Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

yolov3-tiny implementation #87

Open
wants to merge 25 commits into
base: master
Choose a base branch
from
Open

yolov3-tiny implementation #87

wants to merge 25 commits into from

Conversation

andrewfurey21
Copy link
Contributor

@andrewfurey21 andrewfurey21 commented Jul 1, 2024 •
edited
Loading

Implementation of the yolov3-tiny architecture for Google Summer of Code.

  • YOLO forward pass
  • YOLO backward pass
  • YOLO Loss forward pass
  • YOLO loss backward pass
  • Implement YOLOv3-tiny
  • Documentation

The mlpack model is used here which has a bunch of preprocessing and post processing steps too needed for inference.

@andrewfurey21 andrewfurey21 marked this pull request as draft July 1, 2024 16:30
@andrewfurey21
added mask and anchors, updated maxpool block
d8f90e6
@andrewfurey21
layer and loss headers
972ea62
@andrewfurey21
fixed pooling layer
9e4d14c
@andrewfurey21
conv block, yolo loss formula
c0a88ef
@andrewfurey21
return layers instead
78133a7
@andrewfurey21
added loss and updated pooling block
55fadee
@andrewfurey21
all layers
548f365
@andrewfurey21
updated forward pass of yolov3 layer
4e84ac4
@andrewfurey21
normalize
c34f308
@andrewfurey21
use heightChannel variable
39b38f0
@andrewfurey21
Merge commit 'c34f308b50048097723ac4111e61659c52369429'
9ff9816
@andrewfurey21
correct dimensions for large object detections model
df617e3
@andrewfurey21
correct output dims for small object detections model
ee7ef21
@andrewfurey21
use multilayer instead of ffn
2aeb78c
@andrewfurey21
upsampled model using multilayer instead of ffn
b7994c4
@andrewfurey21
updated upsampled multilayer
c653ee4
@andrewfurey21
upsampled multilayer input dimensions
3d138a3
@andrewfurey21
updated yolo layer
c1838cf
@andrewfurey21
removed Gradient method since this layer has no weights
0a9ff95
@andrewfurey21
dont need anchors, mask and classes inside yolo layer
6d6b430
@andrewfurey21
corrected small object detection ffn
0f991dc
@andrewfurey21
forward yolo loss
09dbbc1
@andrewfurey21
correct loss function width and height
8822d59
@andrewfurey21
width and height are params for loss
4667a38
@shrit shrit marked this pull request as ready for review August 31, 2024 11:40
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
s: needs review s: unanswered s: unlabeled
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@andrewfurey21