-
根据DeepKE/tree/main/example/llm/OneKE.md的教程的"快速运行"来执行代码,代码是直接复制下来的,报错:显存不足,指写了0号GPU,总显存是32G的,满足教程里的要求,conda环境也是按教程来安装的。麻烦看看是怎么回事。
代码:
显存信息:
报错信息:
-
https://github.com/zjunlp/DeepKE/blob/main/example/llm/InstructKGC/README_CN.md#22训练数据转换
小白来请教大家,请问在这里的KG训练数据的schema中,是否支持多个实体及其属性,如果我想做三元组的抽取任务,应该参考什么样的格式?
hlcle updated
5 months ago
-
**Describe the question**
> A clear and concise description of what the question is.
NER 任务中,换成自己的标注数据集进行训练时会出现KeyError报错:KeyError: 'B-Disease'。
我已经在get_labels函数中添加了训练集中的标注标签,还是未能解决此问题,希望大佬给予帮助。
…
-
Hi there! I ran ‘yen list’ and found that only 4 versions of python are supported here. I wonder if python 3.8 can be supported later? Thanks.
```Available Pythons:
3.12.3
3.11.9
3.10.14
3.9.19
…
-
![image](https://github.com/user-attachments/assets/25f0c866-b4fa-49c7-b3a9-a8295cfecd9c)
-
```
export 'PYTORCH_CUDA_ALLOC_CONF=max_split_size_mb:1024'
CUDA_VISIBLE_DEVICES="0,1,2" torchrun --nproc_per_node=3 --master_port=1287 src/finetune.py \
--do_train --do_eval \
--overwrite…
-
**1、Describe the bug**
> instructKGC/infer_scripts/infer_llama.bash 执行报错
脚本参数:
CUDA_VISIBLE_DEVICES=3 python src/inference.py \
--stage sft \
--model_name_or_path '/data/.cache/modelscope…
-
Hello, I wanted to ask if you could please make available the vanilla prompts used for the CodeKGC paper. I am working on reproducing the results and I would like to know what the vanilla prompt looke…
-
环境:
vllm版本: 0.5.0.post1
执行命令:
`CUDA_VISIBLE_DEVICES=4,5 /data/anaconda3/envs/knowlm/bin/python -m vllm.entrypoints.openai.api_server --model /data/knowlm/zjunlp/OneKE --dtype auto --api-key toke…
-
你好!
目前我已经微调好oneke并生成了如下的checkpoint文件
![image](https://github.com/zjunlp/DeepKE/assets/58768751/18cc6f14-2bad-4449-9bc3-923dd93345e0)
想问下是否有什么办法能通过以下类似的代码进行checkpoint模型的加载呢?