GitHub / eosphoros-ai/db-gpt-hub issues and pull requests
#308 - 能否拨冗查看下这个问题TypeError: output tensor must have the same type as input tensor
Issue -
State: open - Opened by ZephryLiang 5 months ago
#307 - use_auth_token报错
Issue -
State: open - Opened by seven17777777 5 months ago
#306 - 运行模型微调 报错
Issue -
State: closed - Opened by seven17777777 5 months ago
#305 - Unable to pre-compile async_io
Issue -
State: open - Opened by annen-stack 6 months ago
#304 - 依赖问题
Issue -
State: open - Opened by LingJingMaster 6 months ago
- 4 comments
#303 - 資料集 & 模型的選擇
Issue -
State: open - Opened by JonathanHuangC 6 months ago
#302 - Support execution result evaluation for text2gql
Pull Request -
State: closed - Opened by SonglinLyu 7 months ago
#301 - How do I get func_timeout
Issue -
State: closed - Opened by JonathanHuangC 7 months ago
#300 - 找不到lora权重文件
Issue -
State: open - Opened by 759212482 8 months ago
- 1 comment
#299 - bf16
Issue -
State: closed - Opened by hangongzi 8 months ago
#298 - 使用多卡训练报错 AssertionError: no_sync context manager is incompatible with gradient partitioning logic of ZeRO stage 2
Issue -
State: open - Opened by lhhchanger 8 months ago
- 2 comments
#297 - AttributeError: module 'transformers.utils.logging' has no attribute 'basicConfig'
Issue -
State: open - Opened by moyanxinxu 9 months ago
#296 - cuda11.8 版本怎么修改呀?
Issue -
State: open - Opened by zxjhellow2 9 months ago
#295 - ModuleNotFoundError: No module named 'transformers.deepspeed',已经安装了transformers和deepspeed
Issue -
State: open - Opened by xiangzhangpang 9 months ago
- 1 comment
#294 - 有大佬知道,怎么使用这个项目微调成功后并且整合模型后的模型进行本地部署?
Issue -
State: open - Opened by ychuest 10 months ago
- 1 comment
#293 - 使用Qwen2___5-Coder-7B-Instruct进行微调,参数如下,出现如下报错,求助求助!!!!
Issue -
State: open - Opened by ychuest 10 months ago
- 10 comments
#292 - 基础模型的微调
Issue -
State: open - Opened by ychuest 10 months ago
- 3 comments
#291 - 文件缺少,脚本全是bug!
Issue -
State: open - Opened by cristianohello 11 months ago
#290 - 意图识别报错!
Issue -
State: open - Opened by cristianohello 11 months ago
#289 - The main branch does not have pyproject.toml
Issue -
State: closed - Opened by dusx1981 11 months ago
#288 - gql语料生成
Issue -
State: closed - Opened by ccp123456789 11 months ago
- 1 comment
#287 - Add Text2GQL fine tuning framework and provide TuGraph examples
Pull Request -
State: closed - Opened by SonglinLyu 11 months ago
#286 - Llama 3 performs horribly for certain prompt representations
Issue -
State: open - Opened by oleherbst 11 months ago
#285 - Text2gql
Pull Request -
State: closed - Opened by SonglinLyu 11 months ago
- 1 comment
#284 - feat: Support fine-tuning of NLU tasks
Pull Request -
State: closed - Opened by fangyinc 11 months ago
#283 - Text2gql
Pull Request -
State: closed - Opened by SonglinLyu 11 months ago
#282 - merge test
Pull Request -
State: closed - Opened by SonglinLyu 12 months ago
#281 - AssertionError: Provided path (dbgpt_hub/output/adapter/llama3-sqlcoder-lora) does not contain a LoRA weight.
Issue -
State: open - Opened by zhangsone 12 months ago
- 3 comments
#280 - Gtk code - DO NOT MERGE - WIP
Pull Request -
State: closed - Opened by gaurav274 12 months ago
#279 - 关于支持的模型
Issue -
State: open - Opened by YangwdX 12 months ago
#278 - 如何关闭 FlashAttention ,不使用FlashAttention 加速呢?RuntimeError: FlashAttention only supports Ampere GPUs or newer.
Issue -
State: open - Opened by gongjinghao 12 months ago
#277 - 可以考虑支持下Llama3.1-8B作为基准模型吗
Issue -
State: open - Opened by Oops322 12 months ago
#276 - 怎么生成safetensors 格式的模型?
Issue -
State: open - Opened by DanielSunHub about 1 year ago
#275 - predict_sft.sh中参数如何填写,多次尝试运行脚本文件都报错
Issue -
State: closed - Opened by Oops322 about 1 year ago
- 1 comment
#274 - add paper reference to head
Pull Request -
State: closed - Opened by moutozf about 1 year ago
#273 - add paper reference
Pull Request -
State: closed - Opened by moutozf about 1 year ago
#272 - feat: add qwen-1.5b llm for ner task
Pull Request -
State: closed - Opened by zhanghy-sketchzh about 1 year ago
#271 - 什么时候可以支持glm-4-9b-chat
Issue -
State: open - Opened by KOBEBRYANTand about 1 year ago
- 1 comment
#270 - 使用微调之后的Baichuan2-13B模型到DB-GPT框架后出现报错
Issue -
State: open - Opened by KOBEBRYANTand about 1 year ago
#269 - 预测阶段:poetry run sh ./dbgpt_hub/scripts/predict_sft.sh,Killed
Issue -
State: open - Opened by GuokaiLiu about 1 year ago
- 2 comments
#268 - config.py配置更新
Issue -
State: open - Opened by liujiachi1997 about 1 year ago
- 1 comment
#267 - fix lower case create statement
Pull Request -
State: closed - Opened by initzhang about 1 year ago
#266 - 请问什么时候更新支持最新的模型Qwen1.5 llama3 ,还有数据集支持格式。例如sql-context
Issue -
State: open - Opened by renlongY about 1 year ago
#265 - 评估中每一列分别代表什么?
Issue -
State: open - Opened by tongcu over 1 year ago
- 1 comment
#264 - pip install dbgpt-hub Report an error
Issue -
State: closed - Opened by zhangkuo-zk over 1 year ago
- 3 comments
#263 - 大家有llama3 的微调参数之类的么
Issue -
State: closed - Opened by dusens over 1 year ago
#262 - 切换数据库和知识库,所有的对话都会同步修改
Issue -
State: open - Opened by zhangkuo-zk over 1 year ago
#261 - docs: update README.md
Pull Request -
State: closed - Opened by eltociear over 1 year ago
#260 - 最优模型训练参数
Issue -
State: open - Opened by jiechuangu over 1 year ago
#259 - now 1k star
Issue -
State: closed - Opened by andeyeluguo over 1 year ago
- 1 comment
#258 - Question about template selection
Issue -
State: open - Opened by am4ever over 1 year ago
#257 - 请问目前是仅支持lora和qlora微调吗,全参数微调后续会开放吗?
Issue -
State: open - Opened by CUCldyyyyy over 1 year ago
- 3 comments
#256 - 我在加载数据集时,出现断言错误,请问如何解决?目前使用glm3模型,模型已经导入,目前排查出错在语句dataset = preprocess_dataset(dataset, tokenizer, data_args, training_args, ",sft")后续无法排查。
Issue -
State: open - Opened by hongWin over 1 year ago
- 5 comments
#255 - 微调之后的模型很小
Issue -
State: closed - Opened by yilia1828 over 1 year ago
#254 - fix: rm wechat group code
Pull Request -
State: closed - Opened by csunny over 1 year ago
#253 - lora训练是不支持modules_to_save这个参数吗
Issue -
State: open - Opened by JasonLLLLLLLLLLL over 1 year ago
#252 - A40显卡微调Qwen1.5-7B-Chat报错:RuntimeError: mat1 and mat2 must have the same dtype, but got Float and BFloat16
Issue -
State: open - Opened by lordk911 over 1 year ago
- 2 comments
#251 - Error reported when using Spark Model v3.5 to connect to DB-GPT
Issue -
State: closed - Opened by dj-jack001 over 1 year ago
- 1 comment
#250 - 请问如何使用中文数据集进行训练?
Issue -
State: open - Opened by hanyonggihub over 1 year ago
#249 - 请问支持在Mac M2机器上进行训练吗
Issue -
State: open - Opened by mobguang over 1 year ago
- 1 comment
#248 - torch.cuda.OutOfMemoryError
Issue -
State: closed - Opened by yilia1828 over 1 year ago
- 1 comment
Labels: good first issue
#247 - chore: update wechat group code
Pull Request -
State: closed - Opened by csunny over 1 year ago
#246 - If the fine-tuned model could be used to DB-GPT?
Issue -
State: closed - Opened by nibnahzuh over 1 year ago
- 2 comments
#245 - Can we support the sqlcoder-7b-2
Issue -
State: closed - Opened by yourchanges over 1 year ago
- 1 comment
#244 - 请问为什么合并模型的时候does not contain a a LORA weight
Issue -
State: closed - Opened by GuohuanFeng0 over 1 year ago
- 5 comments
Labels: good first issue
#243 - 请问怎么自定义数据集
Issue -
State: open - Opened by Guohuan-Feng over 1 year ago
- 2 comments
#242 - 麻烦更新一下微信群的二维码,谢谢~
Issue -
State: closed - Opened by ChanghaoLau over 1 year ago
#241 - 在windows server上可以安装么?
Issue -
State: closed - Opened by uebhh over 1 year ago
- 1 comment
#240 - Bird eval all zf
Pull Request -
State: closed - Opened by moutozf over 1 year ago
- 3 comments
#239 - codellama70B probably needs how much memory to train the spwider dataset?
Issue -
State: open - Opened by likenamehaojie over 1 year ago
- 1 comment
#238 - RuntimeError: expected mat1 and mat2 to have the same dtype, but got: float != c10::Half
Issue -
State: open - Opened by HwzGit over 1 year ago
- 3 comments
#237 - chore: update wechat group QR code
Pull Request -
State: closed - Opened by csunny over 1 year ago
#236 - predict_sft.sh 推理速度好慢
Issue -
State: open - Opened by wangyongshuai88 over 1 year ago
#235 - 网页刷新后 每个会话的模型选择恢复到默认模型 无法模型选择记忆化
Issue -
State: closed - Opened by wxy1105952676 over 1 year ago
- 2 comments
#234 - Prompt for CodeLlama model
Issue -
State: closed - Opened by tail-recursion over 1 year ago
- 1 comment
#233 - chore: update wechat QRcode
Pull Request -
State: closed - Opened by csunny over 1 year ago
- 1 comment
#232 - 模型训练完进行合并权重时,显示does not contain a LoRA weight
Issue -
State: closed - Opened by zhuyubaiyu over 1 year ago
- 3 comments
#231 - CodeLlama SFT
Issue -
State: open - Opened by weirukai over 1 year ago
#230 - 可以公开一下hugging face上的lora模块的微调参数吗
Issue -
State: closed - Opened by Mucalinda2436 over 1 year ago
- 1 comment
#229 - Bird数据集评估的时候要传入的predict_dev.json文件的格式是什么样的?
Issue -
State: open - Opened by Mucalinda2436 over 1 year ago
#228 - 三张3090卡可以不开量化用lora在BIRD数据集上微调吗
Issue -
State: closed - Opened by Mucalinda2436 over 1 year ago
- 3 comments
#227 - HUB #226: baseline bugfix
Pull Request -
State: closed - Opened by oushu1zhangxiangxuan1 over 1 year ago
#226 - Baseline execution accuracy metric error
Issue -
State: closed - Opened by oushu1zhangxiangxuan1 over 1 year ago
- 1 comment
#225 - sparc multi-turn data set processing script
Pull Request -
State: closed - Opened by zhanghy-sketchzh over 1 year ago
#224 - 请问推理的时候为什么不使用批量的方式呢?
Issue -
State: closed - Opened by Mucalinda2436 over 1 year ago
- 2 comments
#223 - 请问怎么用bird数据集微调codellama呢?
Issue -
State: closed - Opened by Mucalinda2436 over 1 year ago
- 2 comments
#221 - Text2SQL评估指标EX和TS
Issue -
State: closed - Opened by 123qwe1234512 over 1 year ago
- 1 comment
#219 - 有微调好的大模型吗?
Issue -
State: closed - Opened by FB-wh over 1 year ago
- 3 comments
#218 - update wechat qr code
Pull Request -
State: closed - Opened by csunny over 1 year ago
#217 - 多轮对话的训练数据格式
Issue -
State: open - Opened by wangweihua11 over 1 year ago
- 1 comment
#214 - Provided path (dbgpt_hub/output/adapter/Qwen-14B-Chat-sql-lora) does not contain a LoRA weight.
Issue -
State: closed - Opened by Z-Diviner over 1 year ago
- 8 comments
#212 - update wechat qrcode
Pull Request -
State: closed - Opened by csunny over 1 year ago
#210 - 执行poetry install下载报错
Issue -
State: closed - Opened by Yokixixi over 1 year ago
- 5 comments
#208 - Update sql_data_process.py
Pull Request -
State: closed - Opened by zhanghy-sketchzh over 1 year ago
- 2 comments
#206 - RuntimeError: expected scalar type Float but found BFloat16
Issue -
State: closed - Opened by yoguoo over 1 year ago
- 8 comments
#204 - Checkpoint_dir初始值设置None存在问题
Issue -
State: closed - Opened by 123qwe1234512 over 1 year ago
- 6 comments
#202 - 路径报错:[Errno 2] No such file or directory: 'dbgpt_hub/data/dataset_info.json'
Issue -
State: closed - Opened by polar-bear1234 over 1 year ago
- 2 comments
#196 - 中文模型训练是否支持?
Issue -
State: closed - Opened by WeiYue0517 over 1 year ago
- 4 comments
#193 - in _load_from_state_dict self.bias.data = bias_data.to(self.bias.data.device) AttributeError: 'NoneType' object has no attribute 'to'
Issue -
State: closed - Opened by wangzaistone over 1 year ago
- 4 comments
#191 - Update ci.yml with MacOS Env
Pull Request -
State: closed - Opened by qidanrui over 1 year ago