Repository Issue Activity (beta)

#171 создание художественно-научного текста

#155 vLLM (0.11.0>=) is not working with Dolphin. Got the following issue: ModuleNotFoundError: No module named 'vllm.model_executor.models.bart'

#165 Minimum System Requirements

#164 加载Dolphin-v2 demo_page.py 运行一直卡在这

#159 License Issue

#163 目前Dolphin是否支撑在晟腾(NPU)上部署使用？

#130 [Bug/UX] Dolphin: Mac + Python 3.11/3.12 installation and dtype pitfalls – Only pip works, uv/pdm/poetry fail due to numpy/pytorch/package quirks

#161 更正一下：模型可以直接用vllm推理的

#162 [question] use converted code in my project, is that ok?

#160 Dolphin 2 not installable: No module named 'torch'

#153 幻觉问题异常严重

#140 Footnote edge cases for the model

#158 code-element transcription produces hallucinated code in some cases

#157 the partner of creating dolphin training dataset by myself

#149 Bug: Result folder is empty

#79 Cannot parse a simple PDF

#127 A question about the Table Recognition result reported by paper.

#141 Unable to process JPEG files

#147 Bug Report: Pretrained checkpoint and current DonutModel mismatch

#133 Dependency conflict with numpy versions

#21 Call for Bad Cases

#144 Fine-tuning the Dolphin model encounters a strange issue.

#62 Can you please share a finetuning script

#143 对于page_imgs的demo，运行结果中表格识别出现了不可接受的错位

#139 Inconsistent Prediction Results Between VLLM and Native Transformers