mean_mAP issue

[mama110]

I'm training VOC2007, afer 1 epoch, an error shows up:

F:\pytorch-yolov3-master-ul\test.py:122: RuntimeWarning: invalid value encountered in double_scalars
print('%15s: %-.4f' % (c, AP_accum[i] / AP_accum_count[i]))
aeroplane: nan
bicycle: nan
bird: nan
boat: nan
bottle: nan
bus: nan
car: nan
cat: nan
chair: nan
cow: nan
diningtable: nan
dog: nan
horse: nan
motorbike: nan
person: nan
pottedplant: nan
sheep: nan
sofa: nan
train: nan
tvmonitor: nan
Traceback (most recent call last):
File "train.py", line 268, in
var=opt.var,
File "train.py", line 224, in train
img_size=img_size,
File "F:\pytorch-yolov3-master-ul\test.py", line 125, in test
return mean_mAP, mean_R, mean_P
UnboundLocalError: local variable 'mean_mAP' referenced before assignment

What's the problem?

[nirbenz]

This is because mean_mAP isn't initialized and if a snapshot failed to detect any objects (quite normal for the first few evaluation rounds) it will fail.
Add this before the train loop.

mean_mAP, mean_R, mean_P = 0.0, 0.0, 0.0

@glenn-jocher I'd PR this but it's a single line :)

[mama110]

@nirbenz Thanks.

[glenn-jocher]

@nirbenz thanks, I added this line to test.py in the latest commit!

[pprp]

After running python train.py I got the error below:

WARNING Using GPU0 Only. Multi-GPU issue: https://github.com/ultralytics/yolov3/issues/21
Using cuda:0

      Image      Total          P          R        mAP
      Image      Total          P          R        mAP

mAP Per Class:
/media/cieserver/209b4bc6-b4ad-4a37-b5ee-cde3c165103e/dpj/yolov3/test.py:114: RuntimeWarning: invalid value encountered in double_scalars
  print('%15s: %-.4f' % (c, AP_accum[i] / AP_accum_count[i]))
           head: nan
            eye: nan
           nose: nan

I am using the latest commits and I get the error same as @mama110

mean_mAP, mean_R, mean_P = 0.0, 0.0, 0.0

this solution don't work.

[glenn-jocher]

Do you have more than one GPU?

On Sun, 17 Feb 2019 at 03:09, dpj ***@***.***> wrote: After running python train.py I got the error below: WARNING Using GPU0 Only. Multi-GPU issue: #21 Using cuda:0 Image Total P R mAP Image Total P R mAP mAP Per Class: /media/cieserver/209b4bc6-b4ad-4a37-b5ee-cde3c165103e/dpj/yolov3/test.py:114: RuntimeWarning: invalid value encountered in double_scalars print('%15s: %-.4f' % (c, AP_accum[i] / AP_accum_count[i])) head: nan eye: nan nose: nan I am using the latest version and I get the error same as @mama110 <https://github.com/mama110> mean_mAP, mean_R, mean_P = 0.0, 0.0, 0.0 this solution don't work. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#50 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AZlyGdvXXoxCJ9xVpx7Ih7YHb0tSazECks5vOLndgaJpZM4ZYJM_> .

-- Glenn Jocher Founder and CEO, Ultralytics LLC +1 301-237-6695

[pprp]

Yes , I have two GPU but I used CUDA_VISIBLE_DEVICES=0

[pprp]

After the training, the weight file can detect the target, so it is very likely that the test part calculation is a problem. In addition, I have a question, how to draw pictures below?

[pprp]

I found a problem with my valid file. The path is wrong and I solved it. Thank you very much for your quick response.

[glenn-jocher]

@pprp you can make the plots below using plot_results() in utils.py. You can see more examples here:
https://github.com/ultralytics/yolov3/wiki/Single-Image-Training-Example

[pprp]

Thank you for your kind response! 😄

[kenrubiooo]

Hi @mama110, @glenn-jocher,

I also encountered RuntimeWarning: invalid value encountered in double_scalars. I got the APs but the Targets, P, R, mAP, and F1 are 0 and nans. What could be the problem here? Thank you!

[glenn-jocher]

@kenrubiooo 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 >= 1.0, etc.

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!

[kenrubiooo]

Hi @glenn-jocher

To clarify, I am just experiencing an error in the test.py when trying to determine the mAP.

[glenn-jocher]

@kenrubiooo run this code exactly and paste your terminal output:

sudo rm -rf yolov3  # remove exising repo
git clone https://github.com/ultralytics/yolov3 && cd yolov3 # git clone latest
python3 test.py  # verify test