291c3ec9c7
updates
2019-07-08 15:02:20 +02:00
68b50f5cb6
updates
2019-07-08 12:43:15 +02:00
7a2d356297
GIoU to default
2019-07-07 23:53:56 +02:00
af3c5d0e35
GIoU to default
2019-07-07 23:42:24 +02:00
2b863c3bf2
Merge remote-tracking branch 'origin/master'
2019-07-07 23:24:43 +02:00
70f6379601
GIoU to default
2019-07-07 23:24:34 +02:00
b0aadff56f
Update requirements.txt
2019-07-05 12:34:17 +02:00
32a52dfb02
GIoU to default
2019-07-05 12:33:37 +02:00
429bd3b8a9
GIoU to default
2019-07-05 11:41:43 +02:00
b649a95c9a
GIoU to default
2019-07-05 00:36:37 +02:00
7246dd855c
updates
2019-07-04 22:50:03 +02:00
abf59f1565
updates
2019-07-04 22:10:46 +02:00
d0eace6cec
updates
2019-07-04 21:34:33 +02:00
5d5a7e0273
Merge remote-tracking branch 'origin/master'
2019-07-04 20:45:02 +02:00
1283a1e7e5
updates
2019-07-04 20:43:20 +02:00
a5592093ef
Merge remote-tracking branch 'origin/master'
2019-07-04 14:03:21 +02:00
7a353a9c70
updates
2019-07-04 14:03:13 +02:00
109991198c
updates
2019-07-03 16:18:08 +02:00
1e62ee2152
updates
2019-07-03 16:17:46 +02:00
ab141fcc1f
updates
2019-07-03 15:37:04 +02:00
1d0a4a3ace
updates
2019-07-03 14:42:11 +02:00
a8cf64af31
updates
2019-07-02 18:21:28 +02:00
ccf757b3ea
changed the criteria for the best weight file ( #356 )
...
* changed the criteria for the best weight file
changed the criteria for the best weight file from loss to mAP
I trained the model on my custom dataset. But I failed to get a good results when I load the weight file that has the lowest loss on test dataset.
I thought that the loss used in YOLO is not proper criteria for detection performance. So I changed the criteria from loss to mAP.
what do you think of this?
* Update train.py
2019-07-02 12:24:18 +02:00
1fd871abd8
updates
2019-07-01 17:44:42 +02:00
f43ee6ef94
updates
2019-07-01 17:17:29 +02:00
cf51cf9c99
updates
2019-07-01 17:14:42 +02:00
05358accbb
updates
2019-07-01 15:23:30 +02:00
c4409aa2ed
updates
2019-07-01 15:22:22 +02:00
b0d62e5204
updates
2019-07-01 15:21:06 +02:00
5e2b802f68
updates
2019-07-01 14:48:44 +02:00
09d065711a
updates
2019-07-01 01:27:32 +02:00
63036deeb7
updates
2019-07-01 00:41:13 +02:00
32f5ea955b
updates
2019-06-30 17:47:10 +02:00
db2674aa31
updates
2019-06-30 17:34:29 +02:00
5927d12aa7
Merge remote-tracking branch 'origin/master'
2019-06-30 15:24:43 +02:00
388b66dcd0
updates
2019-06-30 15:24:34 +02:00
1990cd8013
Update README.md
2019-06-30 00:38:32 +02:00
eeae43c414
updates
2019-06-28 00:38:52 +02:00
b202baa31c
update parse_model_cfg() ( #350 )
...
Removing the two lines for adding batch_normalize key to convolutional layers causes the parsing of the model to break when parsing it in models.py
2019-06-28 00:35:24 +02:00
c1bb037cbe
Merge remote-tracking branch 'origin/master'
2019-06-26 11:28:06 +02:00
cbfc5a00e5
updates
2019-06-26 11:27:36 +02:00
37fe87ccd9
cleanup- delete a variable "yolo_layer_count" ( #347 )
2019-06-26 11:10:52 +02:00
45540c787f
updates
2019-06-25 19:36:11 +02:00
1d76751e1f
updates
2019-06-25 12:37:24 +02:00
e4cc830690
updates
2019-06-25 12:21:27 +02:00
9a56d97059
updates
2019-06-25 11:54:19 +02:00
2244c72a1b
updates
2019-06-25 11:45:38 +02:00
6b222df35d
updates
2019-06-24 15:56:20 +02:00
d208f006a1
updates
2019-06-24 14:46:00 +02:00
a3fcf20385
updates
2019-06-24 14:19:20 +02:00
Glenn Jocher
Yonghye Kwon
Jeremy Hu