Haitao Xiao

Results 27 comments of Haitao Xiao

[Question]: EPRSTMT数据集不存在

> ### 请提出你的问题 > Couldn't find 'eprstmt' on the Hugging Face Hub either: FileNotFoundError: Couldn't find file at https://raw.githubusercontent.com/huggingface/datasets/main/datasets/eprstmt/eprstmt.py No module named 'paddlenlp.datasets.EPRSTMT'

[Question]: EPRSTMT数据集不存在

> FewCLUE 中的 EPRSTMT 加载方式为 > > ``` > from paddlenlp.datasets import load_dataset > data_ds = load_dataset("fewclue", name="eprstmt", splits=["train_0"]) > ``` thankyou

[Question] | ERROR | stderr | requests.exceptions.ConnectionError: HTTPConnectionPool(host='localhost', port=10000): Max retries exceeded with url: /refresh_all_workers (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7f16ba0cb340>: Failed to establish a new connection: [Errno 111] Connection refused'))

你好 有解决方案吗

[Bug]: can't make the first connection

> https://milvus.io/docs/install_standalone-helm.md#Connect-to-Milvus 你是用operator/helm安装的吗？其实你可以用同样的方式连接到milvus，因为它们都安装在k8s中 我得到了相同的问题。。

当我使用推理命令的时候出现网络错误，无法构建推理的接口

有木有老哥碰到过呢

当我使用推理命令的时候出现网络错误，无法构建推理的接口

分两条命令启动推理模型

Data-dependent control flow exploration

> A "prototype" torch.cond exists today: https://pytorch.org/docs/main/generated/torch.cond.html#torch.cond. Try it out and please let us know how it goes by opening issues over at github.com/pytorch/pytorch. cc @ydwu4 Hi,Can I avoid this...

怎么从你们的检查点启动呢

> Hi, I'm not sure what's the meaning of "start from checkpoint". Did you mean continue training from the pre-trained checkpoint? Hi,没错QAQ 我可能不太专业哈,同样的基础模型，你们微调了2个阶段后，我只想微调你们用数据预训练后的基础模型，同样的型号，只要用同样的bash文件吗？QAQ 我是这么理解的。 比如LLAMA和miniGemini-Llama，我想微调的是miniGemini的-llama

怎么从你们的检查点启动呢

> 同问 我估摸着是这样，还是等作者回答把

怎么从你们的检查点启动呢

> ` from minigemini.model.builder import load_pretrained_model > > # loal Mini-Gemini-2B > tokenizer, model, image_processor, context_len = load_pretrained_model(model_path="work_dirs/Mini-Gemini-2B", model_base=None, model_name='YanweiLi/Mini-Gemini-2B', load_8bit=False, load_4bit=False, device_map="auto", device="cuda", use_flash_attn=False) ` https://huggingface.co/YanweiLi/Mini-Gemini-8x7B/blob/main/config.json [ elif "mixtral"...