yolov7 icon indicating copy to clipboard operation
yolov7 copied to clipboard

YOLOV7-pose algorithm.

Open aoqiangma opened this issue 2 years ago • 8 comments

This error occurred when I ran the YOLOV7-pose algorithm.

train: Scanning 'F:\code\yolov7-pose\point\train2017' images and labels... 18 found, 0 missing, 0 empty, 18 corrupted: 100%|██████████| 18/18 [00:00<00:00, 1199.57it/s] train: New cache created: F:\code\yolov7-pose\point\train2017.cache Traceback (most recent call last): File "F:/code/yolov7-pose/train.py", line 562, in train(hyp, opt, device, tb_writer) File "F:/code/yolov7-pose/train.py", line 204, in train image_weights=opt.image_weights, quad=opt.quad, prefix=colorstr('train: '), kpt_label=kpt_label) File "F:\code\yolov7-pose\utils\datasets.py", line 74, in create_dataloader kpt_label=kpt_label) File "F:\code\yolov7-pose\utils\datasets.py", line 414, in init labels, shapes, self.segments = zip(*cache.values()) ValueError: not enough values to unpack (expected 3, got 0) train: WARNING: Ignoring corrupted image and/or label F:\code\yolov7-pose\point\images\train2017\000000000036.jpg: cannot reshape array of size 55 into shape (2)

Friends, I'm stuck in this problem. Can someone help me?

aoqiangma avatar Nov 16 '22 01:11 aoqiangma

This error occurred when I ran the YOLOV7-pose algorithm.

train: Scanning 'F:\code\yolov7-pose\point\train2017' images and labels... 18 found, 0 missing, 0 empty, 18 corrupted: 100%|██████████| 18/18 [00:00<00:00, 1199.57it/s] train: New cache created: F:\code\yolov7-pose\point\train2017.cache Traceback (most recent call last): File "F:/code/yolov7-pose/train.py", line 562, in train(hyp, opt, device, tb_writer) File "F:/code/yolov7-pose/train.py", line 204, in train image_weights=opt.image_weights, quad=opt.quad, prefix=colorstr('train: '), kpt_label=kpt_label) File "F:\code\yolov7-pose\utils\datasets.py", line 74, in create_dataloader kpt_label=kpt_label) File "F:\code\yolov7-pose\utils\datasets.py", line 414, in init labels, shapes, self.segments = zip(*cache.values()) ValueError: not enough values to unpack (expected 3, got 0) train: WARNING: Ignoring corrupted image and/or label F:\code\yolov7-pose\point\images\train2017\000000000036.jpg: cannot reshape array of size 55 into shape (2)

Friends, I'm stuck in this problem. Can someone help me?

I am also troubled by this problem. Please let me know if there is a solution

jq9898 avatar Nov 17 '22 03:11 jq9898

Ok, but I'm also a beginner. If any of us solve this problem, please tell each other the solution in the mailbox.

| | @.*** | | @.*** |

---- Replied Message ---- | From | @.> | | Date | 11/17/2022 11:03 | | To | @.> | | Cc | @.@.> | | Subject | Re: [WongKinYiu/yolov7] YOLOV7-pose algorithm. (Issue #1110) |

This error occurred when I ran the YOLOV7-pose algorithm.

train: Scanning 'F:\code\yolov7-pose\point\train2017' images and labels... 18 found, 0 missing, 0 empty, 18 corrupted: 100%|██████████| 18/18 [00:00<00:00, 1199.57it/s] train: New cache created: F:\code\yolov7-pose\point\train2017.cache Traceback (most recent call last): File "F:/code/yolov7-pose/train.py", line 562, in train(hyp, opt, device, tb_writer) File "F:/code/yolov7-pose/train.py", line 204, in train image_weights=opt.image_weights, quad=opt.quad, prefix=colorstr('train: '), kpt_label=kpt_label) File "F:\code\yolov7-pose\utils\datasets.py", line 74, in create_dataloader kpt_label=kpt_label) File "F:\code\yolov7-pose\utils\datasets.py", line 414, in init labels, shapes, self.segments = zip(*cache.values()) ValueError: not enough values to unpack (expected 3, got 0) train: WARNING: Ignoring corrupted image and/or label F:\code\yolov7-pose\point\images\train2017\000000000036.jpg: cannot reshape array of size 55 into shape (2)

Friends, I'm stuck in this problem. Can someone help me?

I am also troubled by this problem. Please let me know if there is a solution

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

aoqiangma avatar Nov 17 '22 03:11 aoqiangma

Insert the --kpt-label argument and it will solve the issue

Liagos avatar Dec 18 '22 13:12 Liagos

how can i train on my custom pose classes?

akashAD98 avatar Dec 23 '22 05:12 akashAD98

first delete train.cache and then run python train.py --kpt-label, it works well

Franksongsong avatar May 29 '23 07:05 Franksongsong

yes i delete train.cache and then run python train.py --kpt-label, it is not work how to solve

schemaphicGopa avatar Oct 31 '23 07:10 schemaphicGopa

i use --kpt-label but not solve problem

schemaphicGopa avatar Oct 31 '23 08:10 schemaphicGopa

I suspect the issue might be related to the annotation file.

mohammedAljadd avatar Dec 14 '23 08:12 mohammedAljadd