Yolov3: shape

[MichaelCong]

RuntimeError: shape '[128, 64, 3, 3]' is invalid for input of size 44878

[MichaelCong]

python train.py
Namespace(accumulate=8, backend='nccl', batch_size=8, cfg='cfg/yolov3-spp.cfg', data_cfg='data/coco_64img.data', dist_url='tcp://127.0.0.1:9999', epochs=100, evolve=False, giou=False, img_size=416, nosave=False, notest=False, num_workers=4, rank=0, resume=False, single_scale=False, transfer=False, var=0, world_size=1)
Using CUDA device0 _CudaDeviceProperties(name='GeForce GTX 1080 Ti', total_memory=11175MB)
device1 _CudaDeviceProperties(name='GeForce GTX 1080 Ti', total_memory=11178MB)
device2 _CudaDeviceProperties(name='GeForce GTX 1080 Ti', total_memory=11178MB)
device3 _CudaDeviceProperties(name='GeForce GTX 1080 Ti', total_memory=11178MB)

Traceback (most recent call last):
File "train.py", line 330, in
accumulate=opt.accumulate,
File "train.py", line 103, in train
cutoff = load_darknet_weights(model, weights + 'darknet53.conv.74')
File "/home/rencong/yolov3/models.py", line 316, in load_darknet_weights
conv_w = torch.from_numpy(weights[ptr:ptr + num_w]).view_as(conv_layer.weight)
RuntimeError: shape '[128, 64, 3, 3]' is invalid for input of size 44878

[glenn-jocher]

Hello, thank you for your interest in our work! This is an automated response. Please note that most technical problems are due to:

• Your changes to the default repository. If your issue is not reproducible in a fresh git clone version of this repository we can not debug it. Before going further run this code and ensure your issue persists:

sudo rm -rf yolov3  # remove exising repo
git clone https://github.com/ultralytics/yolov3 && cd yolov3 # git clone latest
python3 detect.py  # verify detection
python3 train.py  # verify training (a few batches only)
# CODE TO REPRODUCE YOUR ISSUE HERE

• Your custom data. If your issue is not reproducible with COCO data we can not debug it. Visit our Custom Training Tutorial for exact details on how to format your custom data. Examine train_batch0.jpg and test_batch0.jpg for a sanity check of training and testing data.
• Your environment. If your issue is not reproducible in a GCP Quickstart Guide VM we can not debug it. Ensure you meet the requirements specified in the README: Unix, MacOS, or Windows with Python >= 3.7, Pytorch >= 1.1, etc. You can also use our Google Colab Notebook to test your code in working environment.

If none of these apply to you, we suggest you close this issue and raise a new one using the Bug Report template, providing screenshots and minimum viable code to reproduce your issue. Thank you!

[sanazss]

Hi,

Thanks for your help so far. My images are square, do you think I need to tweak any part of the code to suit this? when I implement the letterbox function on my images. the img doesn't get the 128values and the last line of the function doesn't work on my images. do you have any suggestions. I get poor results for training and testing.
Also, I got "Nan" as loss when I printed my losses. Any hint on this?

[glenn-jocher]

@sanazss I see you are posting many issues and trying many things, but I believe you are misdirecting your efforts. My suggestions are very simple:

1. Modify your training data with the correct format.
2. git pull the latest repo. Do not modify anything.
3. Train to 100 epochs.
4. Plot your training results using from utils import utils; utils.plot_results(). Upload your train_batch0.jpg, test_batch0.jpg, and results.png image here.

Without these 3 images I cant provide you any suggestions.