AdelaiDet

AdelaiDet copied to clipboard

How to change the ABCNetV2 backbone?

1

I am new to the AdelaiDet toolkit. I am interested in ABCNetV2. in the `configs/BAText/ReCTS/v2_chn_attn_R_50.yaml` the backbone is `build_fcos_resnet_bifpn_backbone` which come from ResNet50, that too large for me, how can...

neverstoplearn

Does BlendMask supports export to onnx?

42

Does BlendMask supports export to onnx?

lucasjinreal

ValueError: cannot convert float NaN to integer

4

``` Traceback (most recent call last): File "tools/train_net.py", line 243, in args=(args,), File "/home/wangqinyu/detectron2/detectron2/engine/launch.py", line 59, in launch daemon=False, File "/home/wangqinyu/anaconda3/envs/Bezier/lib/python3.7/site-packages/torch/multiprocessing/spawn.py", line 200, in spawn return start_processes(fn, args, nprocs, join,...

Wang-Qinyu

How to transform our owm dataset label to your totalText label

TinkingLoeng

请问total-text数据集的标签您是怎么转换的呢

1

原始total-text数据集的标签和您提供的abcnet_custom_dataste_example中的标签不一样呀。请问abcnet_custom_dataste_example中的totaltext的标签方便分享么，万分感谢

ocrhei

【SOLVED】【解决】A frequently error when fcos export to onnx

7

using Torch1.9 , A frequently error can be come out when fcos export to onnx. 模型拓展到onnx总会出现问题。 ```bash ArrayRef: invalid index Index = 15; Length = 15 ``` becase empty list,...

mozheng

RuntimeError: Address already in use

4

When I run a program using multi GPUs, the code can be trained correctly. But if I opened another similar program, which only changed a few params, the RuntimeError was...

Willy0919

The checkpoint state_dict contains keys that are not used by the model: fc1000_b fc1000_w conv1_b Traceback (most recent call last): File "train_sh.py", line 361, in args=(args,), File "/media/lc/soft/zh/xinde/AdelaiDet-master-blendmask/detectron2/detectron2/engine/launch.py", line 62,...

zhenghan408

There are some bug when load the optimizer and scheduler in checkpoint, we change the "train_net.py" according to the codes in "FreeSOLO" that committed by Xinlong Wang.

Lyan-ing

CUDA out of memory, althougt setting batch_size = 2 on 2 gpus

1

Hi everyone. I'm trying to finetune ABCnet v2 from CTW1500 on my custom dataset. However, I got an error `CUDA of out memory` after training some iterations (over 500 iters)....

Pxtri2156