From efaf684cb20b996db0cd6d99d20fabb4bc26859a Mon Sep 17 00:00:00 2001 From: AlexeyAB <alexeyab84@gmail.com> Date: Mon, 04 Jun 2018 12:38:18 +0000 Subject: [PATCH] By default letter_box is disabled --- README.md | 12 ++++++++++-- 1 files changed, 10 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index b56299e..8e93f9b 100644 --- a/README.md +++ b/README.md @@ -218,7 +218,7 @@ More information about training by the link: http://pjreddie.com/darknet/yolo/#train-voc - **Note:** If during training you see `nan` values in some lines then training goes well, but if `nan` are in all lines then training goes wrong. + **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. ## How to train with multi-GPU: @@ -317,7 +317,7 @@ * Also you can get result earlier than all 45000 iterations. - **Note:** If during training you see `nan` values in some lines then training goes well, but if `nan` are in all lines then training goes wrong. + **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. ### How to train tiny-yolo (to detect your custom objects): @@ -421,6 +421,14 @@ * for training with a large number of objects in each image, add the parameter `max=200` or higher value in the last layer [region] in your cfg-file + * for training for small objects - set `layers = -1, 11` instead of https://github.com/AlexeyAB/darknet/blob/6390a5a2ab61a0bdf6f1a9a6b4a739c16b36e0d7/cfg/yolov3.cfg#L720 + and set `stride=4` instead of https://github.com/AlexeyAB/darknet/blob/6390a5a2ab61a0bdf6f1a9a6b4a739c16b36e0d7/cfg/yolov3.cfg#L717 + + * General rule - you should keep relative size of objects in the Training and Testing datasets the same: + + * `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 2. After training - for detection: -- Gitblit v1.10.0