GroundingDINO

GroundingDINO copied to clipboard

why。。。。

7

PS G:\BaiduSyncdisk\AIproject\GroundingDINO> pip install -e . Looking in indexes: https://pypi.org/simplehttps://pypi.doubanio.com/simple, https://pypi.tuna.tsinghua.edu.cn/simple Obtaining file:///G:/BaiduSyncdisk/AIproject/GroundingDINO Installing build dependencies ... done Checking if build backend supports build_editable ... done Getting requirements to build...

tangbob

关于文本不同 导致的 训练精度差异的问题

2

您好！我在利用自己的数据作训练的过程中发现了一个问题：比如当文本是truck . truck mixer . heavy truck；再比如文本是insulator . dirty insulator . damadge insulator等，这种多类别包含了相同词汇的文本时，得到的预测结果有很多是 truck truck mixer、insulator dirty insulator等。然后我改变了类别的定义，比如说truck . concrete mixer . heavy让它们不再包含相同词汇，识别率会提升很多。 起初我以为是模型对某两个类别的特征区分能力比较差 导致它认为某物体会同时是这两个物体。后来我想了下，跟文本特征提取模块也有关系吧？像yolo这种没有文本特征提取分支的模型，相同的训练和验证集识别率就相对高一点 麻烦帮忙分析一下！我应该如何有效地解决这个问题。

wangshuang-jiayou

infrence the python demo was wrong!!!!

17

the infrence code : from groundingdino.util.inference import load_model, load_image, predict, annotate import cv2 model = load_model("groundingdino/config/GroundingDINO_SwinT_OGC.py", "weights/groundingdino_swint_ogc.pth") IMAGE_PATH = "0.jpg" TEXT_PROMPT = "cat" BOX_TRESHOLD = 0.35 TEXT_TRESHOLD = 0.25 image_source,...

sssssshf

TensorRT-LLM multimodal

2

Whether GroundingDINO can support TensorRT-LLM multimodal ？ [TensorRT-LLM multimodal ](https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/multimodal/README.md)

MouseSun846

ayushpixis5941

the version 3.x and 4.x use the different api (Image) `pip install gradio` will install the `latest gradio` > also found that the #295 is for gradio 4.x api.

junxnone

Why is the effect better on the web page?

2

https://deepdataspace.com/playground/grounded_sam。Why can you get better results in this link, but worse in github?   BOX_THRESHOLD = 0.35; TEXT_THRESHOLD = 0.25

pzhren

error: command '/usr/bin/nvcc' failed with exit code 1

6

(Grounding-DINO) hx@FitServer-R4200-V5:/mnt/sda1/Deep_learning/code/GroundingDINO-main$ pip install -e . Obtaining file:///mnt/sda1/Deep_learning/code/GroundingDINO-main Preparing metadata (setup.py) ... done Requirement already satisfied: torch in /home/hx/anaconda3/envs/Grounding-DINO/lib/python3.8/site-packages (from groundingdino==0.1.0) (1.10.0+cu113) Requirement already satisfied: torchvision in /home/hx/anaconda3/envs/Grounding-DINO/lib/python3.8/site-packages (from groundingdino==0.1.0)...

IronmanVsThanos

[Training & Finetuning] Training Code for Grounding-DINO by Community

2

Thank you very much to the community for paying attention to our algorithm. We have listed some very excellent open-source training code here, hoping it can help everyone use Grounding-DINO...

rentainhe

running under python in cpu only mode

1

I don't have a gpu so i can't use cuda, but when trying to annotate a test image i get an error saying that torch has been compiled without cuda....

SevenChords