Conversation
This will detect if a cat is on the screen. By which I mean displayed on the screen, not sitting on your laptop. This is meant as a simple demo of using MSS for AI. It works as-is, but needs to be documented, and there's some bits that could do with cleanup. There are a lot of additional features that could be added, such as showing a window with bounding boxes, but that's probably more complexity than is called for here.
|
I like it, great inspiration! |
halldorfannar
left a comment
halldorfannar
left a comment
There was a problem hiding this comment.
Excellent PR. Well documented, enjoyable to read. Just some minor improvements suggested.
demos/cat-detector.py
Outdated
| # Performance | ||
| # =========== | ||
| # | ||
| # The biggest determinant of performance is whether the model runs on a GPU or on the CPU. GPUs are extremely |
There was a problem hiding this comment.
Maybe we should mention here, right away, that this particular model will work on both? I know this becomes clearer in the end of this section, when GPU vs CPU performance comparisons are discussed.
There was a problem hiding this comment.
Adding to the next commit
demos/cat-detector.py
Outdated
| import torchvision.models.detection | ||
| import torchvision.transforms.v2 | ||
|
|
||
| # You'll also need to "pip install mss pillow". |
There was a problem hiding this comment.
Rather than assume the user is leveraging pip (I'm a fan of uv) I would suggest the more general:
| # You'll also need to "pip install mss pillow". | |
| # You'll also need to install mss and pillow. |
This also aligns with the earlier text where pip is suggested but it is left open for the user how to do this specifically.
There was a problem hiding this comment.
I prefer very specific commands in user examples, but I'll make the "pip" command an example rather than sounding like a requirement. I had kind of assumed that the user, if using something other than pip, would know how to adapt the command to their needs.
The reason I pointed the user to the PyTorch "Get Started" page wasn't to handle different package managers; that page still only gives commands for pip (plus building from source, and downloading C++ / Java packages). The main difference for our purposes is that it gives different --index-url flags depending on whether you want a CUDA, ROCm, or CPU-only build.
I'll make a change to soften the suggestion of package manager, but still leave the specific command present.
Co-authored-by: Halldor Fannar <halldorfannar@users.noreply.github.com>
My last version accidentally had an intermediate copy. This version prevents that.
4 participants
This will detect if a cat is on the screen. By which I mean displayed on the screen, not sitting on your laptop.
This is meant as a simple demo of using MSS for AI. It works as-is, but needs to be documented, and there's some bits that could do with cleanup.
There are a lot of additional features that could be added, such as showing a window with bounding boxes, but that's probably more complexity than is called for here.
Changes proposed in this PR
./check.shpassed - N/A (doesn't check demos)