Loki

ToDo: Rework final tuning to use a Validation Dataset

@jkroll-aws On a more general note, would you rather (1) we created new notebooks for every framework release, deprecating notebooks associated with unsupported versions or (2) modify existing notebooks to...

Hi, Thanks for bringing this to our attention. Can you provide more details: 1. What sequence length are you using ? 2. Are you using a public dataset ? If...

/rerun

/rerun

Is there any update on this ? Is anybody working on a PR ? I see a TF_CONFIG setup is already implemented in https://github.com/aws/sagemaker-tensorflow-training-toolkit/blob/master/src/sagemaker_tensorflow_container/training.py#L37. It would only need some minor...

> Hey @Lokiiiiii so is the aim of your feature to add a distribution argument for multi worker mirrored strategy in Sagemaker? i.e. `estimator = sagemaker.tensorflow.TensorFlow(entry_point='script.py', ..........distribution = {})` What...

> tf.distribute.InputContext is unable to properly identify the number of replicas in sync or number of input pipelines This seems like a discussions for https://github.com/tensorflow/tensorflow/issues?q=is%3Aissue+multiworkermirroredstrategy+

The cluster setup required for `MultiWorkerMirroredStrategy` is now available as part of the SageMaker Training ToolKit from TF >=2.9. Since we are still working on the SageMaker SDK UI for...

Thanks, however I am still seeing an error: ``` Traceback (most recent call last): File "", line 1, in File "/opt/conda/lib/python3.8/site-packages/torchaudio/__init__.py", line 1, in from torchaudio import _extension # noqa:...