Proposal for Exercise in Unit 2 #295
michaelshekasta
started this conversation in
02. Convolutional Neural Networks
Replies: 1 comment 2 replies
-
For graded assignment - We can utilize a well-established dataset like ImageNet and evaluate the outcomes using the test dataset. The results can be showcased on a leaderboard hosted by Gradio. Furthermore, I suggest incorporating additional components such as labs. In my opinion, delving into the contrasting approaches—widely covering single CNN with numerous kernels versus depth (employing two smaller CNNs sequentially)—will provide valuable insights. What do you think? |
Beta Was this translation helpful? Give feedback.
2 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Greetings everyone,
I'd like to propose a slight adjustment to our approach for Unit 2. Rather than focusing solely on assignments and grades, I believe it would be beneficial to emphasize tasks that foster a deeper understanding among learners.
In my opinion, this unit could be structured without formal grading, but instead, we could introduce a task that encourages hands-on exploration. One idea I had is to delve into a simple yet insightful task, such as utilizing the "minst" dataset, and experimenting with different CNN architectures. For instance, we could compare the performance of a single CNN with numerous kernels against the effectiveness of employing two CNNs sequentially, among other variations.
I'm eager to gather your feedback and insights regarding this proposal.
Best regards,
Michael
Beta Was this translation helpful? Give feedback.
All reactions