Category : amazon-sagemaker

I am using SageMaker to train a model with multiple GBs of data. My data is loaded using huggingface’s datasets.load_dataset method. Since data is huge and I want to re-use it, I want to store it in S3 bucket. I tried below: from datasets import load_dataset dataset = load_dataset(‘s3://bucket_name/some_dir/data’, ‘oscar’, ‘unshuffled_deduplicated_en’) but this results in: ..

Read more

I’m currently testing sagemaker reinforcement learning, specifically dueling double deep q network. My model prediction (endpoint deployment) is totally different from evaluation step taken by the agent. The operation that I used for output nodes is ‘main_level/agent/main/online/network_0/dueling_q_values_head_0/output’ type=Add`. Any help would be appreciated. Source: Python..

Read more

I am running a JupyterLab on AWS SageMaker. Kernel: conda_amazonei_mxnet_p27 The number of fields found: saw 9 increments by 1, each run. Error: ParserError: Error tokenizing data. C error: Expected 2 fields in line 50, saw 9 Code: Invocation (Error doesn’t appear when running all cells before this but does when this is ran): train ..

Read more

I need to import function from different python scripts, which will used inside preprocessing.py file. I was not able to find a way to pass the dependent files to SKLearnProcessor Object, due to which I am getting ModuleNotFoundError. Code: from sagemaker.sklearn.processing import SKLearnProcessor from sagemaker.processing import ProcessingInput, ProcessingOutput sklearn_processor = SKLearnProcessor(framework_version=’0.20.0′, role=role, instance_type=’ml.m5.xlarge’, instance_count=1) sklearn_processor.run(code=’preprocessing.py’, ..

Read more

I have a SageMaker endpoint, which works just as expected from Postman. And I have tried to send the same HTTP POST request from python3, but always get some sort of error, while Postman seems to work fine. I would like to know what I am missing with signature in my Python code and what ..

Read more

I use Brave browser for my jupyter notebook (not jupyterlab) works under Amazon Sagemaker, but I also use Chrome for my Internet activities, so last week the update symbol appeared on Chrome, so I updated it, and inmediately my notebooks on Brave stopped rendering well the following tag <hr style="border:3px solid black"> </hr>, which shows ..

Read more

I use Brave browser for my jupyter notebook (not jupyterlab) works under Amazon Sagemaker, but I also use Chrome for my Internet activities, so last week the update symbol appeared on Chrome, so I updated it, and immediately my notebooks on Brave stopped rendering well the following tag <hr style="border:3px solid black"> </hr>, which shows ..

Read more