160eddddc4
Minor fix
2018-05-27 16:53:42 +03:00
b36512ea2e
classifier.c - add the awesome training chart and make sure "top" is not bigger than "classes" in datacfg file.
2018-05-26 21:42:53 +08:00
65e430588d
Merge pull request #909 from jing-vision/master
...
classifier.c - set batch=1 in "predict", "demo" and "try" mode.
2018-05-26 16:36:30 +03:00
6b6f0f0ca3
classifier.c - set batch=1 in "predict", "demo" and "try" mode.
2018-05-26 21:18:30 +08:00
6390a5a2ab
Update Readme.md
2018-05-23 19:11:10 +03:00
f52040f0bf
Classifier uses C++ OpenCV cv::VideoCapture that hasn't bugs.
2018-05-23 18:43:51 +03:00
eef9f8e5bb
Minor fixes
2018-05-23 18:35:08 +03:00
ec68838342
Fixed memory leaks for Yolo: train, test
2018-05-23 18:27:18 +03:00
768b190b10
Minor fix: warning-message if in labels class_id >= classes in cfg-file
2018-05-22 23:00:22 +03:00
60cd28bdfb
Fixed seg-fault that occurred if <class_id> in labels more than classes= in the cfg-file.
2018-05-22 20:10:04 +03:00
a2666d5b3c
Fixed replace_image_to_label(): replaces directories only for Pascal VOC and COCO, replaces only extensions for other.
2018-05-22 18:23:48 +03:00
032acd408a
Experimental reinforcement learning.
2018-05-22 16:38:04 +03:00
f9ecf6fd3f
For Release build - equality check: output of [convolutional] == input of [yolo].
2018-05-22 02:37:52 +03:00
d01a686246
Fixed compile
2018-05-22 00:33:24 +03:00
a720bf74e2
Added flag -ext_output for detector demo. Fixed proportional resize of network for random=1
2018-05-22 00:15:04 +03:00
46028c6621
Fixed demo for video files: is used do_nms_sort() instead of do_nms_obj().
2018-05-21 19:12:55 +03:00
b88b7b645b
Save jpg-images for detector demo if used flag: -prefix <filename>
2018-05-21 15:04:10 +03:00
4403e71b33
Supported files for training: jpg, jpeg, bmp, png, ppm - and UPPER CASE extensions
2018-05-20 18:43:15 +03:00
175c68b9b3
Merge pull request #856 from vinjn/patch-1
...
darknet.py - improve dll loading under Windows
2018-05-20 16:25:48 +03:00
c6ca54b411
darknet.py - apply the same changes to build/darknet/x64/
2018-05-20 21:12:35 +08:00
dbcdac88a5
darknet.py - loading dll files by absolute name
2018-05-20 14:20:55 +08:00
efdeb01a3a
darknet.py - remove unnecessary check of CUDA_HOME
2018-05-20 13:57:21 +08:00
7b7c669f49
Added flag -save_labels to save detections as training labels txt-files (pseudo labeling concept)
2018-05-20 00:39:03 +03:00
c1bb8c129d
Fixed xnor for random=1
2018-05-19 16:52:05 +03:00
94c84da850
Update Readme.md
2018-05-18 23:31:20 +03:00
8b9af80a5b
Merge pull request #851 from NadeenUdantha/patch-1
...
ojbects -> objects
2018-05-18 23:27:59 +03:00
985ba2720f
yolov3-tiny_obj.cfg - is example for training custom objects
2018-05-18 23:27:27 +03:00
ae9fb2abbd
ojbects -> objects
2018-05-18 14:19:19 +05:30
24f563ce71
Example of usage cv2.imread() in the darknet.py
2018-05-17 23:26:55 +03:00
71437a936d
Update Readme.md
2018-05-17 22:50:27 +03:00
9ba0beb413
Fixed resize of network for low resolution network like 128x128
2018-05-17 17:02:42 +03:00
e39d101109
Merge pull request #844 from salbatron/patch-1
...
Create get_openimages_dataset.py
2018-05-17 16:36:50 +03:00
0ecbeeff42
Create get_openimages_dataset.py
...
Script that downloads images with selected classes from OpenImages dataset and create labels in "yolo format"
2018-05-17 15:22:49 +02:00
fb9fcfb3ba
Support of .ppm files is realised
2018-05-17 03:04:42 +03:00
e29fcb703f
Fixed time for ./darknet detector valid
2018-05-16 15:23:34 +03:00
8587c0a161
Fixed OPENCV_VERSION
2018-05-16 15:19:01 +03:00
b34c913d2e
darknet.py change libdarknet.so to darknet.so
2018-05-16 15:16:41 +03:00
89c11f83ed
Fixed darknet.py - uses batch=1 by default
2018-05-16 15:05:46 +03:00
573d7e8081
Added def array_to_image(arr): to the darknet.py
...
Usage:
import scipy.misc
sci_image = scipy.misc.imread(image)
im, arr = array_to_image(sci_image)
2018-05-16 03:03:23 +03:00
8b5344ee2d
Added BFLOPs output for network configurations
2018-05-14 13:34:40 +03:00
3ebcc647b6
Fixed network resizing (random=1) for non-square networks
2018-05-14 12:20:38 +03:00
cda8171feb
Merge pull request #765 from HotChick91/AlexeyAB-code
...
Exit with nonzero status on error
2018-05-10 01:38:02 +03:00
812d143ebd
Merge branch 'master' of github.com:AlexeyAB/darknet
2018-05-08 19:09:58 +03:00
e96401774b
Fixed compile error
2018-05-08 19:09:43 +03:00
2c2ba0f8fa
Update Readme.md
2018-05-08 18:08:21 +03:00
fb56f6d569
Fixed README.md
2018-05-08 18:02:23 +03:00
0948df52b8
Merge pull request #741 from IlyaOvodov/Fix_detector_output
...
Output improvements for detector results:
2018-05-08 17:27:45 +03:00
1b5afb4583
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
Upd:
* struct detection changes reverted to keep it unchanged
* cmd line key "-ext_output" is added to manage extended output (rect coords in detector test mode)
# Conflicts:
# src/detector.c
2018-05-08 14:09:35 +03:00
0045074f56
Get pre-trained weights for Yolo v3 tiny
2018-05-07 23:33:10 +03:00
ae4ec1a06d
Exit with nonzero status on error
2018-05-07 22:11:57 +02:00
