From e1e5abe19197f2bb56a9d320b31275db4706aa4d Mon Sep 17 00:00:00 2001
From: AlexeyAB <alexeyab84@gmail.com>
Date: Fri, 15 Jun 2018 22:27:11 +0000
Subject: [PATCH] Fixed some security issues
---
README.md | 4 +++-
1 files changed, 3 insertions(+), 1 deletions(-)
diff --git a/README.md b/README.md
index 5f81cd3..b40d531 100644
--- a/README.md
+++ b/README.md
@@ -320,6 +320,8 @@
**Note:** If during training you see `nan` values for `avg` (loss) field - then training goes wrong, but if `nan` is in some other lines - then training goes well.
+ **Note:** If you changed width= or height= in your cfg-file, then new width and height must be divisible by 32.
+
### How to train tiny-yolo (to detect your custom objects):
Do all the same steps as for the full yolo model as described above. With the exception of:
@@ -432,7 +434,7 @@
* `train_network_width * train_obj_width / train_image_width ~= detection_network_width * detection_obj_width / detection_image_width`
* `train_network_height * train_obj_height / train_image_height ~= detection_network_height * detection_obj_height / detection_image_height`
- * to speedup training (with decreasing detection accuracy) do Fine-Tuning instead of Transfer-Learning, set param `stopbackward=1` in one of the penultimate convolutional layers before the 1-st `[yolo]`-layer, for example here: https://github.com/AlexeyAB/darknet/blob/0039fd26786ab5f71d5af725fc18b3f521e7acfd/cfg/yolov3.cfg#L598
+ * to speedup training (with decreasing detection accuracy) do Fine-Tuning instead of Transfer-Learning, set param `stopbackward=1` here: https://github.com/AlexeyAB/darknet/blob/6d44529cf93211c319813c90e0c1adb34426abe5/cfg/yolov3.cfg#L548
2. After training - for detection:
--
Gitblit v1.10.0