pytorch/vision

process.wait(0) when traing faster_rcnn

Open

#1,643 opened on Dec 6, 2019

View on GitHub
 (3 comments) (0 reactions) (0 assignees)Python (6,858 forks)batch import
awaiting responsehelp wantedmodule: reference scriptsneeds discussionneeds reproductiontopic: object detection

Repository metrics

Stars
 (15,050 stars)
PR merge metrics
 (Avg merge 12d 8h) (14 merged PRs in 30d)

Description

I use the provided code to train my data using faster_rcnn on 2 GPUs. The program always stops at epoch 27(just stop, no more output), and when I use ctrl+c to interrupt the program, the content is as shown in the pic.

Contributor guide

Research direction
Investigate the multiprocessing dataloader and training loop for potential deadlock. Check if the issue is related to the number of workers or the way data is loaded, especially around epoch 27.
Tech stack
pythonpytorch
Domain
computer visionmachine learningai
Issue type
Bug
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
2
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
Half day
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Needs investigation
Prerequisites
PythonPyTorch
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
45