### Build Log: Zero to AI – **Day 2**
Hey everyone! Welcome back to Day 2 of the Zero to AI journey. Today was filled with a mix of excitement, a bit of frustration, and some solid progress. Let’s dive in!
First off, I managed to set up the development environment. After a few hiccups with library dependencies (seriously, why does everyone use different versions?), I got everything installed and running. I went with Python and TensorFlow for the initial framework since it’s widely used and has tons of resources available. Honestly, the setup took way longer than I anticipated — note to self: double-check compatibility before diving in next time.
Once the environment was ready, I started sketching out the basic outline of the AI model. The plan is to create a simple neural network that can classify images. I spent some time researching different architectures and settled on a CNN (Convolutional Neural Network) as it’s pretty standard for image classification tasks. I’m feeling good about this choice for now, but I’m sure I’ll second-guess it later!
The challenge today came when I tried to find a suitable dataset. I initially thought I could just grab a few images online, but it quickly became clear that I need a well-structured dataset. I stumbled upon the CIFAR-10 dataset, which includes 60,000 32×32 color images across 10 classes. Perfect! But I found myself combing through the documentation and feeling a bit lost in the nuances of preprocessing the data. It’s a little overwhelming, but I’m keeping the mindset that this is all part of the learning curve.
So, what’s next? My priority for tomorrow is to get the CIFAR-10 images loaded and preprocessed for training. I also want to prototype the CNN architecture so that I can start running some preliminary tests. I plan to dive into the Keras API for building the model since it’s relatively straightforward. It seems like a great way to get the architecture up and running without needing to code everything from scratch.
That’s about it for today! Overall, it feels good to be making progress, even though it’s a bit slow. The challenges are just part of the journey, and I’m learning so much already. Here’s to tackling more tomorrow! Catch you all then!