YoloV9: Exception in thread Thread-14 (plot_images):

Search before asking

• I have searched the Roboflow Notebooks issues and found no similar bug report.

Notebook name

YOLOv9 Object Detection
(https://github.com/roboflow-ai/notebooks/blob/main/notebooks/train-yolov9-object-detection-on-custom-dataset.ipynb)

Bug

Hi everybody. I am training a custom dataset with the yoloV9 model and the following error message appears. Does someone have a similar problem or know how to fix this exception ? When I start the train.py the following exception (Exception in thread Thread-14 (plot_images)) is raising up, but it does not interrupt the program. Here is the full code.

0/299      10.8G      1.432      5.195      1.115         74        640:  Exception in thread Thread-14 (plot_images):
Traceback (most recent call last):
  File "/opt/conda/lib/python3.10/threading.py", line 1016, in _bootstrap_inner
    self.run()
  File "/opt/conda/lib/python3.10/threading.py", line 953, in run
    self._target(*self._args, **self._kwargs)
  File "/kaggle/working/yolov9/utils/plots.py", line 300, in plot_images
    annotator.box_label(box, label, color=color)
  File "/kaggle/working/yolov9/utils/plots.py", line 86, in box_label
    w, h = self.font.getsize(label)  # text width, height
AttributeError: 'FreeTypeFont' object has no attribute 'getsize'
      0/299      10.8G      1.432      5.195      1.115         74        640:  /kaggle/working/yolov9/train.py:302: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with torch.cuda.amp.autocast(amp):
/opt/conda/lib/python3.10/site-packages/torch/nn/parallel/parallel_apply.py:79: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with torch.cuda.device(device), torch.cuda.stream(stream), autocast(enabled=autocast_enabled):
      0/299      10.9G      1.364      4.798      1.082         80        640:  Exception in thread Thread-17 (plot_images):
Traceback (most recent call last):
  File "/opt/conda/lib/python3.10/threading.py", line 1016, in _bootstrap_inner
    self.run()
  File "/opt/conda/lib/python3.10/threading.py", line 953, in run
    self._target(*self._args, **self._kwargs)
  File "/kaggle/working/yolov9/utils/plots.py", line 300, in plot_images
    annotator.box_label(box, label, color=color)
  File "/kaggle/working/yolov9/utils/plots.py", line 86, in box_label
    w, h = self.font.getsize(label)  # text width, height
AttributeError: 'FreeTypeFont' object has no attribute 'getsize'
      0/299      10.9G      1.407      4.783      1.132         62        640:  Exception in thread Thread-20 (plot_images):
Traceback (most recent call last):
  File "/opt/conda/lib/python3.10/threading.py", line 1016, in _bootstrap_inner
    self.run()
  File "/opt/conda/lib/python3.10/threading.py", line 953, in run
    self._target(*self._args, **self._kwargs)
  File "/kaggle/working/yolov9/utils/plots.py", line 300, in plot_images
    annotator.box_label(box, label, color=color)
  File "/kaggle/working/yolov9/utils/plots.py", line 86, in box_label
    w, h = self.font.getsize(label)  # text width, height
AttributeError: 'FreeTypeFont' object has no attribute 'getsize'_**

Environment

Ultralytics 8.3.18 🚀 Python-3.10.14 torch-2.4.0 CUDA:0 (Tesla T4, 15095MiB)
Setup complete ✅ (4 CPUs, 31.4 GB RAM, 5934.0/8062.4 GB disk)

Minimal Reproducible Example

Starting training for 300 epochs...
/opt/conda/lib/python3.10/multiprocessing/popen_fork.py:66: RuntimeWarning: os.fork() was called. os.fork() is incompatible with multithreaded code, and JAX is multithreaded, so this will likely lead to a deadlock.
  self.pid = os.fork()

      Epoch    GPU_mem   box_loss   cls_loss   dfl_loss  Instances       Size
  0%|          | 0/183 00:00/kaggle/working/yolov9/train.py:302: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with torch.cuda.amp.autocast(amp):
/opt/conda/lib/python3.10/site-packages/torch/nn/parallel/parallel_apply.py:79: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with torch.cuda.device(device), torch.cuda.stream(stream), autocast(enabled=autocast_enabled):
      0/299      10.8G      1.432      5.195      1.115         74        640:  Exception in thread Thread-14 (plot_images):
Traceback (most recent call last):
  File "/opt/conda/lib/python3.10/threading.py", line 1016, in _bootstrap_inner
    self.run()
  File "/opt/conda/lib/python3.10/threading.py", line 953, in run
    self._target(*self._args, **self._kwargs)
  File "/kaggle/working/yolov9/utils/plots.py", line 300, in plot_images
    annotator.box_label(box, label, color=color)
  File "/kaggle/working/yolov9/utils/plots.py", line 86, in box_label
    w, h = self.font.getsize(label)  # text width, height
AttributeError: 'FreeTypeFont' object has no attribute 'getsize'
      0/299      10.8G      1.432      5.195      1.115         74        640:  /kaggle/working/yolov9/train.py:302: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with torch.cuda.amp.autocast(amp):
/opt/conda/lib/python3.10/site-packages/torch/nn/parallel/parallel_apply.py:79: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with torch.cuda.device(device), torch.cuda.stream(stream), autocast(enabled=autocast_enabled):
      0/299      10.9G      1.364      4.798      1.082         80        640:  Exception in thread Thread-17 (plot_images):
Traceback (most recent call last):
  File "/opt/conda/lib/python3.10/threading.py", line 1016, in _bootstrap_inner
    self.run()
  File "/opt/conda/lib/python3.10/threading.py", line 953, in run
    self._target(*self._args, **self._kwargs)
  File "/kaggle/working/yolov9/utils/plots.py", line 300, in plot_images
    annotator.box_label(box, label, color=color)
  File "/kaggle/working/yolov9/utils/plots.py", line 86, in box_label
    w, h = self.font.getsize(label)  # text width, height
AttributeError: 'FreeTypeFont' object has no attribute 'getsize'
      0/299      10.9G      1.407      4.783      1.132         62        640:  Exception in thread Thread-20 (plot_images):
Traceback (most recent call last):
  File "/opt/conda/lib/python3.10/threading.py", line 1016, in _bootstrap_inner
    self.run()
  File "/opt/conda/lib/python3.10/threading.py", line 953, in run
    self._target(*self._args, **self._kwargs)
  File "/kaggle/working/yolov9/utils/plots.py", line 300, in plot_images
    annotator.box_label(box, label, color=color)
  File "/kaggle/working/yolov9/utils/plots.py", line 86, in box_label
    w, h = self.font.getsize(label)  # text width, height
AttributeError: 'FreeTypeFont' object has no attribute 'getsize'
      0/299      10.9G      1.037      1.106     0.9808         32        640: 1
                 Class     Images  Instances          P          R      mAP50   
                   all        812        838      0.964      0.943      0.987      0.667
/opt/conda/lib/python3.10/multiprocessing/popen_fork.py:66: RuntimeWarning: os.fork() was called. os.fork() is incompatible with multithreaded code, and JAX is multithreaded, so this will likely lead to a deadlock.
  self.pid = os.fork()

Additional

No response

Are you willing to submit a PR?

• Yes I'd like to help by submitting a PR!

Start with the YOLOv9 notebook and inspect utils/plots.py, especially box_label at line 86 and plot_images at line 300. Reproduce the training run in the reported Python and PyTorch environment, then verify that image plotting completes without the FreeTypeFont exception.