Keras VS MLflow

Keras

As a beginner, I was looking for something simple and flexible for developing deep learning models and that is when I found Keras. Many AI/ML professionals appreciate Keras for its simplicity and efficiency in prototyping and developing deep learning models, making it a preferred choice, especially for beginners and for projects requiring rapid development.

After setting the variables for the environment, the next step is to install dependencies. To use Gemma, KerasNLP is the dependency used. KerasNLP is a collection of natural language processing (NLP) models implemented in Keras and runnable on JAX, PyTorch, and TensorFlow.

All breaking changes are listed here: https://github.com/keras-team/keras/issues/18467

You can use this migration guide to identify and fix each of these issues (and further, making your code run on JAX or PyTorch): https://keras.io/guides/migrating_to_keras_3/

I'm guessing the "real" keras code is coming from the keras repository. Is that a correct assumption? How does that version of Keras get there? If I wanted to write my own activation layer next to ELU, where exactly would I do that?

Other popular machine learning tools include PyTorch, Keras, and Scikit-learn. PyTorch is an open-source machine learning library developed by Facebook that is known for its ease of use and flexibility. Keras is a high-level neural networks API that is written in Python and is known for its simplicity. Scikit-learn is a machine learning library for Python that is used for data analysis and data mining tasks.

Click to Learn more...

I'm not aware of anything off-the-shelf, but if you have sufficient programming experience, one way to do this would be to build a large dataset of reference images and pictures and use something like keras to train a convolutional neural network on them.

How can this be? The current state of practice in AI/ML work requires adaptivity, which is uncommon in classical computational fields. There are myriad tools that capture the work across the many instances of the AI/ML lifecycle. The idea that any one tool could sufficiently capture the dynamic work is unrealistic. Take, for example, an experiment tracking tool like W&B or MLFlow; some form of experiment tracking is necessary in typical model training lifecycles. Such a tool requires some notion of a dataset. However, a tool focusing on experiment tracking is orthogonal to the needs of analyzing model performance at the data sample level, which is critical to understanding the failure modes of models. The way one does this depends on the type of data and the AI/ML task at hand. In other words, MLOps is inherently an intricate mosaic, as the capabilities and best practices of AI/ML work evolve.

MLflow is an open-source platform for managing the end-to-end machine learning lifecycle. It includes features for experiment tracking, model versioning, and deployment, enabling developers to track and compare experiments, package models into reproducible runs, and manage model deployment across multiple environments.

Platforms such as MLflow monitor the development stages of machine learning models. In parallel, Data Version Control (DVC) brings version control system-like functions to the realm of data sets and models.

MLFlow - http://mlflow.org

MLflow:

In search for a tool that needs comparable configuration I looked into mlflow and found this. https://github.com/mlflow/mlflow/blob/master/mlflow/environment_variables.py There they define a class _EnvironmentVariable and create many objects out of it, for any variable they need. The get method of this class is in principle a decorated os.getenv. Maybe that is something I can take as orientation.

I have been using DVC and MLflow since then DVC had only data tracking and MLflow only model tracking. I can say both are awesome now and maybe the only factor I would like to mention is that IMO, MLflow is a bit harder to learn while DVC is just a git practically.