Sacred VS bodywork

🔗 https://github.com/IDSIA/sacred

I'm using Omniboard (https://github.com/vivekratnavel/omniboard) with Sacred (https://github.com/IDSIA/sacred) for tracking experiments. You can specify custom Observers in Sacred so the model metrics and logs will be saved to a local directory or to a remote DB (e.g., MongoDB). I use a MongoDB database hosted on Atlas. Unlike other suggested options, Sacred and Omniboard are free. Atlas free tier comes with 512MB of free storage which is a huge amount if you're uploading only log files to it.

I'm using Omniboard (https://github.com/vivekratnavel/omniboard) with Sacred (https://github.com/IDSIA/sacred) for tracking experiments. You can specify custom Observers in Sacred so the model metrics and logs will be saved to a local directory or to a remote DB (e.g., MongoDB). I use a MongoDB database hosted on Atlas. Unlike other suggested options, Sacred and Omniboard are free. Atlas free tier comes with 512MB of free storage which is a huge amount if you're uploading only log files to it. ex = Experiment() ex.observers.append(FileStorageObserver(EXPERIMENTS_ROOT)) ex.observers.append(MongoObserver(url=MONGODB_URL, db_name='sacred'))

sacred helped me log my experiments. I did setup my environment only once 4 years ago, and since then I have a list of all my training runs with the hyperparameters and results.

For 1, setup an experiment tracking framework. I found Sacred to be helpful https://github.com/IDSIA/sacred.

how to deploy it to Kuberentes using Bodywork.

I usually turn to generative models - e.g. probabilistic programs and Bayesian inference. I’ve written-up my thoughts on how to engineer these into a ‘production system’ deployed to Kubernetes, using PyMC and Bodywork (an open-source ML deployment tool that I contribute to).

I’m one of the core contributors to Bodywork, an open-source tool for deploying machine learning projects developed in Python, to Kubernetes.

I’ve had many successes with this approach. With this in mind, I’ve put together an example of how to make this Agile approach to developing machine learning systems a reality, by demonstrating that it takes under 15 minutes to deploy a Scikit-Learn model, using FastAPI with Bodywork (an open-source MLOps tool that I have built).