In this tutorial we'll walk through a simple convolutional neural network to classify the images in CIFAR10 using PyTorch.
We’ll also set up Weights & Biases to log models metrics, inspect performance and share findings about the best architecture for the network. In this example we're using Google Colab as a convenient hosted environment, but you can run your own training scripts from anywhere and visualize metrics with W&B's experiment tracking tool.
Results will be logged to a shared W&B project page.
Let’s review the key wandb commands we used in the Colab notebook above.
We encourage you to tweak these and run this cell again to see if you can achieve improved model performance!
One you’ve trained your model you can visualize the predictions made by your model, its training and loss, gradients, best hyper-parameters and review associated code.
To view runs created by people in this public project:
Click through to a single run to see more details about that run. For example, on this run page you can see the gradients I logged when I ran this script.
You can visualize predictions made at every step by clicking on the Media tab. Here we can see an example of true labels and predictions made by our model on the CIFAR dataset.
The overview tab picks up a link to the code. In this case, it's a link to the Google Colab. If you're running a script from a git repo, we'll pick up the SHA of the latest git commit and give you a link to that version of the code in your own GitHub repo.
Use a parallel coordinates chart to see the relationship between hyperparameters and output metrics. Here, I'm looking at how the learning rate and other metrics I saved in "config" affect my loss and accuracy.
We encourage you to fork this colab notebook, tweak some hyperparameters and see if you can beat the leading model! Your goal is to maximize Test Accuracy. Good luck!
We're always free for academics and open source projects. Email firstname.lastname@example.org with any questions or feature suggestions. Here are some more resources:
Enter your email to get updates about new features and blog posts.
We're building lightweight, flexible experiment tracking tools for deep learning. Add a couple of lines to your python script, and we'll keep track of your hyperparameters and output metrics, making it easy to compare runs and see the whole history of your progress. Think of us like GitHub for deep learning.
We are building our library of deep learning articles, and we're delighted to feature the work of community members. Contact Carey to learn about opportunities to share your research and insights.