please make sure to call init_process_group

Published by onesixx on 

RuntimeError: Default process group has not been initialized, please make sure to call init_process_group.

cascade_mask_rcnn_convnext-s_p4_w7_fpn_giou_4conv1f_fp16_ms-crop_3x

single GPU이면 error

멀티 GPU로

CUDA_VISIBLE_DEVICES=2,3 port=29506 sixx/dist_train.sh “sixxconfigs/cascade_mask_rcnn_convnext-s_p4_w7_fpn_giou_4conv1f_fp16_ms-crop_3x_coco_sixx.py” 2

Categories: vision

onesixx

Blog Owner

Subscribe
Notify of
guest

0 Comments
Inline Feedbacks
View all comments

Related Posts

vision

Nuclio 모델배포 : mmdetect

Main.py main.py를 만들기 위해 참고할 inference코드 inference demo main.py의 handler작성을 위해 inference demo확인– init_detector– inference_detector 2. 로컬에서 DL모델을 실행하기 위한 소스코드를 Nuclio 플랫폼에 적용 2-1 모델을 메모리에 로딩 (init_context(context)함수를 사용하여) https://nuclio.io/docs/latest/concepts/best-practices-and-common-pitfalls/#use-init_context-instead-of-global-variable-declarations-or-function-calls 2-2 아래 프로세스를 위해 Read more…

vision

Nuclio 모델배포 : custom model – Mask

cvat/serverless/pytorch/shiyinzhang serverless/tensorflow/matterport/mask_rcnn/nuclio

vision

Nuclio 모델배포 : custom model 

모델배포 : custom model – mmdetection 0. inference demo 확인 https://onesixx.com/inference-demo/ main.py의 handler작성을 위해 inference demo확인 ( init_detector, inference_detector, [show_result_pyplot]) https://onesixx.com/mmdet-inference/ demo/image_demo.py —[copy]—>demo_img.py 1. MakeConfig https://onesixx.com/mmdet-train/ 참고 Dataset 수집/구성/> Model 선정/checkpoint > config 수정 > Training Read more…

0
Would love your thoughts, please comment.x
()
x