Distributed training with TensorFlow
In this section, we are going to learn how to take large image files and build custom deep learning models such as object detection or image classification using TensorFlow. By doing so, we’ll learn how to distribute across multiple virtual machines to achieve faster performance for training.
Creating a training job Python file to process
Follow these steps to create a dataset that leverages the user interface:
- Go to https://ml.azure.com and select your workspace.
- Go to Compute and click Start to start the compute instance.
- Wait for the compute instance to start; then, click Jupyter to start coding.
- If you don’t have a compute cluster, please follow the instructions in the previous chapters to create a new one. A compute instance with a CPU is good for development; we will use GPU-based content for model training.
- If you don’t have enough quotas for your GPU, please create a Service Ticket in the...