From 0c30a743bd51846727447b633a4d545e8944958f Mon Sep 17 00:00:00 2001 From: Alexey <AlexeyAB@users.noreply.github.com> Date: Wed, 11 Jan 2017 15:14:49 +0000 Subject: [PATCH] Update Readme.md - Custom object training --- README.md | 10 ++++++++++ 1 files changed, 10 insertions(+), 0 deletions(-) diff --git a/README.md b/README.md index e86062a..ca4a1f2 100644 --- a/README.md +++ b/README.md @@ -211,6 +211,16 @@ 8. Start training by using the command line: `darknet.exe detector train data/obj.data yolo-obj.cfg darknet19_448.conv.23` +9. After training is complete - get result `yolo-obj_final.weights` from path `build\darknet\x64\backup\` + + * Also you can get result earlier than all 45000 iterations, for example, usually sufficient 2000 iterations for each class(object). I.e. for 6 classes to avoid overfitting - you can stop training after 12000 iterations and use `yolo-obj_12000.weights` to detection. + +### Custom object detection: + +Example of custom object detection: `darknet.exe detector test data/obj.data yolo-obj.cfg yolo-obj_3000.weights` + +|  |  | +|---|---| ## How to mark bounded boxes of objects and create annotation files: -- Gitblit v1.10.0