From 028696bf15efeca3acb3db8c42a96f7b9e0f55ff Mon Sep 17 00:00:00 2001
From: iovodov <b@ovdv.ru>
Date: Thu, 03 May 2018 13:33:46 +0000
Subject: [PATCH] Output improvements for detector results: When printing detector results, output was done in random order, obfuscating results for interpreting. Now: 1. Text output includes coordinates of rects in (left,right,top,bottom in pixels) along with label and score 2. Text output is sorted by rect lefts to simplify finding appropriate rects on image 3. If several class probs are > thresh for some detection, the most probable is written first and coordinates for others are not repeated 4. Rects are imprinted in image in order by their best class prob, so most probable rects are always on top and not overlayed by less probable ones 5. Most probable label for rect is always written first Also: 6. Message about low GPU memory include required amount
---
README.md | 48 ++++++++++++++++++++++++++++++++----------------
1 files changed, 32 insertions(+), 16 deletions(-)
diff --git a/README.md b/README.md
index 0830e12..f0549c9 100644
--- a/README.md
+++ b/README.md
@@ -17,9 +17,10 @@
-|  |  https://pjreddie.com/media/files/papers/YOLOv3.pdf |
+|  |  mAP (AP50) https://pjreddie.com/media/files/papers/YOLOv3.pdf |
|---|---|
+* Yolo v3 source chart for the RetinaNet on MS COCO got from Table 1 (e): https://arxiv.org/pdf/1708.02002.pdf
* Yolo v2 on Pascal VOC 2007: https://hsto.org/files/a24/21e/068/a2421e0689fb43f08584de9d44c2215f.jpg
* Yolo v2 on Pascal VOC 2012 (comp4): https://hsto.org/files/3a6/fdf/b53/3a6fdfb533f34cee9b52bdd9bb0b19d9.jpg
@@ -45,7 +46,7 @@
* **OpenCV 3.4.0**: https://sourceforge.net/projects/opencvlibrary/files/opencv-win/3.4.0/opencv-3.4.0-vc14_vc15.exe/download
* **or OpenCV 2.4.13**: https://sourceforge.net/projects/opencvlibrary/files/opencv-win/2.4.13/opencv-2.4.13.2-vc14.exe/download
- OpenCV allows to show image or video detection in the window and store result to file that specified in command line `-out_filename res.avi`
-* **GPU with CC >= 2.0** if you use CUDA, or **GPU CC >= 3.0** if you use cuDNN + CUDA: https://en.wikipedia.org/wiki/CUDA#GPUs_supported
+* **GPU with CC >= 3.0**: https://en.wikipedia.org/wiki/CUDA#GPUs_supported
##### Pre-trained models for different cfg-files can be downloaded from (smaller -> faster & lower quality):
* `yolov3.cfg` (236 MB COCO **Yolo v3**) - require 4 GB GPU-RAM: https://pjreddie.com/media/files/yolov3.weights
@@ -124,6 +125,7 @@
Before make, you can set such options in the `Makefile`: [link](https://github.com/AlexeyAB/darknet/blob/9c1b9a2cf6363546c152251be578a21f3c3caec6/Makefile#L1)
* `GPU=1` to build with CUDA to accelerate by using GPU (CUDA should be in `/usr/local/cuda`)
* `CUDNN=1` to build with cuDNN v5-v7 to accelerate training by using GPU (cuDNN should be in `/usr/local/cudnn`)
+* `CUDNN_HALF=1` to build for Tensor Cores (on Titan V / Tesla V100 / DGX-2 and later) speedup Detection 3x, Training 2x
* `OPENCV=1` to build with OpenCV 3.x/2.4.x - allows to detect on video files and video streams from network cameras or web-cams
* `DEBUG=1` to bould debug version of Yolo
* `OPENMP=1` to build with OpenMP support to accelerate Yolo by using multi-core CPU
@@ -156,6 +158,8 @@
4.2 (right click on project) -> properties -> Linker -> General -> Additional Library Directories: `C:\opencv_2.4.13\opencv\build\x64\vc14\lib`
+5. If you have GPU with Tensor Cores (nVidia Titan V / Tesla V100 / DGX-2 and later) speedup Detection 3x, Training 2x:
+ `\darknet.sln` -> (right click on project) -> properties -> C/C++ -> Preprocessor -> Preprocessor Definitions, and add here: `CUDNN_HALF;`
### How to compile (custom):
@@ -166,7 +170,7 @@
`C:\opencv_3.0\opencv\build\include;..\..\3rdparty\include;%(AdditionalIncludeDirectories);$(CudaToolkitIncludeDir);$(cudnn)\include`
- (right click on project) -> Build dependecies -> Build Customizations -> set check on CUDA 9.1 or what version you have - for example as here: http://devblogs.nvidia.com/parallelforall/wp-content/uploads/2015/01/VS2013-R-5.jpg
-- add to project all .c & .cu files from `\src`
+- add to project all `.c` & `.cu` files and file `http_stream.cpp` from `\src`
- (right click on project) -> properties -> Linker -> General -> Additional Library Directories, put here:
`C:\opencv_3.0\opencv\build\x64\vc14\lib;$(CUDA_PATH)lib\$(PlatformName);$(cudnn)\lib\x64;%(AdditionalLibraryDirectories)`
@@ -215,15 +219,20 @@
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.
+
## How to train with multi-GPU:
1. Train it first on 1 GPU for like 1000 iterations: `darknet.exe detector train data/voc.data cfg/yolov3-voc.cfg darknet53.conv.74`
-2. Then stop and by using partially-trained model `/backup/yolo-voc_1000.weights` run training with multigpu (up to 4 GPUs): `darknet.exe detector train data/voc.data cfg/yolov3-voc.cfg /backup/yolo-voc_1000.weights -gpus 0,1,2,3`
+2. Then stop and by using partially-trained model `/backup/yolov3-voc_1000.weights` run training with multigpu (up to 4 GPUs): `darknet.exe detector train data/voc.data cfg/yolov3-voc.cfg /backup/yolov3-voc_1000.weights -gpus 0,1,2,3`
https://groups.google.com/d/msg/darknet/NbJqonJBTSY/Te5PfIpuCAAJ
## How to train (to detect your custom objects):
+(to train old Yolo v2 `yolov2-voc.cfg`, `yolov2-tiny-voc.cfg`, `yolo-voc.cfg`, `yolo-voc.2.0.cfg`, ... [click by the link](https://github.com/AlexeyAB/darknet/tree/47c7af1cea5bbdedf1184963355e6418cb8b1b4f#how-to-train-pascal-voc-data))
+
+Training Yolo v3:
1. Create file `yolo-obj.cfg` with the same content as in `yolov3.cfg` (or copy `yolov3.cfg` to `yolo-obj.cfg)` and:
@@ -238,7 +247,7 @@
* https://github.com/AlexeyAB/darknet/blob/0039fd26786ab5f71d5af725fc18b3f521e7acfd/cfg/yolov3.cfg#L689
* https://github.com/AlexeyAB/darknet/blob/0039fd26786ab5f71d5af725fc18b3f521e7acfd/cfg/yolov3.cfg#L776
- So if `classes=1` then should be `filters=18`. If `classes=2` then write `filters=31`.
+ So if `classes=1` then should be `filters=18`. If `classes=2` then write `filters=21`.
**(Do not write in the cfg-file: filters=(classes + 5)x3)**
@@ -268,15 +277,17 @@
4. Put image-files (.jpg) of your objects in the directory `build\darknet\x64\data\obj\`
-5. Create `.txt`-file for each `.jpg`-image-file - in the same directory and with the same name, but with `.txt`-extension, and put to file: object number and object coordinates on this image, for each object in new line: `<object-class> <x> <y> <width> <height>`
+5. You should label each object on images from your dataset. Use this visual GUI-software for marking bounded boxes of objects and generating annotation files for Yolo v2 & v3: https://github.com/AlexeyAB/Yolo_mark
+
+It will create `.txt`-file for each `.jpg`-image-file - in the same directory and with the same name, but with `.txt`-extension, and put to file: object number and object coordinates on this image, for each object in new line: `<object-class> <x> <y> <width> <height>`
Where:
* `<object-class>` - integer number of object from `0` to `(classes-1)`
- * `<x> <y> <width> <height>` - float values relative to width and height of image, it can be equal from 0.0 to 1.0
+ * `<x> <y> <width> <height>` - float values relative to width and height of image, it can be equal from (0.0 to 1.0]
* for example: `<x> = <absolute_x> / <image_width>` or `<height> = <absolute_height> / <image_height>`
* atention: `<x> <y>` - are center of rectangle (are not top-left corner)
- For example for `img1.jpg` you should create `img1.txt` containing:
+ For example for `img1.jpg` you will be created `img1.txt` containing:
```
1 0.716797 0.395833 0.216406 0.147222
@@ -301,17 +312,21 @@
9. After training is complete - get result `yolo-obj_final.weights` from path `build\darknet\x64\backup\`
- * After each 1000 iterations you can stop and later start training from this point. For example, after 2000 iterations you can stop training, and later just copy `yolo-obj_2000.weights` from `build\darknet\x64\backup\` to `build\darknet\x64\` and start training using: `darknet.exe detector train data/obj.data yolo-obj.cfg yolo-obj_2000.weights`
+ * After each 100 iterations you can stop and later start training from this point. For example, after 2000 iterations you can stop training, and later just copy `yolo-obj_2000.weights` from `build\darknet\x64\backup\` to `build\darknet\x64\` and start training using: `darknet.exe detector train data/obj.data yolo-obj.cfg yolo-obj_2000.weights`
+
+ (in the original repository https://github.com/pjreddie/darknet the weights-file is saved only once every 10 000 iterations `if(iterations > 1000)`)
* 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.
+
### 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:
-* Download default weights file for tiny-yolo-voc: http://pjreddie.com/media/files/tiny-yolo-voc.weights
-* Get pre-trained weights tiny-yolo-voc.conv.13 using command: `darknet.exe partial cfg/tiny-yolo-voc.cfg tiny-yolo-voc.weights tiny-yolo-voc.conv.13 13`
-* Make your custom model `tiny-yolo-obj.cfg` based on `tiny-yolo-voc.cfg` instead of `yolo-voc.2.0.cfg`
-* Start training: `darknet.exe detector train data/obj.data tiny-yolo-obj.cfg tiny-yolo-voc.conv.13`
+* Download default weights file for yolov2-tiny-voc: http://pjreddie.com/media/files/yolov2-tiny-voc.weights
+* Get pre-trained weights yolov2-tiny-voc.conv.13 using command: `darknet.exe partial cfg/yolov2-tiny-voc.cfg yolov2-tiny-voc.weights yolov2-tiny-voc.conv.13 13`
+* Make your custom model `yolov2-tiny-obj.cfg` based on `cfg/yolov2-tiny-voc.cfg` instead of `yolov3.cfg`
+* Start training: `darknet.exe detector train data/obj.data yolov2-tiny-obj.cfg yolov2-tiny-voc.conv.13`
For training Yolo based on other models ([DenseNet201-Yolo](https://github.com/AlexeyAB/darknet/blob/master/build/darknet/x64/densenet201_yolo.cfg) or [ResNet50-Yolo](https://github.com/AlexeyAB/darknet/blob/master/build/darknet/x64/resnet50_yolo.cfg)), you can download and get pre-trained weights as showed in this file: https://github.com/AlexeyAB/darknet/blob/master/build/darknet/x64/partial.cmd
If you made you custom model that isn't based on other models, then you can train it without pre-trained weights, then will be used random initial weights.
@@ -363,6 +378,7 @@
* **mAP** (mean average precision) - mean value of `average precisions` for each class, where `average precision` is average value of 11 points on PR-curve for each possible threshold (each probability of detection) for the same class (Precision-Recall in terms of PascalVOC, where Precision=TP/(TP+FP) and Recall=TP/(TP+FN) ), page-11: http://homepages.inf.ed.ac.uk/ckiw/postscript/ijcv_voc09.pdf
+**mAP** is default metric of precision in the PascalVOC competition, **this is the same as AP50** metric in the MS COCO competition.
In terms of Wiki, indicators Precision and Recall have a slightly different meaning than in the PascalVOC competition, but **IoU always has the same meaning**.
@@ -397,12 +413,12 @@
* increase network resolution in your `.cfg`-file (`height=608`, `width=608` or any value multiple of 32) - it will increase precision
* recalculate anchors for your dataset for `width` and `height` from cfg-file:
- `darknet.exe detector calc_anchors data/obj.data -num_of_clusters 9 -width 416 -heigh 416`
+ `darknet.exe detector calc_anchors data/obj.data -num_of_clusters 9 -width 416 -height 416`
then set the same 9 `anchors` in each of 3 `[yolo]`-layers in your cfg-file
- * desirable that your training dataset include images with objects at diffrent: scales, rotations, lightings, from different sides
+ * desirable that your training dataset include images with objects at diffrent: scales, rotations, lightings, from different sides, on different backgrounds
- * desirable that your training dataset include images with objects (without labels) that you do not want to detect - negative samples
+ * desirable that your training dataset include images with non-labeled objects that you do not want to detect - negative samples without bounded box (empty `.txt` files)
* 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
--
Gitblit v1.10.0