Comments (3)
I dont think we should teardown. Data shouldn't change helps me in trusting that there was no regression.
from sagemaker-debugger.
Agreed that the data shouldn't change. The data should come from the test though, that's the point of a "unit" test.
from sagemaker-debugger.
Agreed that the data shouldn't change. The data should come from the test though, that's the point of a "unit" test.
If we are creating and deleting data in s3, then there is chance that we would mess with the data.
This is not unit test. If we want unit test, we should mock s3 client and not hit s3.
Hope this makes it clear.
Also, if this is part of s3, then people who dont have s3 won't be able run the unit test. May be we should move all test which hits s3 as part of CI only rather than unit test(use skipif test annotation for test which are trying to access s3). This way the tests will not fail and people doesn't need to have s3 account or access.
If CI s3 test breaks on existing data, then we want some eyes on it, rather than making it easy to modify the s3 data.
from sagemaker-debugger.
Related Issues (20)
- FileNotFoundError when using SageMaker Debugger with PyTorch Distributed Training on SageMaker HOT 2
- Extend Logs to report time and memory usage
- smdebug causes an OperatorNotAllowedInGraphError inside a function decorated with tf.function HOT 1
- Turn off debugger hooks in PyTorch? HOT 4
- Sagemaker debugger hooks for keras unet
- TypeError: get() takes no keyword arguments - breaks training jobs HOT 5
- TypeError: os.environ.get() takes no keyword argument (breaking all PyTorch training jobs)
- Compatibility with gradient accumulation HOT 1
- Can we save tensors that match a regex pattern only for a particular collection
- TF keras.py _wrap_tape_gradient breaks for arrays
- Understanding of how sagemaker-debugger works HOT 1
- Error while running sagemaker-debugger with custom pytorch container and custom model
- Is there any way to disable smdebug to start training the model? HOT 1
- [Feature Request] TensorBoardOutputConfig local output path
- test_pytorch_integration.py::test_pytorch[False-False] is incompatible with PyTorch >=1.7
- Cannot run a custom container using smdistributed/dataparallel unless USE_SMDEBUG is turned off
- Update the Hook callback to be compatible with xgboost>1.3.0 callback style HOT 2
- smdebug crashes with newer numpy versions HOT 1
- The latest smdebug in pypi is 1.0.12 and Released: Aug 26, 2021 HOT 1
- pre-commit changes current master?
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 sagemaker-debugger.