Comments (2)
Yes, it seems the new backend has solved that issue (under Ubuntu). We don't have annoying messages from the zombie process in the shell after termination. I didn't checked how it was programmed, but I believe subprocess is now better managed.
from wandb.
Hi @christopher5106 - I am sorry to hear you are experiencing this issue. Would you mind sharing some additional information to help us troubleshoot this:
- What platform are you running your W&B experiments on?
- What version of the SDK do you currently have installed?
- Does this happen with all your W&B experiments? It would be useful to know what objects are you logging (metrics, Images, Artifacts, Tables)
- Do you have the URL for one of the interrupted Runs as well as the
debug.log
anddebug-internal.log
files you can find in the./wandb/run-<date_time>-<runid>/logs
folder?
Finally, with version 0.17.0+ we have been releasing a new backend for the SDK [wandb-core](https://github.com/wandb/wandb/tree/main/core#readme)
as part of the wandb
package. Among other features, this should improve the robustness of the client so it would be worth testing if you are still experiencing the same issue after enabling it.
The new backend can be enabled setting wandb.require("core")
at the beginning of your script.
from wandb.
Related Issues (20)
- Distributed training with wandb.sweep HOT 6
- does not exist and causes issue with numpy 2.0 release HOT 15
- [Solved][App]: Unable to see the runs in workspace even if the run is taking place successfully HOT 2
- [App]: runs showing from link but not in project page HOT 3
- [Q] Upgrade to gqlparser 2.5.14 in next release? HOT 1
- [App]: All tests fail to run because some dependencies are missing HOT 5
- [Feature]: Halving Random Grid Search for Hyperparameter Tuning HOT 1
- Issue showing runs in the groups from mobile browsers HOT 4
- [Q] "None" was logged when wandb.sweep using pytorch_lightning HOT 3
- [CLI]: wandb write data error HOT 2
- [Q] Download an artifact without the API/CLI HOT 4
- [CLI]: Offline wandb sync failed HOT 7
- [CLI]: init wandb sdk but the /tmp/code directory was not created HOT 4
- [Q] why some steps didn't be logged during training? HOT 2
- [Q] wandb stream ID error HOT 2
- [Q]wandb: ERROR Internal wandb error: file data was not synced wandb: While tearing down the service manager. The following error has occurred: Python int too large to convert to C long HOT 4
- [Q] Does per-sample logging bottleneck batching? HOT 2
- [CLI]: Logging an external artifact folder in Azure Storage Account (HNS) results in a directory stub being logged HOT 2
- [CLI]: wandb.errors.UsageError: Agent user not valid HOT 1
- [Q] Any docs on the settings argument to `wandb.init`? HOT 6
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 wandb.