Comments (8)
Hi @kbiyani33
To validate, you may pass --validate
into train scripts. You may check your format of annotations if they are correct. Btw, I suggest you start with a baseline model like Faster R-CNN.
from gcnet.
Hi, thank you for your response sir.
I have already trained with similar annotations with Relation Networks for Object Detection.
Even that was trained in COCO. So I am assuming that coco annotations format is correct.
Am I doing anything wrong that I am getting -1 AP and AR?
from gcnet.
I am not quite sure. Since Relation Networks use different codebase with GCNet. You may start with some baseline methods like Faster R-CNN from official MMDetecion to see whether you could get the correct output.
from gcnet.
Oh, understood sir.
Will try it out and get back. Thanks.
from gcnet.
Hello Sir, apparently even faster RCNN is having the same issue. Still getting AP=AR=-1. So what might be the issue?
Is it because of the annotations??
The log file is still giving similar issues.
from gcnet.
Are you using the official MMDetection or GCNet?
You may follow this guide to check your dataset.
from gcnet.
I am using the MMDetection pipeline sir.
from gcnet.
Issue resolved sir. Thanks.
from gcnet.
Related Issues (20)
- what's the T in cosine learning rate when training on ImageNet? HOT 3
- gc HOT 7
- Train on custom data HOT 6
- What‘s the value of transform module mean? HOT 3
- Did anyone use GCNet on Optical Flow features?
- How can I use gc block in resnet18? HOT 3
- Simplified NL HOT 1
- gcnet performs not good on segmentation tasks. HOT 1
- Questions about training
- Attention maps in Different query position HOT 1
- GCNet with pretrianed model on COCO detection?
- Could it be used in 3D data? HOT 1
- possible replacements for layernorm HOT 1
- 找个GC Block这么难? HOT 3
- Mask for training
- Visualization code wanted HOT 1
- change mask-rcnn to faster-rcnn?
- Does GCNet have 1d? HOT 1
- some of the problems HOT 3
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from gcnet.