-
**运行
~/mmdetection$ python tools/train.py configs/cascade_rcnn_x101_32x4d_fpn_1x.py --validate
时**
2019-07-25 10:13:28,296 - INFO - Distributed training: False
2019-07-25 10:13:30,073 - INFO - l…
-
gtx-16@gtx16-Lenovo:~/Detectron-Cascade-RCNN$ sudo nvidia-docker run --rm -it detectron:c2-cuda9-cudnn7 python2 detectron/tests/test_batch_permutation_op.py
No handlers could be found for logger "caf…
-
请问是什么原因?
-
```import torch
# torch.backends.cudnn.benchmark = True
# torch.backends.cudnn.enabled = True
from mmdet.apis import init_detector, inference_detector, show_result_pyplot
import mmcv
# Load mod…
-
Thanks for your excellent work!
When I use --num-gpus=2, I get the following error, I can't solve the problem, and I hope to get your help.
Traceback (most recent call last):
File "tools/lazyco…
-
I still have this problem.
my project version:
```
commit 0a1a74b16a6712a11570ad06c0378584fe2095f2
Merge: 7589307 96a4037
Author: Jianwei Yang
Date: Thu May 9 12:39:35 2019 -0400
Mer…
-
## 📌 Description
Wrap-up Report를 작성하기 위해 자신이 작성하고자 하는 내용을 댓글로 남겨주시면 좋을 것 같아요.
인당 하나의 댓글을 작성해주시고, Edit 기능을 통해 내용을 수정하는 방향으로 가볼까요?!
## ✅ CheckList
- [ ] 작성 완료
- [ ] 노션으로 Migration
-
model = dict(
type='FCOS',
pretrained=None,
backbone=dict(
type='SwinTransformer',
embed_dim=96,
depths=[2, 2, 6, 2],
num_heads=[3, 6, 12, 24],
…
-
看了一下Objects365 2019 Challenge夺冠模型的配置,没看到seed的设置,另外在paddlex那儿也说目前没设定随机数以便复现结果的方法
-
Can't get rid of this error. It seems like error comes from the model while loading the model