When training a model, there is always a chance that something might fail unexpectedly. Proper checkpointing provides a safety net during failures that enables users to restore the state of the model and trainer from a checkpoint file. In Lightning, checkpointing is a core feature in the Trainer and is turned on by default to create a checkpoint after each epoch. But checkpointing provides more than just a safety net in case of failure. Often we care about keeping track of the "best" model weights encountered during the course of training, because in practice not every new epoch leads to an improved generalization error (unstable optimization, overfitting).

The input is a single text file and the model learns to predict the next character in the sequence. Hoping to see many public char-rnn micro-api s with different models spring up, so we can experiment together more easily. This code implements multi-layer Recurrent Neural Network (RNN, LSTM, and GRU) for training/sampling from character-level language models. The input is a single text file and the model learns to predict the next character in the sequence. The context of this code base is described in detail in my blog post.

Today we'll be exploring deploying Machine Learning workloads to a Kubernetes cluster deployed on the Microsoft Azure cloud. The main focus of this is not a deep dive on the ML methods used, but to focus instead on the infrastructure considerations when deploying these types of workloads. So it was a good time to try out these technologies used by many ML shops. For a video on how ML companies are using Kubernetes for their workloads check out Building the Infrastructure that Powers the Future of AI . This post will cover the basics of pushing ML jobs onto a Kubernetes cluster.

At the time of writing this post, most of the big tech companies (such as IBM, Google, Microsoft, and Amazon) have easy-to-use visual recognition APIs. Some smaller companies also provide similar offerings, such as Clarifai. But none of them offer object detection. The following images were both tagged using the same Watson Visual Recognition default classifier. The first one, though, has been run through an object detection model first.

The goal of this tutorial is to show how to use TensorFlow to train and evaluate a simple feed-forward neural network for handwritten digit classification using the (classic) MNIST data set. The intended audience for this tutorial is experienced machine learning users interested in using TensorFlow. These tutorials are not intended for teaching Machine Learning in general. Please ensure you have followed the instructions to install TensorFlow. Simply run the fully_connected_feed.py file directly to start training: MNIST is a classic problem in machine learning.