Eusto
0
I could do
mlflow model serve -m <RUN_ID> --p 1234 --no-conda
and
mlflow sagemaker run-local -m <MODEL_PATH> -p 1234
Are they not the same anyway as both can do model serving so what's the hassle deploying it to Sagemaker?
I'm a beginner at this so if anyone can help me out with my understanding that will be great. Thank you so much in advance!
MLFlow Tracking lets users to log parameters and metrics for monitoring and visualizing the model’s performance.
SageMaker Model Monitor enables monitoring the model’s endpoint and automates monitoring with its monitoring schedulers. SageMaker SDK monitors the fraction of dataset sent to the model’s endpoint and alerts on any data drift affecting model’s performance.
Maybe this article would be helpful: https://neptune.ai/blog/amazon-sagemaker-alternatives
Another article by Amazon about Managing your machine learning lifecycle with MLflow and Amazon SageMaker - https://aws.amazon.com/blogs/machine-learning/managing-your-machine-learning-lifecycle-with-mlflow-and-amazon-sagemaker/