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.

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

re-enable the "partial" command #83

Open
stephanecharette opened this issue Oct 18, 2024 · 0 comments
Open

re-enable the "partial" command #83

stephanecharette opened this issue Oct 18, 2024 · 0 comments
Assignees
Labels
enhancement New feature or request

Comments

@stephanecharette
Copy link
Collaborator

From the previous versions of Darknet:

Download file with the first 29-convolutional layers of yolov4-tiny: https://github.com/AlexeyAB/darknet/releases/download/darknet_yolo_v4_pre/yolov4-tiny.conv.29 (Or get this file from yolov4-tiny.weights file by using command: darknet.exe partialcfg/yolov4-tiny-custom.cfg yolov4-tiny.weights yolov4-tiny.conv.29 29

And:

darknet partial <CFG_FILE> <OLD_WEIGHTS_FILE> <NEW_WEIGHTS_FILE> <NUM_OF_LAYERS_TO_CUT>

The new V3 code that recognizes if loading the weights has failed due to either under-reading or attempting to over-read weights is currently triggering, as reported by a user on the discord tonight (#training-questions):

I'm attempting to use the "partial" command to cut the first n layers' weight values and utilize them when initiating a new training. However, when I try to load the cut weights, Darknet raises an error from the "xfread" method (expected to read 64 fields, but only read 0).

@stephanecharette stephanecharette added the enhancement New feature or request label Oct 18, 2024
@stephanecharette stephanecharette self-assigned this Oct 18, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

1 participant