keypoint_rcnn_training_pytorch
keypoint_rcnn_training_pytorch copied to clipboard
How do I get `evaluate()` to work?
First off, thanks for the tutorial! You're a real lifesaver. I'm trying to get evaluate
to work at the moment. I have num_keypoints=4
. The model trains smoothly when I just comment out evaluate
during training, but I need to evaluate the performance now so I'm trying to get that to work. I did what you instructed in your tutorial:
Update. It’s possible not to edit
pycocotools/cocoeval.py
file in pycocotools library to changekpt_oks_sigmas
, but to editcoco_eval.py
file, as Diogo Santiago suggested:
# self.coco_eval[iou_type] = COCOeval(coco_gt, iouType=iou_type)
coco_eval = COCOeval(coco_gt, iouType=iou_type)
coco_eval.params.kpt_oks_sigmas = np.array([.5, .5]) / 10.0
self.coco_eval[iou_type] = coco_eval
Since I have num_keypoints=4
, I instead wrote np.array([.5, .5, .5, .5]) / 10.0
for kpt_oks_sigmas
.
But I'm still getting the same error: ValueError: operands could not be broadcast together with shapes (4,) (17,)
Hoping you can help me with this! 🤞
😃 Fixed!
Apparently, the solution on the article was right. But for some reason, when I edit on Colab and save, it doesn't actually get saved. So I had to already have changed coco_eval.py
before I even cloned it on my Colab runtime.
I basically did the same thing, but this time I downloaded the training scripts from PyTorch and created my own repository to clone:
- Download the scripts from pytorch/vision/references/detection and created my own repo storing it
- Open
coco_eval.py
- Edit the lines somewhere under
__init__
for iou_type in iou_types:
- self.coco_eval[iou_type] = COCOeval(coco_gt, iouType=iou_type) + coco_eval = COCOeval(coco_gt, iouType=iou_type) + coco_eval.params.kpt_oks_sigmas = np.array([.5, .5, .5, .5]) / 10.0 # 4 keypoints + self.coco_eval[iou_type] = coco_eval
- At the start of the Training Notebook:
!git clone https://github.com/<username>/<repo-name>.git %cd <repo-name>
Thanks, man. I've been struggling with this error. Apparently, making direct changes to the file in Colab doesn't update it. weird.