Colab update for robotics #1002
Conversation
|
Note Gemini is unable to generate a summary for this pull request due to the file types involved not being currently supported. |
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
github-actions
bot
added
status:awaiting review
|
View / edit / reply to this conversation on ReviewNB Giom-V commented on 2025-10-23T12:17:27Z You need to decide if you want to show the code or not, and you should only use the colab @title magics for specific case when you want to hide things as those titles won't show in GitHub. I think this is such a case here.
|
|
View / edit / reply to this conversation on ReviewNB Giom-V commented on 2025-10-23T12:17:28Z I think you should explain why thinking = 0 |
|
View / edit / reply to this conversation on ReviewNB Giom-V commented on 2025-10-23T12:17:29Z Add a line to explain that you are loading the example images. |
|
View / edit / reply to this conversation on ReviewNB Giom-V commented on 2025-10-23T12:17:30Z You should explain what pointing is |
|
View / edit / reply to this conversation on ReviewNB Giom-V commented on 2025-10-23T12:17:30Z Explain the difference with the previous case. |
|
View / edit / reply to this conversation on ReviewNB Giom-V commented on 2025-10-23T12:17:31Z I think the explanation would be better as text than as a comment |
|
View / edit / reply to this conversation on ReviewNB Giom-V commented on 2025-10-23T12:17:32Z I think you should really explain in detail what's the difference with the previous cases |
|
View / edit / reply to this conversation on ReviewNB Giom-V commented on 2025-10-23T12:17:33Z Add an explanation about what your doing (IIUC extracting the frames from the GIF and sending them to the model) |
|
View / edit / reply to this conversation on ReviewNB Giom-V commented on 2025-10-23T12:17:33Z I think a minimal explanation on what bounding boxes and what the format is needed (just copy the one from the spatial understanding notbook). Also explain that you are loading more example images. |
|
View / edit / reply to this conversation on ReviewNB Giom-V commented on 2025-10-23T12:17:34Z Explain what trajectories are |
|
View / edit / reply to this conversation on ReviewNB Giom-V commented on 2025-10-23T12:17:35Z I think you need to remove the @title here |
|
View / edit / reply to this conversation on ReviewNB Giom-V commented on 2025-10-23T12:17:35Z Explain what's added compared to the previous case |
|
View / edit / reply to this conversation on ReviewNB Giom-V commented on 2025-10-23T12:17:36Z Explain what spatial reasoning is. |
|
View / edit / reply to this conversation on ReviewNB Giom-V commented on 2025-10-23T12:17:36Z I think video analysis should be a higher title and needs an introductory explanation. |
|
View / edit / reply to this conversation on ReviewNB Giom-V commented on 2025-10-23T12:17:37Z Why are we back to simple image understanding while we were reviewing videos? I think this section and the next one should be moved somewhere else. |
|
View / edit / reply to this conversation on ReviewNB Giom-V commented on 2025-10-23T12:17:38Z Explain why you are suddenly enabling thinking. What's special in this use-case? (also explain that thinking=-1 means thinking with automatic budget) |
|
View / edit / reply to this conversation on ReviewNB Giom-V commented on 2025-10-23T12:17:39Z Why are you suddenly loading new images without a new section or explanation? Maybe you should group all the wget together at the beginning?
Other possibility, add a "Multi-view" section and explain what it is (you should do taht anyway) |
|
View / edit / reply to this conversation on ReviewNB Giom-V commented on 2025-10-23T12:17:39Z Explain what code execution is and link to the relevant notebook |
|
View / edit / reply to this conversation on ReviewNB Giom-V commented on 2025-10-23T12:17:40Z Explain what segmentation is and maybe link to the relevant notebook |
|
View / edit / reply to this conversation on ReviewNB Giom-V commented on 2025-10-23T12:17:41Z Add links to the spatial understanding and spatial 3d notebooks |
|
Considering the length of the notebook, we should add a table of content and |
|
So we ended up with a robotics-samples repo. I think I'm going to break this one into multiple smaller colabs there, then use this as a table of contents. |
2 participants
No description provided.