charosen

the sorting keys?

3

Hi: Thanks for the awesome contributions. I notice there is `sorting keys` params in your config, but it seems like `_create_batches` of BagIterator doesn't use those keys to sort Bags/Instances...

我发现请求一次api并不能获得到准确的100个商户信息，请求api时返回的商户个数会变化，所以使用100为偏移量offset来翻页会不会导致爬取时存在bug，导致run()中统计并打印的获取商户个数有误？我改成了一次请求返回limit=17，返回的商户个数能够准确在17个，用17为偏移量offset来翻页统计的商户个数应该是对的，不知道我理解的对不对。

# Add from_file method to message prompt template **Feature**: This PR adds `from_file` class method to BaseStringMessagePromptTemplate. This is useful to help user to create message prompt templates directly from...

feat(module): add param ids to ElasticVectorSearch.from_texts method

1

# add param ids to ElasticVectorSearch.from_texts method. - Description: add param ids to ElasticVectorSearch.from_texts method. - Issue: NA. It seems `add_texts` already supports passing in document ids, but param `ids`...

[TGI] V100部署qwen2-7b推理服务问题

1

请问qwen2-7b模型可以通过TGI或者vllm框架部署在v100的gpu上吗？ 我使用tgi的容器镜像来直接部署，发现报错不支持flash attention的问题，用了USE_FLASH_ATTENTION=false，tgi部署过程中仍然报错 ``` MASTER_PORT="$(shuf -n 1 -i 60000-65000)" docker run -e CUDA_VISIBLE_DEVICES=3 \ -e USE_FLASH_ATTENTION=false \ --gpus all \ --network host \ --ipc host \ --shm-size="1G" \ --privileged \...