InternVL

InternVL copied to clipboard

[Bug] Slow loading and inference

2

### Checklist - [X] 1. I have searched related issues but cannot get the expected help. - [X] 2. The bug has not been fixed in the latest version. -...

JunsuKimSustainment

### Checklist - [X] 1. I have searched related issues but cannot get the expected help. - [X] 2. The bug has not been fixed in the latest version. -...

GuiQuQu

[Bug] Failed to merge lora, using tools/merge_lora.py

4

### Checklist - [x] 1. I have searched related issues but cannot get the expected help. - [x] 2. The bug has not been fixed in the latest version. -...

quanlzheng

如何在internvl2上做增量预训练

3

想做增量预训练，大概需要多大的数据量，以及增量预训练的样本分布需要遵循什么原则吗。 可以提供internvl2的stage1的MLP的权重以及训练脚本吗

guanyanchu

用InternVL2-4B 微调后遇到到灾难性遗忘的问题

6

### Motivation 用InternVL2-4B 在一个700多张图片的数据集上微调一个垂直领域的图片理解任务。微调后发现输入通用prompt，比如“请简单描述图片的内容”或“请问图片中是否有车辆”，但输出基本都是微调数据集上的答案。好像通用的能力消失了。不知道这样的问题有没有解决方案？谢谢。 ### Related resources _No response_ ### Additional context _No response_

crmlei

大佬，如何重新训练InternViT‑300M‑448px模型，有开源代码吗

5

### Checklist - [ ] 1. I have searched related issues but cannot get the expected help. - [ ] 2. The bug has not been fixed in the latest...

wuxiaolianggit

[Feature] how to do in-context learning(few-shot learning) with internvl2

4

### Motivation I notice that internval_chat/eval/evaluate_vqa.py has parameters for few-shot learning but have not been implemented correctly. My question is: How can we do few-shot learning with internvl2? 1. should...

Claude-Liu

请问internVL2系列的模型的训练数据集是什么？

2

请问internVL2系列的模型的训练数据集是什么？数据集是跟internVL1.5的技术报告里面提到的数据集一致吗？https://arxiv.org/abs/2404.16821

ChingKwanCheung

[Bug] TypeError: TrainerState.__init__() got an unexpected keyword argument 'stateful_callbacks'

1

### Checklist - [ ] 1. I have searched related issues but cannot get the expected help. - [ ] 2. The bug has not been fixed in the latest...

humphery0sh

Fine-tune InternVL2 using SFTTrainer from hugging face

3

Is the fine-tuning of InternVL supported by hugging face SFTTrainer? I got the following error when using the SFTTrainer: ```python model = AutoModel.from_pretrained( "OpenGVLab/InternVL2-8B", device_map="auto", torch_dtype=torch.float16, trust_remote_code=True ) tokenizer =...

Liyan06