sagemaker-debugger
sagemaker-debugger copied to clipboard
awslabs
Amazon SageMaker Debugger provides functionality to save tensors during training of machine learning jobs and analyze those tensors
Usecase - For BYO analysis, current supported way is fetch tensor one by one, we should expose an API : trial.get_tensors({"tname" :[steps])
Current;y, we access s3 for downloading a tensor, instead we should allow disk to be used as cache, rather than going to s3 each time.
So that people don't get confused by the different ways of doing the same thing
This is an usecase especially for the keras model_to_estimator flow.
### Description of changes: #### Style and formatting: I have run `pre-commit install && pre-commit run --all-files` to ensure that auto-formatting happens with every commit. #### Issue number, if available...
from xgboost version 1.3.0 and later, all the callbacks taken by xgb.train() must be instances of xgb.callback.TrainingCallback https://xgboost.readthedocs.io/en/stable/python/python_api.html#callback-api From the usage : https://github.com/awslabs/sagemaker-debugger/blob/master/smdebug/xgboost/hook.py#L64 If we pass Hook together with other...
When do pip install smdebug, the lastest version it can install is 1.0.12, and that is release 2 years ago. But in github, the latest version is 1.0.31. I wonder...
### Description of changes: Using `time.time()` in the definition of the method would freeze the default to the time at the definition. Defining it in the body allows to query...
When I check out the current master branch and run `pre-commit install && pre-commit run --all-files`, it changes a bunch of files. Should the master be fixed, or pre-commit updated?