tahaemara / yolo-custom-object-detector

Making custom object detector using Yolo (Java and Python)

Building a custom object detector using YOLO

This repository contains the code and the dataset for the tutorials (Part1 and Part2) I wrote about making custome object detector using YOLO in Java and Python. It also contains a dockerfile to build a docker image contains darknet framwork, OpenCV 3.3, and CUDA.

yolo-custom-object-detector's Issues

Using Protobuff model files from Darknet/Darkflow in Java

Instead of .data model files generated from the DL4J training API, can we use .pb files from Darknet or Darkflow training to be used for inference. Basically, I already have my trained .pb file and I want to input it to the detection component of your tutorial (because it seems to be faster than Tensorflow's inference).

OpenCV(3.4.5) + DNN (-215:Assertion failed) separator_index < line.size() in function 'ReadDarknetFromCfgStream'

python3 yolo_opencv.py -c custom/yolov3-tiny.cfg -w backup/yolov3-tiny_120000.weights -cl custom/objects.names
['bola']
Traceback (most recent call last):
File "yolo_opencv.py", line 47, in
net = cv2.dnn.readNet(args.weights,args.config)
cv2.error: OpenCV(3.4.5) /io/opencv/modules/dnn/src/darknet/darknet_io.cpp:507: error: (-215:Assertion failed) separator_index < line.size() in function 'ReadDarknetFromCfgStream'

I have this error on try your code... In darknet we can Run inference, and get results. You know the reason?

Can you share the train.py file to have the yolov3-tiny_final.weights?

I want to have a .weights file with my own data. Can you share the train.py file to us follow? Thanks.

False detection after many iterations

Hello all: I trained yolo net based on the procedure; after 10000 iterations, I got some false results (for your kindly information avg loss is about 0.001); please advise me. Best, Ghasem

Use it for Ubuntu 18.04

I want to use this custom yolo object detector in my Ubuntu 18.04 LTS. What are the things which I've to change in order to successfully build this docker for the aforementioned OS?

avg loss is too high while training of yolov3 tiny

i am using images of 640*480 pixels. my data is not too large and i used yolov3 tiny.cfg for training with darknet52.448 weight file not using any GPU.

i get very high avg loss from starting. i dont know why?

Region 16 Avg IOU: 0.544776, Class: 0.499554, Obj: 0.499566, No Obj: 0.499584, .5R: 0.625000, .75R: 0.000000, count: 8
Region 23 Avg IOU: -nan, Class: -nan, Obj: -nan, No Obj: 0.499721, .5R: -nan, .75R: -nan, count: 0
Region 16 Avg IOU: 0.567904, Class: 0.499282, Obj: 0.499585, No Obj: 0.499586, .5R: 0.714286, .75R: 0.142857, count: 7
Region 23 Avg IOU: -nan, Class: -nan, Obj: -nan, No Obj: 0.499723, .5R: -nan, .75R: -nan, count: 0
Region 16 Avg IOU: 0.528688, Class: 0.499494, Obj: 0.499567, No Obj: 0.499584, .5R: 0.500000, .75R: 0.000000, count: 8
Region 23 Avg IOU: -nan, Class: -nan, Obj: -nan, No Obj: 0.499722, .5R: -nan, .75R: -nan, count: 0
Region 16 Avg IOU: 0.693195, Class: 0.499361, Obj: 0.499523, No Obj: 0.499583, .5R: 0.833333, .75R: 0.333333, count: 6
Region 23 Avg IOU: -nan, Class: -nan, Obj: -nan, No Obj: 0.499722, .5R: -nan, .75R: -nan, count: 0
Region 16 Avg IOU: 0.575871, Class: 0.499474, Obj: 0.499555, No Obj: 0.499586, .5R: 0.833333, .75R: 0.166667, count: 6
Region 23 Avg IOU: -nan, Class: -nan, Obj: -nan, No Obj: 0.499720, .5R: -nan, .75R: -nan, count: 0
Region 16 Avg IOU: 0.551334, Class: 0.499472, Obj: 0.499480, No Obj: 0.499583, .5R: 0.750000, .75R: 0.000000, count: 8
Region 23 Avg IOU: -nan, Class: -nan, Obj: -nan, No Obj: 0.499722, .5R: -nan, .75R: -nan, count: 0
Region 16 Avg IOU: 0.600088, Class: 0.499542, Obj: 0.499506, No Obj: 0.499584, .5R: 0.800000, .75R: 0.000000, count: 5
Region 23 Avg IOU: -nan, Class: -nan, Obj: -nan, No Obj: 0.499721, .5R: -nan, .75R: -nan, count: 0
Region 16 Avg IOU: 0.572092, Class: 0.499428, Obj: 0.499505, No Obj: 0.499584, .5R: 0.666667, .75R: 0.000000, count: 6
Region 23 Avg IOU: -nan, Class: -nan, Obj: -nan, No Obj: 0.499724, .5R: -nan, .75R: -nan, count: 0
Region 16 Avg IOU: 0.553508, Class: 0.499499, Obj: 0.499542, No Obj: 0.499585, .5R: 0.875000, .75R: 0.000000, count: 8
Region 23 Avg IOU: -nan, Class: -nan, Obj: -nan, No Obj: 0.499720, .5R: -nan, .75R: -nan, count: 0
Region 16 Avg IOU: 0.553497, Class: 0.499381, Obj: 0.499442, No Obj: 0.499584, .5R: 0.555556, .75R: 0.111111, count: 9
Region 23 Avg IOU: -nan, Class: -nan, Obj: -nan, No Obj: 0.499722, .5R: -nan, .75R: -nan, count: 0
Region 16 Avg IOU: 0.580341, Class: 0.499422, Obj: 0.499497, No Obj: 0.499584, .5R: 0.833333, .75R: 0.000000, count: 6
Region 23 Avg IOU: 0.305393, Class: 0.499000, Obj: 0.499338, No Obj: 0.499720, .5R: 0.000000, .75R: 0.000000, count: 1
Region 16 Avg IOU: 0.550048, Class: 0.499547, Obj: 0.499663, No Obj: 0.499584, .5R: 0.666667, .75R: 0.000000, count: 3
Region 23 Avg IOU: 0.454548, Class: 0.499584, Obj: 0.499320, No Obj: 0.499721, .5R: 0.000000, .75R: 0.000000, count: 1
Region 16 Avg IOU: 0.558823, Class: 0.499531, Obj: 0.499570, No Obj: 0.499583, .5R: 0.625000, .75R: 0.000000, count: 8
Region 23 Avg IOU: -nan, Class: -nan, Obj: -nan, No Obj: 0.499721, .5R: -nan, .75R: -nan, count: 0
Region 16 Avg IOU: 0.615219, Class: 0.499384, Obj: 0.499519, No Obj: 0.499586, .5R: 1.000000, .75R: 0.000000, count: 6
Region 23 Avg IOU: -nan, Class: -nan, Obj: -nan, No Obj: 0.499723, .5R: -nan, .75R: -nan, count: 0
Region 16 Avg IOU: 0.564179, Class: 0.499493, Obj: 0.499501, No Obj: 0.499588, .5R: 0.833333, .75R: 0.000000, count: 6
Region 23 Avg IOU: -nan, Class: -nan, Obj: -nan, No Obj: 0.499722, .5R: -nan, .75R: -nan, count: 0
Region 16 Avg IOU: 0.546191, Class: 0.499529, Obj: 0.499633, No Obj: 0.499584, .5R: 0.571429, .75R: 0.000000, count: 7
Region 23 Avg IOU: -nan, Class: -nan, Obj: -nan, No Obj: 0.499721, .5R: -nan, .75R: -nan, count: 0
1: 439.307251, 439.307251 avg, 0.000000 rate, 262.300388 seconds, 64 images
Loaded: 0.000048 seconds

and this is the avg loss and rate readings

1: 439.307251, 439.307251 avg, 0.000000 rate, 262.300388 seconds, 64 images
2: 437.796631, 439.156189 avg, 0.000000 rate, 260.419217 seconds, 128 images
3: 436.743683, 438.914948 avg, 0.000000 rate, 263.185402 seconds, 192 images
4: 439.118439, 438.935303 avg, 0.000000 rate, 260.007587 seconds, 256 images
5: 440.381744, 439.079956 avg, 0.000000 rate, 261.064861 seconds, 320 images
6: 437.774689, 438.949432 avg, 0.000000 rate, 259.742917 seconds, 384 images
7: 438.066528, 438.861145 avg, 0.000000 rate, 257.299960 seconds, 448 images
8: 439.212860, 438.896301 avg, 0.000000 rate, 257.397450 seconds, 512 images
9: 438.903320, 438.897003 avg, 0.000000 rate, 258.272195 seconds, 576 images
10: 438.605133, 438.867828 avg, 0.000000 rate, 257.986108 seconds, 640 images
11: 383.567566, 433.337799 avg, 0.000000 rate, 217.779761 seconds, 704 images
12: 385.344147, 428.538422 avg, 0.000000 rate, 217.042249 seconds, 768 images
13: 383.292480, 424.013824 avg, 0.000000 rate, 217.776042 seconds, 832 images
14: 384.244659, 420.036896 avg, 0.000000 rate, 217.518784 seconds, 896 images
15: 383.643707, 416.397583 avg, 0.000000 rate, 217.775383 seconds, 960 images
16: 382.793121, 413.037140 avg, 0.000000 rate, 217.932038 seconds, 1024 images
17: 383.037781, 410.037201 avg, 0.000000 rate, 217.793683 seconds, 1088 images
18: 384.058350, 407.439331 avg, 0.000000 rate, 217.527402 seconds, 1152 images
19: 384.619659, 405.157349 avg, 0.000000 rate, 217.347416 seconds, 1216 images
20: 384.284943, 403.070099 avg, 0.000000 rate, 217.261834 seconds, 1280 images
21: 557.822144, 418.545288 avg, 0.000000 rate, 327.631583 seconds, 1344 images
22: 558.407227, 432.531494 avg, 0.000000 rate, 326.945323 seconds, 1408 images
23: 558.074524, 445.085785 avg, 0.000000 rate, 327.198063 seconds, 1472 images
24: 558.270020, 456.404205 avg, 0.000000 rate, 327.856426 seconds, 1536 images
25: 558.338806, 466.597656 avg, 0.000000 rate, 327.192073 seconds, 1600 images
26: 558.290649, 475.766968 avg, 0.000000 rate, 327.586775 seconds, 1664 images
27: 558.376099, 484.027893 avg, 0.000000 rate, 326.799530 seconds, 1728 images
28: 557.339539, 491.359070 avg, 0.000000 rate, 327.356004 seconds, 1792 images
29: 558.731384, 498.096313 avg, 0.000000 rate, 327.647542 seconds, 1856 images
30: 558.595276, 504.146210 avg, 0.000000 rate, 326.950869 seconds, 1920 images
31: 692.322754, 522.963867 avg, 0.000000 rate, 411.091572 seconds, 1984 images
32: 692.104065, 539.877869 avg, 0.000000 rate, 411.196345 seconds, 2048 images
33: 692.929993, 555.183105 avg, 0.000000 rate, 411.557181 seconds, 2112 images
34: 694.361206, 569.100891 avg, 0.000000 rate, 411.281385 seconds, 2176 images
35: 693.376526, 581.528442 avg, 0.000000 rate, 410.951036 seconds, 2240 images
36: 692.907471, 592.666321 avg, 0.000000 rate, 411.049841 seconds, 2304 images
37: 693.877014, 602.787415 avg, 0.000000 rate, 411.390684 seconds, 2368 images
38: 692.186951, 611.727356 avg, 0.000000 rate, 411.028736 seconds, 2432 images
39: 693.944397, 619.949036 avg, 0.000000 rate, 410.121289 seconds, 2496 images
40: 693.239868, 627.278137 avg, 0.000000 rate, 410.412040 seconds, 2560 images
41: 692.219482, 633.772278 avg, 0.000000 rate, 410.161699 seconds, 2624 images
42: 694.196045, 639.814636 avg, 0.000000 rate, 410.275656 seconds, 2688 images
43: 693.995300, 645.232727 avg, 0.000000 rate, 410.213628 seconds, 2752 images
44: 694.451599, 650.154602 avg, 0.000000 rate, 410.228155 seconds, 2816 images
45: 692.546448, 654.393799 avg, 0.000000 rate, 410.168366 seconds, 2880 images
46: 691.750488, 658.129456 avg, 0.000000 rate, 411.126075 seconds, 2944 images
47: 693.839478, 661.700439 avg, 0.000000 rate, 410.967428 seconds, 3008 images
48: 693.480652, 664.878479 avg, 0.000000 rate, 410.267494 seconds, 3072 images
49: 692.532898, 667.643921 avg, 0.000000 rate, 411.022943 seconds, 3136 images
50: 692.901367, 670.169678 avg, 0.000000 rate, 410.338903 seconds, 3200 images
51: 497.743286, 652.927063 avg, 0.000000 rate, 278.302808 seconds, 3264 images
52: 498.426666, 637.477051 avg, 0.000000 rate, 278.354224 seconds, 3328 images
53: 495.944183, 623.323792 avg, 0.000000 rate, 278.412660 seconds, 3392 images
54: 497.179016, 610.709290 avg, 0.000000 rate, 278.443549 seconds, 3456 images
55: 496.152832, 599.253662 avg, 0.000000 rate, 279.154921 seconds, 3520 images
56: 496.505615, 588.978882 avg, 0.000000 rate, 278.242445 seconds, 3584 images
57: 494.551849, 579.536194 avg, 0.000000 rate, 278.382127 seconds, 3648 images
58: 496.167511, 571.199341 avg, 0.000000 rate, 278.240654 seconds, 3712 images
59: 495.463837, 563.625793 avg, 0.000000 rate, 278.398038 seconds, 3776 images
60: 497.070984, 556.970337 avg, 0.000000 rate, 278.684142 seconds, 3840 images
61: 384.408966, 539.714172 avg, 0.000000 rate, 217.767682 seconds, 3904 images
62: 385.864685, 524.329224 avg, 0.000000 rate, 217.507124 seconds, 3968 images
63: 383.937805, 510.290070 avg, 0.000000 rate, 217.256994 seconds, 4032 images
64: 383.163666, 497.577423 avg, 0.000000 rate, 217.012377 seconds, 4096 images
65: 383.387421, 486.158417 avg, 0.000000 rate, 217.972890 seconds, 4160 images
66: 382.984528, 475.841034 avg, 0.000000 rate, 217.588748 seconds, 4224 images
67: 385.618958, 466.818817 avg, 0.000000 rate, 217.499060 seconds, 4288 images
68: 382.405975, 458.377533 avg, 0.000000 rate, 217.634510 seconds, 4352 images
69: 385.127777, 451.052551 avg, 0.000000 rate, 217.618965 seconds, 4416 images
70: 382.505920, 444.197876 avg, 0.000000 rate, 217.654535 seconds, 4480 images
71: 623.082397, 462.086334 avg, 0.000000 rate, 351.220555 seconds, 4544 images
72: 623.486328, 478.226318 avg, 0.000000 rate, 351.498025 seconds, 4608 images
73: 623.361328, 492.739807 avg, 0.000000 rate, 351.219987 seconds, 4672 images
74: 622.147156, 505.680542 avg, 0.000000 rate, 351.685479 seconds, 4736 images
75: 625.195312, 517.632019 avg, 0.000000 rate, 351.314560 seconds, 4800 images
76: 623.505066, 528.219299 avg, 0.000000 rate, 351.449000 seconds, 4864 images
77: 623.507935, 537.748169 avg, 0.000000 rate, 351.963596 seconds, 4928 images
78: 621.648926, 546.138245 avg, 0.000000 rate, 351.101064 seconds, 4992 images
79: 622.835510, 553.807983 avg, 0.000000 rate, 351.297160 seconds, 5056 images
80: 623.368774, 560.764038 avg, 0.000000 rate, 351.290002 seconds, 5120 images
81: 286.198059, 533.307434 avg, 0.000000 rate, 157.447369 seconds, 5184 images
82: 287.941803, 508.770874 avg, 0.000000 rate, 157.684069 seconds, 5248 images
83: 285.415344, 486.435333 avg, 0.000000 rate, 157.401089 seconds, 5312 images
84: 285.205841, 466.312378 avg, 0.000000 rate, 157.329998 seconds, 5376 images
85: 285.281372, 448.209290 avg, 0.000000 rate, 157.584071 seconds, 5440 images
86: 287.431305, 432.131500 avg, 0.000000 rate, 157.354369 seconds, 5504 images
87: 286.164795, 417.534821 avg, 0.000000 rate, 157.275441 seconds, 5568 images
88: 286.064484, 404.387787 avg, 0.000000 rate, 157.955694 seconds, 5632 images
89: 286.444519, 392.593445 avg, 0.000000 rate, 157.526958 seconds, 5696 images
90: 287.335541, 382.067657 avg, 0.000000 rate, 158.285737 seconds, 5760 images
91: 554.981995, 399.359100 avg, 0.000000 rate, 329.371322 seconds, 5824 images
92: 553.158691, 414.739075 avg, 0.000000 rate, 326.247533 seconds, 5888 images
93: 555.990479, 428.864227 avg, 0.000000 rate, 326.451206 seconds, 5952 images
94: 554.330505, 441.410858 avg, 0.000000 rate, 326.310421 seconds, 6016 images
95: 554.205627, 452.690338 avg, 0.000000 rate, 326.317394 seconds, 6080 images
96: 553.398254, 462.761139 avg, 0.000000 rate, 326.462671 seconds, 6144 images
97: 554.518982, 471.936920 avg, 0.000000 rate, 327.193108 seconds, 6208 images
98: 555.916748, 480.334900 avg, 0.000000 rate, 326.842712 seconds, 6272 images
99: 553.823120, 487.683716 avg, 0.000000 rate, 327.115745 seconds, 6336 images
100: 552.513184, 494.166656 avg, 0.000000 rate, 326.454763 seconds, 6400 images
101: 619.286072, 506.678589 avg, 0.000000 rate, 352.168429 seconds, 6464 images
102: 617.514282, 517.762146 avg, 0.000000 rate, 352.174336 seconds, 6528 images
103: 618.645813, 527.850525 avg, 0.000000 rate, 352.396979 seconds, 6592 images
104: 617.819763, 536.847473 avg, 0.000000 rate, 352.473959 seconds, 6656 images
105: 617.598755, 544.922607 avg, 0.000000 rate, 353.036217 seconds, 6720 images
106: 617.861511, 552.216492 avg, 0.000000 rate, 352.430663 seconds, 6784 images
107: 617.363586, 558.731201 avg, 0.000000 rate, 353.301873 seconds, 6848 images
108: 617.079346, 564.566040 avg, 0.000000 rate, 353.166611 seconds, 6912 images
109: 616.639038, 569.773315 avg, 0.000000 rate, 353.929210 seconds, 6976 images
110: 615.588623, 574.354858 avg, 0.000000 rate, 352.950311 seconds, 7040 images
111: 281.859894, 545.105347 avg, 0.000000 rate, 157.596891 seconds, 7104 images
112: 282.193329, 518.814148 avg, 0.000000 rate, 157.864882 seconds, 7168 images
113: 283.582581, 495.290985 avg, 0.000000 rate, 157.756852 seconds, 7232 images
114: 282.969513, 474.058838 avg, 0.000000 rate, 157.587287 seconds, 7296 images
115: 281.781921, 454.831146 avg, 0.000000 rate, 158.565058 seconds, 7360 images
116: 281.244690, 437.472504 avg, 0.000000 rate, 157.544969 seconds, 7424 images
117: 280.310699, 421.756317 avg, 0.000000 rate, 157.605689 seconds, 7488 images
118: 280.991211, 407.679810 avg, 0.000000 rate, 157.815611 seconds, 7552 images
119: 281.158966, 395.027710 avg, 0.000000 rate, 157.860355 seconds, 7616 images
120: 281.178833, 383.642822 avg, 0.000000 rate, 158.100813 seconds, 7680 images
121: 327.148590, 377.993408 avg, 0.000000 rate, 195.432400 seconds, 7744 images
122: 326.496704, 372.843750 avg, 0.000000 rate, 195.366968 seconds, 7808 images
123: 327.663025, 368.325684 avg, 0.000000 rate, 195.394300 seconds, 7872 images
124: 325.546722, 364.047791 avg, 0.000000 rate, 194.875608 seconds, 7936 images
125: 324.472107, 360.090210 avg, 0.000000 rate, 195.080409 seconds, 8000 images
126: 325.627075, 356.643890 avg, 0.000000 rate, 194.862185 seconds, 8064 images
127: 326.404358, 353.619934 avg, 0.000000 rate, 194.930244 seconds, 8128 images
128: 324.869019, 350.744843 avg, 0.000000 rate, 195.257795 seconds, 8192 images
129: 324.588776, 348.129242 avg, 0.000000 rate, 195.362603 seconds, 8256 images
130: 325.526917, 345.869019 avg, 0.000000 rate, 194.918326 seconds, 8320 images
131: 278.576416, 339.139771 avg, 0.000000 rate, 158.724961 seconds, 8384 images
132: 277.433044, 332.969086 avg, 0.000000 rate, 158.295029 seconds, 8448 images
133: 278.286682, 327.500854 avg, 0.000000 rate, 157.998602 seconds, 8512 images
134: 278.139282, 322.564697 avg, 0.000000 rate, 158.167720 seconds, 8576 images
135: 277.541443, 318.062378 avg, 0.000000 rate, 159.111993 seconds, 8640 images
136: 277.136139, 313.969757 avg, 0.000000 rate, 158.650136 seconds, 8704 images
137: 278.575226, 310.430298 avg, 0.000000 rate, 158.221151 seconds, 8768 images
138: 276.459656, 307.033234 avg, 0.000000 rate, 158.083424 seconds, 8832 images
139: 276.105682, 303.940491 avg, 0.000000 rate, 158.015226 seconds, 8896 images
140: 277.102905, 301.256744 avg, 0.000000 rate, 158.623654 seconds, 8960 images
141: 536.659912, 324.797058 avg, 0.000000 rate, 326.846880 seconds, 9024 images
142: 538.147034, 346.132050 avg, 0.000000 rate, 326.141047 seconds, 9088 images
143: 535.919373, 365.110779 avg, 0.000000 rate, 326.283251 seconds, 9152 images
144: 534.371216, 382.036835 avg, 0.000000 rate, 327.150031 seconds, 9216 images
145: 535.093018, 397.342468 avg, 0.000000 rate, 326.926293 seconds, 9280 images
146: 534.861145, 411.094330 avg, 0.000000 rate, 326.816545 seconds, 9344 images
147: 534.762146, 423.461121 avg, 0.000000 rate, 327.129895 seconds, 9408 images
148: 533.693176, 434.484314 avg, 0.000000 rate, 327.121451 seconds, 9472 images
149: 531.593079, 444.195190 avg, 0.000000 rate, 327.136581 seconds, 9536 images
150: 531.298828, 452.905548 avg, 0.000001 rate, 327.194996 seconds, 9600 images
151: 414.923370, 449.107330 avg, 0.000001 rate, 257.420264 seconds, 9664 images
152: 415.928467, 445.789429 avg, 0.000001 rate, 257.910323 seconds, 9728 images
153: 413.900208, 442.600494 avg, 0.000001 rate, 257.637355 seconds, 9792 images
154: 412.858124, 439.626251 avg, 0.000001 rate, 257.096158 seconds, 9856 images
155: 412.618256, 436.925446 avg, 0.000001 rate, 257.430933 seconds, 9920 images
156: 411.467041, 434.379608 avg, 0.000001 rate, 257.970096 seconds, 9984 images
157: 409.170044, 431.858643 avg, 0.000001 rate, 259.011452 seconds, 10048 images
158: 410.271393, 429.699921 avg, 0.000001 rate, 258.337243 seconds, 10112 images
159: 406.555450, 427.385468 avg, 0.000001 rate, 257.326154 seconds, 10176 images
160: 408.599060, 425.506836 avg, 0.000001 rate, 258.188722 seconds, 10240 images
161: 461.562439, 429.112396 avg, 0.000001 rate, 281.067575 seconds, 10304 images
162: 459.650452, 432.166199 avg, 0.000001 rate, 281.043317 seconds, 10368 images
163: 456.708221, 434.620392 avg, 0.000001 rate, 281.593234 seconds, 10432 images
164: 454.844574, 436.642822 avg, 0.000001 rate, 281.670021 seconds, 10496 images
165: 454.015350, 438.380066 avg, 0.000001 rate, 287.939994 seconds, 10560 images
166: 453.958282, 439.937897 avg, 0.000001 rate, 289.653265 seconds, 10624 images
167: 452.909180, 441.235016 avg, 0.000001 rate, 295.676957 seconds, 10688 images
168: 450.827209, 442.194244 avg, 0.000001 rate, 286.917430 seconds, 10752 images
169: 449.469177, 442.921753 avg, 0.000001 rate, 290.520989 seconds, 10816 images
170: 447.693481, 443.398926 avg, 0.000001 rate, 286.509354 seconds, 10880 images

cv2.error: OpenCV(4.2.0) ../modules/dnn/src/darknet/darknet_io.cpp:519: error: (-215:Assertion failed) separator_index < line.size() in function 'ReadDarknetFromCfgStream'

I'm getting same error while performing prediction on trained YOLOv3 Tiny. I have checked number of anchor are 6 only.

Could someone help me on this please .

I have trained model in my virtual machine Ubuntu 20.04

yolov3 tiny.config
#yoloblock1
[yolo]
mask = 3,4,5
anchors = 10,14, 23,27, 37,58, 81,82, 135,169, 344,319
classes=1
num=6
jitter=.3
ignore_thresh = .7
truth_thresh = 1
random=1

#yoloblock2
[yolo]
mask = 0,1,2
anchors = 10,14, 23,27, 37,58, 81,82, 135,169, 344,319
classes=1
num=6
jitter=.3
ignore_thresh = .7
truth_thresh = 1
random=1

Command:
darknet$ python3 yolo_opencv.py -c /home/ai/darknet/training/yolov3-tiny.cfg -w /home/ai/darknet/backup/yolov3-tiny_60000.weights -cl /home/ai/darknet/training/objects.names

Error

Traceback (most recent call last):
File "yolo_opencv.py", line 47, in
net = cv2.dnn.readNet(args.weights,args.config)
cv2.error: OpenCV(4.2.0) ../modules/dnn/src/darknet/darknet_io.cpp:519: error: (-215:Assertion failed) separator_index < line.size() in function 'ReadDarknetFromCfgStream'

What is the object id of Rubik's cube

I have a query about the custom/train.txt file. this file should contain the bounding box information.

size of image files annotataed

is it necessary that all images are of same size or different image size is okay?

When training on Yolov3 usage of CPU is high and GPU is low

Hello!
I have some trouble figuring out if I am doing my training on custom images correctly, because when I'm running the training this is what i see:

So my feeling is that training should be using much more of GPU RAM and not as much of CPU power. Also my nvidia-smi processes bar does not show any proceesses (but that might be another problem)

My question would be: Is this normal behaviour and if not, what should I do to make Yolo training make use of GPU more?

Im running Yolov3 training from inside docker, which I enter with:
docker run --runtime=nvidia -it --rm --entrypoint "/bin/bash" --env DISPLAY=$DISPLAY -v /home/spacerobo/Documents/Git/yolo-custom-object-detector:/home yolo3-opencv-cuda

and from /home/python/ I start training with this command:
/darknet/./darknet detector train DatasetStones/trainer.data DatasetStones/yolov3-tiny.cfg darknet53.conv.74

The darknet Makefile first couple of lines look like this:

GPU=1
CUDNN=1
OPENCV=1
OPENMP=0
DEBUG=0
CUDA=1

And the Yolo config file first couple of relevant lines:

[net]
# Testing
batch=1
subdivisions=1
# Training
batch=64
subdivisions=16
width=416
height=416
channels=3
momentum=0.9
decay=0.0005
angle=0
saturation = 1.5
exposure = 1.5
hue=.1

learning_rate=0.001
burn_in=1000
max_batches = 500200
policy=steps
steps=400000,45000032
scales=.1,.1

My setup:
Nvidia GeForce GT 740, 2GB
Intel i5-4460 CPU
8GB RAM
Manjaro OS

Cannot add vertex: a vertex with name "outputs" already exists

Exception in thread "main" java.lang.IllegalStateException: Cannot add vertex: a vertex with name "outputs" already exists
at org.nd4j.base.Preconditions.throwStateEx(Preconditions.java:641)
at org.nd4j.base.Preconditions.checkState(Preconditions.java:304)
at org.deeplearning4j.nn.conf.ComputationGraphConfiguration$GraphBuilder.addVertex(ComputationGraphConfiguration.java:924)
at org.deeplearning4j.nn.conf.ComputationGraphConfiguration$GraphBuilder.addLayer(ComputationGraphConfiguration.java:788)
at org.deeplearning4j.nn.transferlearning.TransferLearning$GraphBuilder.addLayer(TransferLearning.java:817)
at com.sax2.muaaz.joosuf.yolo_neural_network.YoloAlgorithm.main(YoloAlgorithm.java:152)

any ideas cause im completely stuck now

i have changed my object names but all it detect is "person"

i only have one class and it is called "Corrosion"

The readme video was taken off of youtube