Comments (3)
Hi!
Thanks for opening the issue. The only explanation I could have is that your environment variable AIDE_CONFIG_PATH
points to a different settings file. Are you using Docker? Then you have to modify the docker/settings.ini
file, not the config/settings.ini
file.
--
To make the system a bit more hiccup-proof, the latest commit now contains a fallback solution to the numWorkers
argument if it cannot be found (defaults to 6
).
from aerial_wildlife_detection.
Hi!
Thanks for opening the issue. The only explanation I could have is that your environment variable
AIDE_CONFIG_PATH
points to a different settings file. Are you using Docker? Then you have to modify thedocker/settings.ini
file, not theconfig/settings.ini
file.--
To make the system a bit more hiccup-proof, the latest commit now contains a fallback solution to the
numWorkers
argument if it cannot be found (defaults to6
).
settings.txt
terminal_logs.txt
Thank you for your reply. The problem is solved! Yes when I use docker I can start the server and log in. I tried to run AIDE on one machine however, when I tried to label the images I uploaded and use built-in model to learn them, I have some questions:
- All of my labels cannot be saved. After I finished labeling some images and quit, I cannot find the labels I did anymore. Is it because of some configuration problems in setting.ini file?
- How can I apply built-in models automatically? I make some configurations under AI model->setting in GUI but nothing happened. The model does not run when I do the labeling. I tried to find answers in build-in model.md but I am sorry I still feel confused what I should do to make it run. What is the configuration JSON file for and how can I use it?
Here are some files I think might help you to understand my questions. Thank you again!
from aerial_wildlife_detection.
Hi!
- Your annotations should be saved automatically whenever you click "Next" (or "Previous") or leave the browser page. AIDE tries to maximize the diversity of images annotators get to see; you therefore won't see already checked images again as long as there are unlabeled ones. However, if you are a project administrator you can still take a look at your labels, either by clicking "review annotations" in the bottom right of the main labeling interface screen (below the label class list), or by clicking the tab "Data management" in the configuration page (in your screenshot above)—in there, you can list images and sort them by number of annotations, for example.
- A little while ago I started to re-write the automated training procedure, as it was a bit unstable. For the time being, automated model training is disabled for this reason; models can currently only be used (training, prediction, etc.) manually by launching tasks through the Workflow designer. Sorry about that. However, the next major release is coming soon, and in there automated model training will be enabled and configurable again through the "AI model" > "Settings" tab. You can take a sneak peek at this on the respective branch.
from aerial_wildlife_detection.
Related Issues (20)
- Decoupling Docker services HOT 2
- Problem uploading images HOT 2
- column "annotationtype" does not exist HOT 4
- AIDE.sh: line 10: 1647 Illegal instruction (core dumped) HOT 7
- sysctl errors HOT 1
- predictions showing after user annotation HOT 3
- Error - invalid characters for a local volume name HOT 2
- "libc10_cuda.so: cannot open shared object file: No such file or directory" HOT 2
- Unexpected Error - Results not being saved (Time Out Error) HOT 1
- Manually stop training without using UI HOT 3
- Way to export null annotations HOT 1
- Celery worker raising IsADirectoryError upon running a task HOT 3
- admin password - invalid salt HOT 6
- Installing database failling HOT 2
- Detectron error HOT 2
- Maximum number of images to predict on at a time not working
- how to best export data (model, images, annotations) to new machine
- Yolov file format not recognized?
- MacOS: ERROR: Could not build wheels for imagecodecs
- Ubuntu 20.04 install fail - failed to build detectron2
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 aerial_wildlife_detection.