastApi报AttributeError: 'ChatGLMForConditionalGeneration' object has no attribute 'chat'

Issue - State: open - Opened by hongmin118 3 months ago - 5 comments

#248 - Update 01-Yuan2.0-2B FastApi 部署调用.md

Pull Request - State: closed - Opened by zhangfanTJU 3 months ago

#247 - Yuan2.0的python request调用方法的示例图片不对

Issue - State: open - Opened by Aomferni 3 months ago - 1 comment

#246 - GLM4教程中“02-GLM-4-9b-Chat 接入 LangChain”代码运行报错

Issue - State: open - Opened by meisongzhu 3 months ago

#245 - glm4 AttributeError: 'ChatGLMForConditionalGeneration' object has no attribute 'chat'

Issue - State: open - Opened by eason9494 3 months ago - 2 comments

#244 - Update OpenELM-3B-Instruct "模型部署" and "Lora微调"

Pull Request - State: closed - Opened by moyitech 3 months ago

#243 - glm4教程中小问题

Issue - State: open - Opened by warmth27 3 months ago

#242 - DeepSeek-Coder-V2推理警告

Issue - State: open - Opened by Qlalq 3 months ago

#241 - 使用 8 x 4090 微调 llama3-1-8B 后，询问无输出反馈。

Issue - State: open - Opened by Sakkana 3 months ago - 5 comments

#240 - 通过vllm部署的GLM-4-9B-Chat模型效果很拉垮

Issue - State: closed - Opened by Duanexiao 3 months ago - 2 comments

#239 - 使用fastapi 的GLM4. 出现 AttributeError: 'ChatGLMForConditionalGeneration' object has no attribute 'stream_generate'. Did you mean: 'can_generate'?

Issue - State: closed - Opened by triumph 3 months ago - 1 comment

#238 - deepseek-chat模型LorA微调完没有adapter_config.json

Issue - State: open - Opened by HolyCrazy 3 months ago - 2 comments

#237 - 输出一直在重复

Issue - State: open - Opened by quanquan07 3 months ago - 1 comment

#236 - GLM-4-9B-chat FastApi，curl调用时抛错

Issue - State: closed - Opened by hongmin118 3 months ago - 1 comment

#235 - LLaMA3_1-8B-Instruct WebDemo 部署打开Web页面报错

Issue - State: open - Opened by vistar-terry 3 months ago - 4 comments

#234 - ”02-Gemma-2-9b-it langchain 接入“章节README中最后的使用案例应该使用Gemma模型

Issue - State: closed - Opened by w251510860 3 months ago - 2 comments

#233 - 大模型Prompt Template格式

Issue - State: open - Opened by fang-siqi 3 months ago - 4 comments

#232 - 请问langchain如何实现自定义开源LLM的 stream 响应模式呢？

Issue - State: open - Opened by yuanasdfgh 3 months ago

#231 - llama3.1-8B微调后推理没有结果

Issue - State: open - Opened by Telogen 3 months ago - 3 comments

#230 - Qwen2-7B-Instruct Lora微调的问题

Issue - State: open - Opened by yinzih 3 months ago - 1 comment

#229 - Llama3 Lora报错：RuntimeError: element 0 of tensors does not require grad and does not have a grad_fn

Issue - State: open - Opened by AndyLuo1029 3 months ago - 1 comment

#228 - 'NoneType' object has no attribute 'shape'

Issue - State: open - Opened by jjzhu0579 4 months ago

#227 - Qwen2-7B-Instruct WebDemo部署的问题

Issue - State: open - Opened by yinzih 4 months ago - 6 comments

#226 - Qwen2-7B-Instruct Langchain 问题

Issue - State: open - Opened by yinzih 4 months ago - 1 comment

#225 - Qwen2-7B-Instruct Langchain 接入的问题

Issue - State: open - Opened by yinzih 4 months ago - 4 comments

#224 - update：GLM-4

Pull Request - State: closed - Opened by sssanssss 4 months ago

#223 - update：GLM-4

Pull Request - State: closed - Opened by sssanssss 4 months ago

#222 - llama3 FastAPI部署报错怎么解决哇ww

Issue - State: open - Opened by UandUfindU 4 months ago - 1 comment

#221 - rlhf 相关

Issue - State: open - Opened by Qiny-dl 4 months ago

#215 - 针对作者的 LLaMA3-8B-Instruct 的微调代码运行所出现的问题

Issue - State: open - Opened by linmyl 4 months ago - 2 comments

#209 - Qwen2 lora微调问题

Issue - State: open - Opened by zbbb31 4 months ago - 2 comments

#208 - DeepSeek-Coder-V2-Lite模型Lora微调，显卡利用率不高

Issue - State: open - Opened by tongye98 4 months ago - 2 comments

#204 - qwen2微调后回复的文本出现大段复制的重复内容

Issue - State: open - Opened by Wesirehour 4 months ago - 2 comments

#199 - GLM-4-9B-chat微调部分运行至加载lora权重推理处报错value error

Issue - State: closed - Opened by Eric374254 4 months ago - 10 comments

#198 - GLM-4-9B-chat微调过程中，加载Lora权重推理阶段报错ValueError: too many values to unpack (expected 2)

Issue - State: closed - Opened by Eric374254 4 months ago - 4 comments

#178 - TypeError: transformers.generation.utils.GenerationMixin.generate() argument after ** must be a mapping, not Tensor

Issue - State: open - Opened by LSK-1 5 months ago - 12 comments

#163 - llama3 langchain

Issue - State: open - Opened by Alexa2077 5 months ago - 2 comments

#161 - 数据集如何构造？

Issue - State: open - Opened by chopin1998 5 months ago - 1 comment

#115 - 使用 llama3 的 lora 微调报错：NotImplementedError: Cannot copy out of meta tensor; no data!

Issue - State: open - Opened by chenmonster 6 months ago - 5 comments

#100 - 【非Issues！讨论帖】通过lora微调的qianwen和直接使用system来预设的区别貌似不是很大

Issue - State: closed - Opened by Roger-G 7 months ago - 3 comments

#99 - Building wheel for flash-attn (setup.py) ... - 卡住了。

Issue - State: closed - Opened by zhangyc 7 months ago - 7 comments

#98 - InternLM2 缺少包

Issue - State: closed - Opened by JinProton 7 months ago - 2 comments

#97 - 报错asyncio.run() cannot be called from a running event loop，辛苦大佬们看看

Issue - State: closed - Opened by chaoskklt 7 months ago - 6 comments

#96 - 想问下在这个项目下的lora微调和Chatglm3官方微调的demo的数据格式怎么不一样呀

Issue - State: closed - Opened by McRays 7 months ago - 1 comment

#95 - peft训练完成，参考的是04-Qwen-7B-Chat Lora 微调.ipynb，但是重新载入模型时，提示peft版本问题，用的是model = AutoModelForCausalLM.from_pretrained("../output/Qwen/checkpoint-1300/", trust_remote_code=True).eval()，提示的错误时ValueError: The version of PEFT you are using is not compatible, please use a version that is greater than 0.5.0

Issue - State: closed - Opened by Roger-G 7 months ago - 2 comments

#94 - 请问LLAMA3，里面是按1.2.3.4的顺序来分别执行吗？

Issue - State: closed - Opened by YadiHe 7 months ago - 6 comments

#93 - LLaMA3-8B-Instruct+lora使用A800（80GB显存）微调长度8192

Issue - State: closed - Opened by 12915494174 7 months ago - 2 comments

#92 - Llama3 web_demo&lora

Pull Request - State: closed - Opened by Hongru0306 7 months ago