ZNF

JSONDecodeError when calling Chinese jar in Python

4

I have download stanfordcorenlp 3.9.1.1 and the newest zip file stanford-corenlp-latest, still download Chinese model jar in this folder, when I run the code, ``` from stanfordcorenlp import StanfordCoreNLP sentence...

LayoutLmv3 | ConnectionError: Couldn't reach https://raw.githubusercontent.com/huggingface/datasets/2.4.0/metrics/seqeval/seqeval.py

1

I tried to fine-tune layoutLMv3 Chinese model on XFUND, and I input those in terminal ``` python examples/run_xfund.py \ --data_dir data --language zh \ --do_train --do_eval \ --model_name_or_path microsoft/layoutlmv3-base-chinese \...

Is there a simple method can directly use layoutlmv3 and test some image samples?

1

I only want to input some images without labels to this model and return a result including bbox, labels, etc. However, the code is too complicate for a novice... I...

Why parameter [timesteps] is fixed?

4

I found that in line 2308, the creating of NoiseScheduler layer depends on `timesteps` in dalle2_pytorch.py. Did it means I **cannot** change (increase or decrese) the value of `timesteps` if...

I am confused about the format,'.gntx' in [CASIA-AHCDB](http://www.nlpr.ia.ac.cn/pal/CASIA-AHCDB.html). How to convert it to a png file, is it similar with 'gnt' format?

I am confused about the format,'.gntx' in [CASIA-AHCDB](http://www.nlpr.ia.ac.cn/pal/CASIA-AHCDB.html). How to convert it to a png file, is it similar with 'gnt' format?

Hi, I am confused about `A: B: C: D:` meaning in this part of dict, ``` { "instruction": "[HM]: 如何种植一个蔬菜园？\nA:准备土壤并进行测试。\nB:选择合适的作物和种子。\nC:根据植物需要准备水分。\nD:定期施肥及杀虫剂等保护措施。\nE:每隔一段时间耕翻土壤。\nF:根据季节进行必要的农事管理工作。\nG:收获成熟的作物。\nH:将植物废弃物回收利用。", "input": "", "output": "如果你想要种植一个蔬菜园，建议按照以下步骤进行：A. 准备土壤并进行测试；B. 选择合适的作物和种子；C. 根据植物需要准备水分；D. 定期施肥及杀虫剂等保护措施；E. 每隔一段时间耕翻土壤；F. 根据季节进行必要的农事管理工作；G....

您好，关于release2.0版本提及的`1.通过数据集清洗再训练，缓解了先前版本经过Agent/工具学习训练后对原有知识的灾难性遗忘，` 能否问一下在SFT中具体采用的方法吗？包括通用数据具体采用了何种数据集，和专业数据的具体比例，以及训练前数据预处理过程？是否需要shuffle？或者别的处理？

Update rouge.py

1

A typo in comment :)

Qwen3-Next-80B-A3B-Thinking hangs during multi-node SFT training with Ray (NCCL timeout on InfiniBand)

1

### Reminder - [x] I have read the above rules and searched the existing issues. ### System Info When training Qwen3-Next-80B-A3B-Thinking across multiple nodes using Ray, the following error occurs:...