X-PLUG/mPLUG-Owl

Issues

• Question for mplug-owl v1 model code

  #223 opened a month ago by bellos1203
  0

• mPLUG-owl2: RuntimeError: probability tensor contains either `inf`, `nan` or element < 0

  #222 opened a month ago by liuxuannan
  0

• Question about Abstractor's FFN and Attention

  #219 opened 2 months ago by jp1924
  0

• The attention mask and pad token id were not set.

  #221 opened 2 months ago by Evanwu1125
  0

• finetuning: No pytorch_model.bin file after running train_it.sh

  #220 opened 2 months ago by vicaranq
  1

• Owl2中使用的Vit-H-16是从哪个版本初始化的？

  #218 opened 2 months ago by OpenJarvisAI
  2

• ms-swift对于mPLUG-Owl2和mPLUG-Owl2.1微调(finetune)的支持

  #217 opened 2 months ago by Jintao-Huang
  0

• mplug-owl2-llama2-7b initialization error

  #216 opened 2 months ago by lihong2303
  0

• mPLUG-Owl2.1输出全是英文

  #215 opened 2 months ago by Cloopen-ReLiNK
  0

• ModuleNotFoundError: No module named 'transformers_modules.mPLUG-Owl2'

  #214 opened 2 months ago by Cloopen-ReLiNK
  0

• How to do few-shot learning or in-context learning with mPLUG-Owl2?

  #213 opened 2 months ago by Deep1994
  0

• 为什么输出全是英文

  #212 opened 2 months ago by Yuuchuin
  0

• Please can you split the model into 4GB chunks rather than 1 x 16GB. SafeTensors would be a nice addition also.

  #211 opened 2 months ago by doreilly257
  0

• QuickStart Code for mplug_owl2.1 has lots of errors.

  #210 opened 2 months ago by Carol-gutianle
  3

• The code and detailed implementation of Figure 4 and Figure 5 in the paper mPLUG-Owl2

  #197 opened 5 months ago by Zlatan-Ibrahi
  1

• Is there any code for fine-tuning the video model?

  #209 opened 3 months ago by Knight-Totem
  0

• Zero3 train: Invalidate trace cache @ step 391: expected module 25, but got module 5

  #208 opened 4 months ago by Zhiyuan-Fan
  0

• cls_token problem with image.

  #207 opened 4 months ago by evertonipx
  3

• The Quick Start Code cannot be executed in mPLUG-Owl2

  #195 opened 6 months ago by ppsmk388
  8

• Issue with gradio webui

  #206 opened 4 months ago by vamsikrishyvk
  0

• Cannot run inference

  #205 opened 4 months ago by findalexli
  0

• how to realize multi-image correlation in vqa task?

  #200 opened 4 months ago by fansticOne
  4

• other downstream tasks available? Like Visual Reasoning, requires the model to predict whether a sentence describes a pair of images

  #204 opened 4 months ago by fansticOne
  1

• The updated owl2 based on qwen14b?

  #203 opened 4 months ago by luohao123
  1

• 对图像进行坐标检测，生成的bbox是resize成正方形之后的值吗？

  #199 opened 5 months ago by zhaop-l
  5

• file configuration_qwen.py not found

  #201 opened 4 months ago by vd001
  1

• No module named 'mplug_owl2.model.multiway

  #202 opened 4 months ago by YHWmz
  1

• Is there model checkpoint for multi-language(mainly chinese) videos?

  #198 opened 5 months ago by abeldafa
  0

• unable to load model

  #191 opened 6 months ago by segalinc
  3

• Encountered flash_attn_2_cuda error while running finetune_lora.sh

  #192 opened 6 months ago by pradipto111
  1

• RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:2 and cuda:0! (when checking argument for argument weight in method wrapper_CUDA__cudnn_convolution)

  #196 opened 6 months ago by wanghanyang123
  0

• cur_input_embeds = torch.cat([cur_input_embeds_1, cur_image_features[0:0], cur_input_embeds_2], dim=0),其中cur_image_features[0:0]表示这是一个没有维度的向量，图像的特征并没有真正加进去

  #193 opened 6 months ago by hangzeli05
  1

• NameError: name 'repeat_kv' is not defined

  #183 opened 7 months ago by nullnameno
  1

• mPLUG-Owl2，finetune，训练卡住没有输出

  #188 opened 6 months ago by LianghuiGuo
  5

• RuntimeError: indices should be either on cpu or on the same device as the indexed tensor (cpu)

  #190 opened 6 months ago by LianghuiGuo
  2

• implementation of pre-training stage

  #194 opened 6 months ago by annopackage
  0

• Can you release the pre-trained ckpt (without sft) for mplug-owl2?

  #189 opened 6 months ago by waltonfuture
  0

• 请问是否支持中文数据微调呀？May I finetune mPLUG-Owl with chinese image-text pair?

  #180 opened 7 months ago by LianghuiGuo
  2

• Thanks for the great work! I am wondering whether mPLUG-Owl2 have OCR ability?

  #187 opened 6 months ago by LianghuiGuo
  0

• 从modelscope下载的权重，用给的示例代码做Inference，有的权重好像没有加载上，这个有影响么

  #184 opened 6 months ago by LianghuiGuo
  2

• Evaluation Script For The Other Benchmarks

  #185 opened 6 months ago by chancharikmitra
  1

• 加载本地模型报错

  #182 opened 6 months ago by LianghuiGuo
  1

• Does mPLUG-Owl2 support Chinese? Or are there plans to release a new multilingual version?

  #181 opened 7 months ago by BaoyanWang
  1

• Have you tuned the vision model in sft stage?

  #179 opened 7 months ago by xmy0916
  2

• Question about the weight decay

  #177 opened 7 months ago by YifanXu74
  1

• how many tuning parameters？

  #176 opened 7 months ago by adda1221
  1

• duplicate outputs when trying the readme snipped inference code

  #178 opened 7 months ago by segalinc
  0

• Mplug_owl 2 support video training?

  #175 opened 7 months ago by YuzhouPeng
  3

• Dose mPLUG-Owl2 support Chinese?

  #174 opened 7 months ago by Little-Yeah
  1

• OOM when finetune lora zero3 mPLUG-Owl2 on 4-A100-40g

  #173 opened 7 months ago by NamiKaze7
  0