Add YOLOv5 integration - #5
Conversation
|
Good job! You can add a |
fanqiNO1
left a comment
fanqiNO1
left a comment
There was a problem hiding this comment.
I think there is no obvious problem.
Gennadiyev
left a comment
Gennadiyev
left a comment
There was a problem hiding this comment.
Quite a lot of revamps or explanations needed. I believe the sub-branch will work as-is, but I'd really like to trim the repo down a bit.
Also, I found that this YOLO implementation contains a lot of GPL-licensed code. This may infringe our MIT license.
| @@ -0,0 +1,19 @@ | |||
| # Yolo_unipose | |||
There was a problem hiding this comment.
I see. And I will improved code quality in my future work.
| @@ -0,0 +1,8 @@ | |||
| """Implements UniPose model. | |||
There was a problem hiding this comment.
Change to YOLO-related implementation.
Wait I don't think you implemented UniPose model in this file...
| from .yolo_unipose import yolo_unipose | ||
|
|
||
|
|
||
| __all__ = ["yolo_unipose"] No newline at end of file |
There was a problem hiding this comment.
Add newline \n to end of file.
| @@ -0,0 +1,653 @@ | |||
| # YOLOv5 🚀 by Ultralytics, GPL-3.0 license | |||
| """ | |||
| Export a YOLOv5 PyTorch model to other formats. TensorFlow exports authored by https://github.com/zldrobit | |||
There was a problem hiding this comment.
We do not need an export feature. Should be removed or re-factored.
| @@ -0,0 +1,649 @@ | |||
| # YOLOv5 🚀 by Ultralytics, GPL-3.0 license | |||
There was a problem hiding this comment.
Unipose convention is to use layers.py instead.
Formalize YOLO README
|
The merge is quite a shortcut and sooner than it should be. Since #4 requires YOLO urgently for data preprocessing, automated annotation and such, this pull request is merged at a pre-mature state. More YOLO-related patches will come to |
3 participants
yolo-unipose, which is a pre-processing tool, is used to to get the main creature in the image, judge its kind, and then return the bounding box of the creature.