wangzaistone

Results 26 comments of wangzaistone

ok,thanks for your reply. 在 2018-02-27 20:00:09,"embedding" 写道: hi 是的,analogy准确率越好,准确率越高代表做对的analogy问题越多;spearman rank也是越高越好,越高代表算法的结果和人类给出的结果越接近 — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it...

in mac runs it ,which reports error, add as follows import os os.environ['KMP_DUPLICATE_LIB_OK']='True' it works refers:https://github.com/dmlc/xgboost/issues/1715

# 2023-12-02 ## 议题 : 1、DB-GPT v0.4.3新版发布 2、DB-GPT-Hub 相关项目打包成api和后续目标。 3、DB-GPT-Web 前端更新 4、paper ## 结论 1、 DB-GPT v0.4.3版本发布,侧重知识库优化尤其是检索召回和完成文档更新等工作 2、DB-GPT-Hub 初步完成关于api开发、后续方向上形成关于基座与微调baseline构建和数据集丰富的两个目标。 3、DB-GPT-Web 完成官网UI 更新和相关召回展示优化 4、主项目的paper预计12月中旬前放出 # Metting Notes ## Agenda: 1、Release of...

@HwzGit 你显卡是怎样的型号的?初步看和显卡对数据精读的支持有关

> @ucas010 ”把config加到路径即可。sys“ 这个具体怎么加的? 能详细说下步骤吗?

> 按照[install文档](https://github.com/imClumsyPanda/langchain-ChatGLM/blob/master/docs/INSTALL.md)安装的, 最后一个执行错误, python loader/image_loader.py Traceback (most recent call last): File "/langchain-ChatGLM/loader/image_loader.py", line 8, in from configs.model_config import NLTK_DATA_PATH ModuleNotFoundError: No module named 'configs' 请教下,我在centos环境下也会有这个问题,这个该如何解决?

> 可以参考[这个](https://github.com/baichuan-inc/Baichuan2/issues/204)解决一下。 After `pip install transformers==4.33.1 tokenizers==0.13.3` ,it will report vllm version eror ,like this : ``` ERROR: pip's dependency resolver does not currently take into account all the packages...

> @Aries-ckt llm model有尝试了vicuna 13b_v1.1 和 chatglm-6b,两种模型的效果在链接数据库都是经常出现错误。另一个类似报错为: {'model': 'vicuna-13b', 'prompt': 'You are an AI designed to answer human questions, please follow the prompts and conventions of the system\'s input for...

have the same issue too ,evaluate_classical_test2.py:64: RuntimeWarning: coroutine 'exec_on_db' was never awaited flg, gold_result = exec_on_db(testcase_path, gold_query, timeout=GOLD_TIMEOUT) RuntimeWarning: Enable tracemalloc to get the object allocation traceback 0%| | 0/196...

模型.chat前增加一行,model = model.half().cuda() 即可。 环境:v100