Today, I completed the introductory "Is it a bird?" notebook and got pretty stoked to see how the resnet18 model can generate predictions with such high accuracy with just a few lines of code.
I'm really interested in how all this works under the hood and excited to dig deeper.
I also finished the first chapter of the book, which covered similar material to the video and notebook. According to the authors, it provides a deeper understanding of the concepts, which I find quite engaging thus far.
Sadly I won't have more time to dedicate to learning today. But, I believe that even a little bit of progress is better than none at all.
What I learned:
- How to use the
resnet18model to fine-tune it to a new dataset
Time commitment:
Total: 1 hour