Deep Learning (convnet) Solution for Image Classification: Hot Dog or Not Hot Dog?

Here I address the simple (or so it seems) question: given an image does it contain a hot dog? The main problem is the lack of training data (images), we augment our data, and then use a 2d convolutional neural network to make our predictions; the optimization algorithm and related hyper-parameters are critical so I discuss those as well.

less than 1 minute read

Mitch Murphy

Improving humanity by teaching machines how to learn.

You can find my Jupyter notebook for this analysis here.

Share on

Twitter Facebook LinkedIn

Cluster Mesh with Cilium Overview

20 minute read

Cluster Mesh with Cilium Overview

Vagrant + RKE2 + Cilium

5 minute read

Use Vagrant to provision and install RKE2 and Cilium

Cilium and Istio Harmony

1 minute read

Update Cilium to work with Istio and create a self-signed certificate for Istio.

Adding Persistent Storage to RKE2 with Longhorn

3 minute read

Configure Longhorn to provide persistent storage to RKE2 clusters using Flux CD

Deep Learning (convnet) Solution for Image Classification: Hot Dog or Not Hot Dog?

Mitch Murphy

Share on

You may also enjoy

Cluster Mesh with Cilium Overview

Vagrant + RKE2 + Cilium

Cilium and Istio Harmony

Adding Persistent Storage to RKE2 with Longhorn