InternImage
InternImage copied to clipboard
OpenGVLab
[CVPR 2023 Highlight] InternImage: Exploring Large-Scale Vision Foundation Models with Deformable Convolutions
I tried to export a classification model to ONNX model according to [README](https://github.com/OpenGVLab/InternImage/blob/master/classification/README.md) in the classification directory. But it was failed with following error messages. ```bash (internimage) ubuntu@ip-10-0-0-189:~/InternImage/classification$ python export.py...
 FileNotFoundError: [Errno 2] No such file or directory: 'data/cityscapes/gtFine/train/strasbourg/strasbourg_000001_031683_gtFine_labelTrainIds.png'
Hi, I currently migrate the InternImage into mmdetection==2.28.1, and I encounter the error that say the CustomLayerDecayOptimizerConstructor is not implemented. But I didn't find the impelment of CustomLayerDecayOptimizerConstructor in the...
add frozen_stages kwarg to InternImage backbone. It will make it consistent with other MMDetection backbones and allows for easier fine-tuning for downstream tasks.
optimizer = dict(type='SGD', lr=0.05, momentum=0.9, weight_decay=0.0001) optimizer_config = dict(grad_clip=None) # learning policy lr_config = dict( policy='step', warmup='linear', warmup_iters=500, warmup_ratio=0.001, step=[8, 11]) runner = dict(type='EpochBasedRunner', max_epochs=20)
I got this error in yolov8 ModuleNotFoundError: No module named 'DCNv3', how to fix it
DCNv4 version 1.0.0.post2 by pip. `pip install DCNv4==1.0.0.post2` config file as follows: img_backbone=dict( type='InternImage', core_op='DCNv3', ## large init_cfg=dict(type='Pretrained', checkpoint='ckpts/cascade_flash_internimage_l_fpn_3x_coco.pth'), # init_cfg=dict(type='Pretrained', checkpoint='ckpts/mask2former_flash_internimage_l_640_160k_ade20k_ss.pth'), channels=160, depths=[5, 5, 22, 5], groups=[10, 20, 40,...
使用mmdeploy转换tensorrt模型后，使用Inference by Model Converter 可以正常推理，并生成结果图像；但使用Python API 进行推理会报如下错误。请问该怎么解决呢？（mmdeploy master分支） `loading libmmdeploy_trt_net.so ... loading libmmdeploy_ort_net.so ... [2023-06-30 07:19:16.008] [mmdeploy] [info] [model.cpp:35] [DirectoryModel] Load model: "/home/project/data/Model/InternImage/20230608_internimage-L_1x" [2023-06-30 07:19:36.276] [mmdeploy] [error] [trt_net.cpp:28] TRTNet: 1:...
Hello! I'm attempting to export detection model from pytorch to tensorrt, but there is an error below: ImportError: cannot import name 'register_all_modules' from 'mmdet.utils.setup_env' (/xxx/.conda/envs/internimage/lib/python3.7/site-packages/mmdet/utils/setup_env.py) Is this mmdet version problem?
I convert the semantic segmentation model to tensorrt engine, then I need to deploy it by C++ environment ,any one experienced could tell me how to infer the end2end.engine of...
